Session 4

National Institute of Technology, Rourkela 769008, Orissa.
B. Chitti Babu*, K. Aroul*, C. Poongothai** * Department of Electrical Engineering, National Institute of Technology Rourkela ** Department of Electrical Engineering, Indian Institute of Technology Madras Abstract- The superior properties of silicon carbide (SiC) power electronic devices compared with silicon (Si) are expected to have a significant impact on next-generation power components. The system-level benefits of using SiC devices include a large reduction in the size, weight, and cost of the power conditioning and/or thermal systems. However, the expected performance characteristics of the various semiconductor devices and the impact that these devices could have in applications are not well understood. In this paper, the progress made in Sic semiconductor technology research and developments are addressed. Attempts are made to highlight the current and future issues involved for the development of Sic technology for future power semiconductor devices in the high temperature and high power applications. 1. INTRODUCTION Since more than one decade, the research and development of SiC power devices have been done at various institutions and companies world wide with many promising results .The driving force for the use of SiC in power electronics is the potential benefit to realize low loss and very fast unipolar diodes and switches with blocking voltages from 600 V up to 4000 V [1]. Compared to the today predominantly used bipolar silicon structures, unipolar devices feature the a linear increase in voltage drop with the forward current and a positive temperature coefficient of the resulting differential on resistance which enable easy paralleling. In addition, the absence of a minority carrier flow in the main operation region enables a switching performance known only for low voltage (VBR<100V) systems today. The reason for the ultra low loss performance of high voltage unipolar devices is the wide bandgap of SiC which leads to a 243
SiC Power Semiconductor Devices- An impact on Next Generation Power Components
10 times higher critical electrical field (compared to Si) and therefore to 10 times "thinner" devices with a corresponding low on-resistance [2]. As mentioned before, SiC power devices, with their close to ideal characteristics assures great performance improvements. Some of the advantages compared with Si based power devices are as follows [3]. 1) SiC devices are thinner and they have lower on resistances. At low breakdown voltages (~50V), SiC unipolar device on-resistances are around 100 times less; and at higher breakdown voltages (~5000), up to 300 times less. With lower Ron, SiC power devices have lower conduction losses and therefore higher overall efficiency. 2) SiC-based power devices have higher breakdown voltages because of their higher electric breakdown field. 3) SiC has a higher thermal conductivity and thus a lower junction-to-case thermal resistance, Rth-jc and thus device temperature increase is slower. 4) SiC can operate at high temperatures. SiC device operation at up to 600C is mentioned in literature [1]. Si devices, on the other hand, can operate at a maximum junction temperature of only 150C. 5) SiC is extremely radiation hard; i.e., radiation does not degrade the electronic properties of SiC. 6) Forward and reverse characteristics of SiC power devices vary only slightly with temperature and time; 7) SiC-based devices have excellent reverse recovery characteristics. With less reverse recovery current, the switching loss and EMI are reduced and there is no need for snubbers. Regarding commercial products, fast SiC Schottky diodes in the voltage range between 300 V and 600 V were introduced on the market 2001 by Infineon for the first time [4]. Because of the
Proceedings of the 1st APE-2009 on Jan. 24-25, 2009
virtually zero reverse charge storage of Scotty diodes these diodes are the nearly ideal choice for large growing market of boost converter topologies in high end power supplies for efficiency improvement. In addition to that, these diodes find use in solar inverters, because they have lower switching losses than the Si PIN freewheeling diodes used in that applications now. The higher costs of the new technology are well compensated by the reduced size of passive components [5]. In this paper, the progress made in Sic semiconductor technology research and developments are addressed. Attempts are made to highlight the current and future issues involved for the development of Sic technology for future power semiconductor devices in the high temperature and high power applications. 2. SiC POWER SEMICONDUCTOR DEVICES As stated above, Silicon have played a one man army and work horse of power electronics community since 1954 for high-voltage high power applications. The situation has changed due to the significant achievements in bulk material growth and in device processing technology. Thus, many high-voltage experimental SiC based two-terminal rectifiers and three-terminal switches have been already demonstrated. 4H-Sic unipolar devices are expected to replace Si bipolar rectifiers in the 6006500 V range and power switches higher than 1.2kV in near the figure [6]. A. SiC Schottky diode a) Structure
The structure of a SiC Schottky power diode and its equivalent circuit are shown in Fig. 1. VFB is the voltage drop across the Schottky barrier; RD is the resistive voltage drop across the lightly doped drift region; RS and RC are the resistance of substrate and contact, respectively. Fig.2 shows the basic V-I characteristic of a first generation Schottky diode as available at the market beginning from 2001 and the following points were observed. b) Static and Dynamic Characteristics It is important to emphasize the main differences compared to the behavior of classical bipolar silicon devices. Firstly, we have to deal with a considerable increase of the forward drop with temperature. On the one hand this feature stabilizes the current sharing in the case of paralleling, however, the static losses increase dramatically. The most severe drawback of this behavior is the limited surge current ability of such a structure. In case of a forced high current through the device, even for a very short time period, the danger of thermal destruction is high and thus, in a lot of the applications today, devices with a higher current handling capability are selected only for managing surge currents during turn on or drop out events. Since higher current capabilities means larger active areas, the cost pressure is very high. It is important to emphasize that today devices area in SiC chips is the most severe cost driver. The second feature which is different compared to standard devices is the reverse current behavior with temperature and in the case on over-voltage events. Since the reverse behavior is dominated by a combination of thermo-ionic emission and tunneling currents [6], the reverse current increases and the virtual blocking voltage decreases with temperature. Avalanche is difficult to achieve because the leakage current over the Schottky barrier will already at electric fields well below the critical field reach current densities which are critical. Thus, over-voltages are a dangerous stress for the devices and should be avoided. This above critical discussion shows, that a simple plug and play cannot be the right strategy if a circuit designer plans to use SiC Schottky diodes. A comprehensive consideration of all features is important for a successful system design. Nevertheless, the most
Fig. 1. Basic structure and equivalent circuit of Schottky power diodes.
244
Fig. 2. Forward characteristic of SiC Schottky power diode and Si Power Diode.
Fig. 3 Turn-off loss of SiC Schottky power diode and Si Power Diode
critical drawbacks mentioned above are addressed by the new, second generation of Schottky barrier diodes which was introduced recently. The new device represents a mixture between a pure Schottky barrier device and a pn structure (a so called merged pin-Schottky diode[7].Regarding the reverse recovery characteristics of SiC Schottky diode, the reverse recovery current does not change with forward current, theoretically, Schottky diodes do not display reverse recovery phenomenon[8].
As a summary, the static characteristics of both of the tested diodes are similar; however, the dynamic characteristics are much different. Si pn diode has high peak reverse recovery currents that result in high diode switching losses and extra IGBT losses since the reverse current has to go through a main switch. Consequently, it is expected for the Si IGBT SiC Schottky diode inverter to perform better than the similar allSi inverter [9].
Table I: Physical Properties of Various Semiconductors for Power Devices
Material Si Ga As 3C-SiC 6H-SiC 4H-SiC GaN GaP C
Eg (eV) 1.12 1.4 2.3 2.9 3.2 3.39 2.26 5.6
(cm /V.s) 1450 8500 1000 415 950 1000 250 2200
(cm /V.s) 450 400 45 90 115 350 150 1800
Vsat (cm/s) 107 2 x10 2.5 x107 2 x107 2 x107 2 x107 3 x107
7
Ec (V/cm) 3 x105 4 x10

5
(W/cm.K) 1.3 0.54 5 5 5 1.3 1.1 20
r
11.7 12.9 9.6 9.7 10 8.9 11.1 5.7
2 x106 2.5 x106 3 x106 5x106 107 5.6x105
245
B. SiC MOSFET IGBTs have been predominantly used for power conversion at 600-V and 1200-V ratings. Previously, Si MOSFETs were handicapped in those applications by their high on-resistance (RDSON). At high breakdown voltages, RDSON increases approximately with the square of the drain-source breakdown voltage VDSMAX. The RDSON of a MOSFET consists of the sum of the channel resistance, the inherent JFET resistance and the drift resistance as shown in fig. 1. The drift resistance (RDRIFT) is the dominant portion of the overall resistance, where d equals drift-layer thickness, q equals electron charge, n equals channel mobility and ND equals doping factor [11].
temperatures greater than 200C, but the engineering sample is limited in temperature to 150C by its TO-247 plastic package. Compared with a Si IGBT, a SiC MOSFET has a substantial advantage in conduction losses, particularly at lower power outputs. By virtue of its univocal nature, it has no tail currents at turn-off, thereby leading to greatly reduced turn-off losses. Table 2 shows the switching loss difference when compared with a standard off-the-shelf 1200-V IGBT [12]. The switching losses of a SiC MOSFET are less than half those of a Si IGBT (1.14 mJ versus 2.6 mJ, respectively). Combining this switching-loss reduction with the lower overall conduction losses, it is clear that the SiC switch is a much more efficient device for high-power-conversion systems.
Table 2. Switching Loss Comparison
Fig.4 Cross-section of DMOSFET power transistor shows its resistive components
The new generation of SiC MOSFETs cuts driftlayer thickness by nearly a factor of 10 while simultaneously enabling the doping factor to increase by the same order of magnitude. The overall effect results in a reduction of the drift resistance to 1/100th of its Si MOSFET equivalent. The improved SiC MOSFET discussed here is an engineering sample of a 1200-V, 20-A device with a 100-mV RDSON at a 15-V gate-source voltage. Besides its inherent reduction in on-resistance, SiC also offers a substantially reduced on-resistance variation over its operating temperature. From 25C to 150C, SiC variations are in the range of 20%, compared with 200% to 300% for Si. The SiC MOSFET die is capable of operation at junction 246
C. UNIPLOAR SiC JFETs SWITCHES For the blocking voltage range from l.2kV to I.5kV, the Si MOSFET is not a realistic option and the Si IGBT shows high dynamic losses requiring fast switching. Sic JFET it may be an excellent alternative since this switch shows an ultra low specific on-resistance and is also able to operate at high temperatures and high frequencies. Infineon has developed a I5kV, 0.5 on-resistance hybrid switch made up of a 1.5kV vertical SiC JFET and a 60V Si MOSFET in cascade configuration. The SIC JFET is a normally-on device, thus a cascade arrangement is necessary. This switch is aimed at resonant converters and power supplies. Recently, a 3x4.1 mm2 1.8kV SIC JFET die has been presented by Infineon with a current capability of 15A at an
on-state voltage of just 2V. The technology is said to be viable at voltages of up to 4SkV. In order to increase the current capability of high voltage (>3kV) JFET switches, new devices have been proposed adding a bipolar component. This is done to create a conductivity modulation inside the drift region and thus reducing the on-resistance as was done in the Si IGBT. In this sense, 1.6kV, 40A, 6mmx6mm SIJFET (Static induction carrier Injected JFET) has been demonstrated [30] with high switching speed and lower on-resistance than equivalent JFET. In a similar way, the BIFET (Bipolar Injection FET), a normally-on JFET with a pn junction at the cathode, which provides conductivity modulation in the drift region, has also been proposed by SiCED [13]. 3.CONCLUSION Summarizing today's position of SiC power devices in relation to the future progress of power electronic systems, we believe that SiC will become a major player on the way to the highest power density, i.e. highest efficiency, lowest volume and weight. At the moment we are on the the border to a deeper market penetration especially taking into account the new generation of Schottky Barrier diodes. A next crucial step will be the introduction of a switch based on SiC. Even if the design engineers would welcome the options provided by an SiC switch, the competition with silicon based charge compensated devices, especially in the large volume 600V class, is strong. For DC/DC converters the limit of Si high power converters up to 100 kW seems to be at a power density of 5W/cm3. The future goal of power density is 30W/cm3 for such DC/DC converters in a few years, which should be only possible by pure SiC design. It should be emphasized that the foreseeable higher device costs do not permit to replace Si power devices in general, but surely in advanced (mobile) application areas. Besides the further development of SiC devices, the realization of a long term stable performance needs intensive efforts to improve the packaging reliable, at elevated operating temperatures. After managing the broad use of SiC devices in mass volume applications and the availability of different
suppliers, next steps like the development of high end devices in the blocking voltage range of several kV, serving as a future tool for efficient energy generation and distribution, will follow.
REFERENCES [I] T. P. Chow, "High voltage SiC Devices for Power Applications-future prospects", in Proc. of the 10th Eurupean Conference on Power Elcctronics and Application, EPE 2003. [2] M. Treu, "A surge current stable and avalanche rugged SiC merged pn-Schottky diode blocking 600V, especially suited for PFC Applications", Proc. Of the ICSCRM 2005, Pittsburgh, September 2005. [3] F. Dahlquist, J. Hancock, M. Treu, R. Rupp, T. Riemann, "2nd Generation 600 V SiC Schottky Diodes Use Merged PN Schottky Structure For Surge Overload Protection", Proc. of the APEC 2006. [4] A Elasser, M. Kheraluwala, M. Ghezzo, R. Steigerwald, N. Krishnamurthy, J. Kretchmer, and T. P. Chow, A comparative evaluation of new silicon carbide diodes and state-of-the-art silicon diodes for power electronic applications, IEEE IAS Annual Meeting Conference Proceedings, pp. 341-345, 1999. [5] W. Wright, J. Carter, P. Alexandrov, M. Pan, M. Weiner, J.H. Zhao, Comparison of Si and SiC diodes during operation in three-phase inverter driving an AC induction motor, IEE Electronics Letters, 37(12), June 2001. [6] J.B. Dufrene, G. Carter, J.B. Casady, I. Sankin, D.C. Sheridan, W. Draper, M. Mazzola, High-voltage (600 to 3 kV) silicon carbide diode development, IEEE APEC 2001, pp. 1253-1257, Anaheim, 2001. [7] A.R. Hefner, Jr., R. Singh, Jih-Sheng Lai, D.W. Berning, S.Bouche, C. Chapuy, SiC power diodes provide break through performance for a wide range of applications, IEEE Transactions on Power Electronics, 16(2) , pp. 273280, March 2001. [8] A.R. Hefner, D. Berning, J. S. Lai, C. Liu, R. Singh, T.Kamgaing, J. Bernstein, Silicon carbide merged PiN schottky diode switching characteristics and evaluation for power supply applications, IEEE IAS Annual Meeting Conf Proc., pp. 2948-2954, 2000. [9] M. Bhatnagar and B. J. Baliga, Comparison of 6H-SiC, 3C-SiC, and Si for power devices, IEEE Trans. On Electron Devices, 40 (3), pp. 645-655, March 1993. [10] K. Shenai, R. S. Scott, B. J. Baliga, Optimum semiconductors for high-power electronics, IEEE Transactions on Electron Devices, 43(9), pp. 1811-1823, Sept. 1989. [11] D.Peters, H.Mitlehner, R.Schorner, R.Elpelt, and D.Stephani, "State of the art and technological challenges of SiC power MOSFETs designed for high blocking voltages", Proc. of the EPE 2003,Toulouse,ISBN 90-75815-07-7. [12] 4. B. Ozpineci, L. M. Tolbert, S. K. Islam, M. Hasanuzzaman, Effects of silicon carbide (SiC) power devices on HEV PWM inverter losses, The Annual
247
Proceedings of the 1st APE-2009 on Jan. 24-25, 2009 Conference of the IEEE Industrial Electronics Society (IECON'01), pp. 1187-1192, 2001. [13] Toumier D. et al. ,"A 4N-SiC High-Power Density VJFET as controlled current limiter ", IEEE Trans. on Industry Applications, vol. 39, 2003, pp. 1508-1513. [14] http://www.infinion.com ]15] http://www.powerelectronics.com
248
National Institute of Technology, Rourkela 769008, Orissa
AC LINK INVERTER BASED ON MULTIPLE CARRIER PWM FOR FUEL CELL

Mrs. Chandwani.H.B Reader in Elect Engg Dept, hinachandwani@yahoo.com Mrs. Matnani.M.K Student M.E (IE), Elect Engg Dept, meeta_matnani@yahoo.com The Maharaja Sayajirao University of Baroda, Vadodara Abstract - The paper suggests the development of a low cost fuel cell inverter system. The approach consists of a three-terminal push-pull inverter to boost the fuel cell voltage 48 V to 220 V ac (1 KHz frequency). High-frequency ac link conversion offers a possible way to reduce the number of power stages, in the form of a cyclo converter. It is shown that the control complexity in this converter can be reduced by adapting a multiple carrier PWM approach as a convenient way to implement a high frequency link inverter. 1. INTRODUCTION For the power conversion system where the linkage with ac system line is required, the HF (high-frequency) link converter topology has been attracting special interests recently, because it enables the power converters to be compact and light-weight. As for the HF link main circuit topology, the combination of a dc/dc converter and a PWM inverter as an interface with ac mains as shown in Fig. l(a), which is often called as dcdc converter type, is the most popular one at present. This kind of configuration, however, is applicable only when the active power flow is unidirectional. Previous attempts to provide power conversion with ac links were based on phase control and phase modulation techniques. In practice, the level of complexity was high. As a result, ac link methods have been confined to very high power levels. This paper suggests ac link-based power conversion with the same control complexity as conventional PWM power conversion. With lowvoltage dc input, the use of a simple ac link reduces the overall system complexity level. To avoid redundancy while keeping the desirable HF link many alternative circuit topologies have been proposed among them is cycloconverter type as shown in Fig 1(b).In this type dc source is 249 chopped to square wave. Then controls similar to ac-ac cycloconverter generate desired output. The power conversion is more direct. As shown in Fig 1(b) there are just three primary magnetic elements: the input interface to minimize fuel cell (dc source) ripple, the ac link transformer, and the output filter inductor. It turns out that the PWM cycloconverter, with or without natural commutation, and the conventional PWM inverter can be unified through a multiple carrier PWM framework. The features provided by such technology are (1) it allows inverters to use a high-frequency ac link bus instead of the conventional dc and (2) it allows multiple signals to be put onto the same PWM carrier or data stream for delivery to multiple switches. It also allows the use of ac circuit components (e.g. high-frequency transformers), in support of PWM converters, gate drives and switch control. This technology makes the power conversion process cheaper and more direct than other methods by eliminating some of the power stage processes and by simplifying switching device control signals. In near future, as the technology in battery cells progresses, the battery energy storage systems combined with other new energy system is expected to become more popular and one of them is Fuel cells. Fuel cells produce no polluting emissions, such as those one would get from a conventional turbine or internal combustion engine, and virtually no moving parts. This paper explores the possibility of a reduced- stage power conversion for a low-voltage fuel cell application. A static current-voltage relationship for a prototype fuel cell is shown in Fig. 2.The fuel source have an open circuit voltage of 1.15 V per cell at 800C and one atmosphere of pressure. Under load, the voltage drops abruptly and is typically 0.6 V per cell or even less.
In practice, a fuel cell stack with 64 cells would generate a working range of 32 V to 52 V Fig 2 provides only the static characteristic for the fuel cell. It is not be feasible to simply add batteries in parallel with the stack. A battery V-I curve is similar to a fuel cell but the operating point cannot be managed with a direct parallel connection, especially as fuel flow rates change. The open cell voltage of the 48V nominal fuel cell can reach 72V and exhibits a large initial drop as current increases.
Fig. 2 Typical fuel cell voltage- current characteristic
dc/dc converter type
Fig.1 High frequency link dc/ac converter
Challenges faced with Fuel cell: i) Slow dynamics ii) Step up needed. Hence while designing points to be considered are i) Isolation ii) Transformer frequency and iii) High voltage dc link or ac link.. Even though this environmental friendly, highly efficient energy resource is promising, the high cost of installing a fuel cell power plant is one of the main obstacles hindering its wide spread deployment. Currently, fuel cell production costs are decreasing, and have nearly achieved energy costs that are competitive with local utility rates. To further assist the reduction of cost, the price of the inverter portion of the fuel cell system must also decrease, while at the same time efficiency, reliability and suitable power quality levels must be maintained. When examining the typical fuel cell polarization and power curves a point of maximum power is observed. The MPPTAlgorithm adjusts the stack current and fuel flow, at optimal values for fuel consumption to supply the power required by the load.
2. FUEL CELL AND BATTERY ARRANGEMENT The fuel cell control system continuously adjusts the fuel flow rate to match the power output. However, these control systems involve pumps and compressors that are slow. The time constants involved in responding to a load change are several tens of seconds, depending on the size of the system. Hence it is necessary to use a temporary energy storage element to support sudden load changes. Battery banks, super capacitors or flywheel based energy storage schemes can be considered. At first glance it might be compelling to place the batteries in parallel with the fuel cell.
Decommut t C1 (t) Comparator 1
Carri 0/180 deg Phase shifter 1
Gate drive Modulating Function 0/180 deg Phase shifter 2 C2 (t) M1 (t) P 1
0/180 deg Phase shifter 3
M2 (t)
Comparat 2
P 2
Fig. 3 Multiple carrier pulse width modulator circuit
However, both the fuel cell and the batteries exhibit similar output V-I curves, representing a KVL problem for a direct connection. Additionally, it is desirable to control the charging of the batteries. Both of these requirements indicate that a power converter stage is needed to 250
interface fuel cell with the batteries. Hence different topologies that can be used are i) Direct low-voltage inverter followed by transformer step-up ii) dcdc converter cascaded with bridge inverter iii) Two-input current-sourced forward converter as the basis for the fuel cell system iv) Forward converter driving batteries and inverter v) Three-stage converter boost/forward/inverter vi) Boost converter, followed by ac link inverter and ac-ac converter for output. 3. OPERATION OF MULTIPLE CARRIER PULSE WIDTH MODULATORS It makes PWM inverters much more convenient from low-voltage dc inputs such as 12 V or 48 V systems, yet also supports conventional highvoltage buses. Allowing multiple signals to be put onto the same PWM carrier or data stream enables simplification of processes such as encoding a signal into left and right channels for a stereo amplifier, a single PWM signal can carry both, and a simple de-coding method recovers them independently. The same approach permits all three signals for a three-phase motor drive to be encoded on a single PWM signal, which can then be used either with three high-frequency links, or as the basis for a single common ac link poly phase inverter. Conventional PWM signals contain both high-frequency switching energy and
low-frequency modulation waveform energy. These waveforms cannot be coupled through small transformers. As a result, extra stages are needed for intermediate dc-dc conversion. The extra stages increases loss, parts count, and costs. Conventional PWM also requires a wide dynamic range of pulse widths from switching devices. This invention overcomes these limitations. This technology uses an enhancement of conventional PWM, namely, intelligent combinations of PWM signals, to build waveforms with pre- determined characteristics. The characteristics can include the elimination of low-frequency content the generation of twolevel, three-level, or multi-level PWM signals with or without low-frequency content, output with PWM characteristics that maintain duty ratios close to 50%, and conventional PWM outputs with effective switching frequencies double those of the actual devices. Under framework of Fig.3 families of multiple carrier PWM sequences can be constructed. Table 1 lists two carrier condition, three carrier condition are also simulated and shown in Fig.4 (d). All the sequences given in Table 1 are intended for HF link applications, the modulation and combination processes are selected to cancel out the carrier either with a phase shift or subtraction.
TABLE 1. COMBINATIONS CONDITIONS AND RESULTING PWM SEQUENCES

Carrier type 1 2 3 4 5 6 Triangle Triangle Ramp Ramp Ramp Ramp Phase Shifter 1 0 180 0 180 180 180
0 0 0
Phase Shifter 2 0 0 0 0 180 0

0
Phase Shifter 3 180 0 180 0 0 180

0
Combining method Add Subtract Add Subtract Add Add
Gate Drive Signal type 2-Level 3-Level 2-Level 3-Level 2-Level 2-Level
Output PWM equivalent Ramp PWM at double fswitch' Triangle PWM Triangle PWM Ramp PWM at double fswitch' Ramp PWM at double fswitch' Ramp PWM at double fswitch'
251
(a)
(b)
(c)
Fig. 4 (d) three carrier PWM sequence
Fig. 4 Two carrier PWM sequence generation process Parts (a), (b), (c) row 4, 6, 2 in table 1 respectively
dc Input Fuel Cell/ Battery
Push-Pull Inverter
HF Transformer
Cycloconverter
L o a d
Carrier Modulating Signal
Multiple Carrier PWM Sequence Generation
Selection of Positive/ Negative Converter
Current Polarity Detector
Fig. 5 Block Diagram for circuit operation
252
A decommutator is common in time-division multiplexing (TDM) base communication systems and has a similar function here, to segment the carrier into time slices that can be used as separate carriers. The decommutation rate fdecom is set to the same frequency as the carrier in case of a ramp waveform, or to twice the frequency of the carrier in the case of a triangle waveform. Each separate decommutator output is compared to the modulating function or with a phase-shifted image. The comparator outputs are combined arithmetically to generate a gate control sequence. To generate HF links, the gate control sequence must avoid low-frequency content yet retain information about the modulating function. There are two constraints in the switching of the cycloconverter a) the high frequency side of the cycloconverter can not be short circuited. b) The output side of the cycloconverter can not be open circuited. Conventional PWM sequences are generated by comparing a triangle or ramp carrier with a modulating function. In the multiple carrier PWM, the modulating process is performed in more than one manner, and the results are combined to produce a useful unified sequence. It is easy to generate multiple carrier PWM. Just as conventional PWM can be created with a sinetriangle combination into a comparator. So multiple carrier PWM can be generated with a bit of logic plus comparators. In the PWM cycloconverter method, the ac link voltage can be either a simple two-level square-wave or a threelevel waveform with short dead times. 4. PROPOSED SCHEME AND CIRCUIT OPERATION There are numerous topologies that can be chosen to meet the design criterion. Perhaps the most obvious topology is to use a push-pull inverter followed by a step-up transformer as shown in Fig 6. The topology extends well to extremely high power levels, since the cycloconverter method supports the use of SCRs as output switching devices. The design keeps the number of stages and magnetic elements low while still providing isolation. 253
An important disadvantage is that each output switch must be bilateral, in the sense that each must carry ac current and block ac voltage. This produces higher output voltage drop than single devices would, but it has the advantage of reducing the individual device current ratings. The output voltage must be sufficient to create a dual 220 V output to provide sufficient voltage for the cycloconversion. The primary-side inverter is push pull inverter using metal oxide semiconductor field effect transistors (MOSFETs) while the output bridge consists of eight unidirectional switches organized in four pairs. The thyristor implementation shown in Fig.6 supports link frequencies up to a few kilohertz. When a three-phase application is to be supported, the output bridge would have six pairs of devices. The multiple carrier PWM sequence, multiplied by a square wave, recovers an output waveform identical to that obtained with conventional PWM. The square wave can be applied to either the input or output bridge, while the multiple carrier sequence is applied to the other. The circuit operates from an input 1-kHz square wave. Simple logic is used with a current comparator to separate the positive and negative current conditions. Here a single ramp serves as carrier while modulating signal is phase shifted as per requirement. Since two carrier modulation is used and no decommutator is required. Cycloconverter commutation at the zero crossing of load current is an important control issue, well understood from conventional ac cycloconverter. In a cycloconverter-type system, the input is already two-level, and it is possible to identify a switch sequence that will deliver the same PWM output as obtained in the conventional case. 5. CONCLUSION The scheme suggested shows that the number of stages can be reduced, leading to a HF-link conversion approach. The reduction came from recognizing redundancy in the power processing. Without a dc link bus, rectifiers and filter components along with their associated losses are eliminated. Applying the techniques of multiple
carrier PWM cycloconversion results are exactly identical to conventional PWM techniques. The combination of a current-controlled input, adjust the average power demand from a fuel cell, battery buffer, and PWM cycloconverter provides a reduced parts count solution compared to conventional boost-forward-inverter cascade topologies. The combined converter isolates the fuel cell from its load both electrically and dynamically while reducing parts count and therefore reduces costs. REFERENCES
[1] P. T. Krein, X. Geng, and R. Balog, High-frequency link inverter based on multiple-carrier PWM, in Proc. IEEE Applied Power Electronics Conf., 2002, pp. 997 1003. [2] P. T. Krein and R. Balog, Low cost inverter suitable for medium-power fuel cell sources, in Proc. IEEE Power Electronics Specialists Conf.,2002, pp. 321326. [3] Rajesh Gopinath, Sangsun Kim,Jae-Hong Hahn, Prasad N. Enjeti, Mark B. Yeary and Jo W. Howze Development of a Low Cost Fuel Cell Inverter System With DSP Control, IEEE Transations On Power Electronics VOL. 19, NO. 5, SEPTEMBER 2004 [4] T. Kawabata, K. Honjo, N. Sashida, K. Sanada, and M. Koyama, High frequency link dc/ac converter with PWM cycloconverter, in Proc.IEEE Industry Applications Soc. Annu. Meeting, 1990, pp. 11191124 [5] P. M. Espelage and B. K. Bose, High-frequency link power conversion, IEEE Trans. Ind. Applicat., vol. IA-13, pp. 387394, Sept./Oct. 1977. [6] Philip T. Krein, Fellow, IEEE, Robert S. Balog, Member, IEEE, and Xin Geng, Student Member, IEEE High-Frequency Link Inverter for Fuel Cells Based on Multiple-Carrier 2004 PWMIEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 19, NO. 5, SEPTEMBER
254
ENHANCING ENERGY EFFICIENCY IN PHOTOVOLTAIC SYSTEM USING POWER CONDITIONING UNIT

Javed Khan Bhutto1, Prof. Mool Singh2
1 2
Department of EIC Engineering, JNIT, Jaipur-302022 Department of Electrical Engineering, MNIT, Jaipur Email-navyboy8686@yahoo.co.in power in PV systems. Furthermore, since solar energy systems are relatively expensive in comparison to other energy sources, it would be advisable to maximize energy efficiency so that the energy provided in relation to the system cost is optimized. Index term-DC-DC converter, Boost converter, Mppt photovoltaic 1. INTRODUCTION Most importantly, since the energy provided by clean energy sources, such as solar energy, can potentially supplant the use of traditional energy sources that pollute, it is environmentally wise to enhance the power output of renewable energy sources. In addition to enhancing efficiency [1], power conditioning is also used to facilitate energy storage. For example, some battery charging systems may require power conditioning circuitry to provide a desired current to charge batteries effectively. For this type of system, the use of an electronic circuit called a DC/DC converter is required to control the current supplied by the photovoltaic array to the battery. To prevent battery damage, the charging current supplied to the battery by the DC/DC converter is determined by the batterys state of charge. In other less sophisticated systems, circuitry may manage battery charging by simply connecting or disconnecting the solar panels from the battery depending on the state of charge of the battery. I have provided a block diagram of a typical battery charging system using a DC/DC converter circuit.
Abstract- Researchers in the field of photovoltaic faced with the task of making solar energy more efficient. There exists a variety of avenues in making this type of technology more productive, one branch of of research deals with the solid state properties of the materials used in solar cells. This type of research focuses primarily on the properties of thin film silicon and finding methods to manufacture high quality silicon at lower manufacturing costs. This paper is motivated by the fact that high quality semiconductor materials can have a significant effect on the energy efficiency of the solar panels themselves. A photovoltaic (PV) system may be a combination of several components such as a battery system, DC/AC conversion circuits, and other power conditioning devices in addition to the solar panels themselves. The paper I am presenting in this body of work concentrates more on the electronic means of enhancing energy efficiency in a PV system as well as describing energy storage and power grid integration techniques. This branch of power electronics is generally called power conditioning and in the present case is used to describe the management of electrical energy to effectively charge batteries, draw maximum power from the solar panels, or provide a high quality AC output. If a PV system is constructed with rudimentary electronics, it is possible for nearly all of the power generated by the solar panels to be wastefully dissipated as heat in the system components. In order to prevent the waste of potentially useful electrical energy, it is worthwhile to investigate modern and sophisticated means of managing electrical 255
Fig-1 Block diagram of a battery charging system.
Electronics can play a large role in enhancing the efficiency of a PV system. Furthermore, the use of appropriate electrical techniques enables solar generated electricity to be integrated into a power grid that may power a home or business. Aside from improving the quality of the solar panels themselves and the materials they require, power electronics can provide another means of improving energy efficiency in PV systems. 2. DC-DC CONVERTER In this section the principles of switching power conversion are introduced and details of the DC-DC boost converter circuits are discussed in steady state. A switching converter consists of capacitors, an inductor, and a switch. All these devices ideally do not consume any power, which is the reason for the high efficiencies of switching converters [3]. The switch is realized with a switched mode semiconductor device, usually a MOSFET. It is switched on and off by the driving square wave signals at the gate. If the semiconductor device is in the off state, its current is zero and hence its power dissipation is zero. If the device is in the on state (i. e. saturated), the voltage drop across it will be close to zero and hence the dissipated power will be very small [8]. During the operation of the converter, the switch will be switched at a constant frequency fs with an on-time of DTs, and an off-time of DTs, where Ts is the switching period 1/fs, D is the duty ratio of the switch and D is (1 - D) . The DC-DC converters can have two distinct modes of operation [9]: (1) continuous current conduction and (2) discontinuous current conduction. In continuous mode, inductor current never falls to zero in one switching cycle, Ts or at least one the switch or diode is conducting. Whereas in discontinuous conduction mode, the inductor current falls to zero before completing one switching cycle, Ts. In practice, a converter may operate in both modes, which have significantly different characteristics.
DC/DC converters are also useful in circuitry designed to draw maximum power from solar panels in what are called maximum power point trackers (MPPT). In the absence of a MPPT and depending on the load connected directly to solar panels, a great deal of the solar panels electrical power may be dissipated in the form of heat. In the interest of maximizing energy efficiency [1],[2]. MPPTs are connected between the solar panels and load to ensure that the solar panels are producing their maximum power despite variations in light intensity and/or other factors that may vary within the system. MPPTs may be used in the presence or absence of battery charging systems but are more often used in grid-connected systems that have no batteries. Considering that solar panels and batteries both supply DC voltages and most electronic devices require AC to power them, circuitry is usually required to convert this DC voltage into an AC voltage by what are called inverters. Once the DC voltage is converted into a 60 Hz waveform with proper amplitude, the AC voltage can then be introduced into a power grid. The presence of the inverter makes it possible to use solar energy to power almost any electronic device that runs on AC power. The following diagram demonstrates how an inverter may be used.
Fig-2 Block diagram of solar system using an inverter.
256
capacitor is assumed to be large to ensure a constant output voltage vo (t) 4. POWER CONSUMPTION IN THE BOOST CONVERTER The components, which cause power loss in the Boost-Converter when operating at high switching frequency, fsw, are the diode conduction loss, inductor conduction loss, MOSFET conduction loss and switching of the MOSFET. The conductance loss in the diode, Pdiode, is given by: Pdiode = Id *Vf *(1 D) ------- (1) where the Voltage Vf is the forward bias of the diode, Id is the input current and D is the duty ratio. Since for the Boost-Converter the inductor current, IL, is equal the input current, Id, equation (2) can be rewritten as: Pdiode = IL *Vf *(1 D) ----------- (2) The conduction loss in the inductor, Pinductor, is given by: Pinductor = IL2 RL -------------(3)
Fig-3 Ideal switch V, Duty ratio D, and switching period
3. BOOST CONVERTER The Boost Converter, as shown in Figure 4, is also known as the step-up converter. As the name implies its typical application is to convert low input-voltage to a high output voltage.
Fig-4 DC-DC Boost converter
where RL is the effective series resistance of the inductor. The conduction loss in PMOSFET,conduction, is given by:
Fig-5 Circuit of boost converter
the
MOSFET,
PMOSFET,conduction = IL2 Rds D -- 4) Where Rds is the MOSFETs drain-to-source on resistance. From equation (2), (3) and (4), the total conduction loss the Boost- Converter is: Pconduction loss = IL Vf (1 D) + IL2 RL + IL2 Rds D --(5) The MOSFET switching loss can be found by considered four different switching periods. They are namely the turn-on, conduction, turn-off and off period. The overlapping of the current and voltage
During the first time interval DTs of the switching period Ts, the closed switch connects the input through the inductor to ground and current starts to flow, the current through the inductor increases and the energy stored in the inductor builds up. The diode is reversed biased so no inductor current flows through the load, thus isolating the output stage. After the switch is opened in the second time interval DTs of the switching period, the output stage receives energy from the inductor as well as from the input. In steady state analysis, the output 257
waveforms during the turn-on period accounts for the switching loss and is given by [11]: Pturn-on = (Idso *Vdd* td *f) + f Ids tr [ Vdd /2 + (Vds,sat Vdd) / 3] -----(6) where f is the switching frequency, Vdd is the drain voltage, td is the turn-on delay time, tr is the turn-on rise time and Vds,sat is the saturated drain-to-source voltage, Idso is the drain-source leakage current of the MOSFET
Acm, then the output voltage of a differential amplifier is given by [10]: Vo = AdVdiff + AcmVcm----------------- (9) For a well-designed differential amplifiers have the differential gain Ad that is much larger than the common-mode gain Acm. A quantitative specification is the common-mode rejection ratio (CMRR), which is defined as the ratio of the magnitude of the differential gain to the magnitude of the common-mode gain. The CMRR is defined as [10]: CMRR = 20 log (|Ad| / |Acm|) --------- (10) From the circuit theory, the output voltage of a differential amplifier is given by the following expression: Vo = (R2 / R1) (V1 V2) ------------------(11) From equation (3.27) differential amplifier gain, G, is
5. DIFFERENTIAL AMPLIFIER
Fig-6 Differential Amplifier
G = (R2 / R1) -----------------------
(12)
A differential amplifier shown in Fig 6 is used as a transducer that converter the current signals into voltage signals. The voltage signals that provide the information about the current therefore can be fed back to the microcontroller for processing. The circuit consists of two input sources V1 and V2. It responds to both the differential signal Vdiff and common-mode signal Vcm. The differential signal is the difference between the input voltages and can be defined as: Vdiff = (V1 V2) -------------------------- (7)
The common mode signal is the average of the input voltages and is given by:
Vcm = 1/ 2 (V1 + V2) ------------------- (8) If the gain for the differential signal is denoted as Ad and the gain for the common-mode signal as 258
6. MAXIMUM POWER POINT TRACKERS At a given temperature and illumination level, solar cells supply maximum power at one particular voltage and current output called the maximum power point. In the interest of optimizing the power output of the PV system, the load should be operated at this maximum power point. However, the maximum power point can vary greatly as the illumination level and temperature of the solar cells change. In addition, the characteristics of the load may vary as well. To ensure ideal power transfer at the maximum power point, a dynamic system called a maximum power point tracker (MPPT) should be inserted between the solar panels and the load. This circuit will then respond to any variations at its input and output by ensuring that maximum power transfer is achieved [3].The MPPT is most commonly constructed out of DC/DC converters that use pulse width modulation (PWM) to vary the duty ratio as well as current and voltage sensors at its input. In order to calculate the power at the MPPTs input, the current and voltage measured by
the sensors are multiplied in a digital circuit. Considering the high power efficiency of the DC/DC converter, nearly all of the power at the input is transferred to the load [5].The MPPT typically uses an iterative process to find the maximum power point. This type of operation dynamically varies the duty ratio one direction (increase or increase in duty ratio) while calculating the power at its input. If the variation produces an increase in power at the MPPT input, the duty ratio is further increased in the same direction. If the power at the input decreases, the duty ratio is adjusted in the reverse direction. It should be mentioned that several other methods not mentioned here do exist [5]. Since the solar cell is simply a p-n junction, the current-voltage relation of the cell is similar to that of a diode. However, because a diodes p-n junction isnt exposed tolight the photocurrent must also be taken into in the solar cell I-V relation [1]. This can be expressed as follows: I-Vcharacteristic of solar cell: I = Iph Io(eqv/kT ) The first term represents the photocurrent which is generated by the incoming light on the p-n junction. It should be noted that the photocurrent, Iph , is directly proportional to the light intensity. The equivalent circuit shown in Figure 7 can model the behavior of an ideal solar cell [6]. The symbol on the left represents a current source.
A solar cell behaves much like a current source in parallel with a diode. When the terminals at the output of the solar cell are shorted, no current will flow through the internal diode but will all flow to the output. In this short circuit condition, the shortcircuit current Isc will be provided at the output and the output voltage will be zero. When the terminals are connected to a load with finite resistance, some but not all of the current will flow through the diode. Lastly, if the load connected approaches the open circuit condition, all the current will flow through the internal diode, the output current will be zero, and the output voltage will be Voc [1].
Fig 8. I-V characteristics of solar cell under illumination and darkness.
Fig-7 The equivalent circuit of a solar cell.
A graphical representation of the I-V characteristic is shown in Figure 8. It can be seen that as the illumination intensity rises, the curve is shifted increasingly downward. This downward shift of the graph occurs because the photocurrent Iph is in the opposite direction of the current produced in a conventional diode. This can also be understood by noting that the two terms are opposite in magnitude. In photovoltaics, the portion of the graph resulting from illumination is what is of interest. To simplify analysis and to make the power provided by the solar cell(s) positive in magnitude, the illuminated portion of the graph in figure 8 is conventionally inverted upside-down [1]. This can be seen by comparing the illuminated I-V curve of figure 3.3 and the customarily inverted I-V curve of figure 3.4. Although a solar cell can be operated anywhere on its current vs. voltage curve, it is desired to operate the array so that maximum
259
power output results. Considering the I-V characteristic of a solar cell, the power is maximized at a unique point on the I-V curve (see figure 3.4) under a specific illumination level and temperature. As the illumination level varies, the IV curve will for the most part be shifted up or down and will consequently create different maximum power points for each illumination level. This shifting of the I-V curve can best be understood by realizing that the photocurrent term Iph in equation will shift the curve vertically along the current axis as Iph is varied. The maximum power point PMPP is located on the knee of the I-V curve and is indicated at a voltage of VMPP and current of IMPP in figure9[1]. This illustration shows both the I-V and power curves on one graph. Note how the maximum power point is located at the knee of the I-V curve.
their respective power conditioning systems. Furthermore, since inverters exhibit problems in terms of reliability, attempts are being made to enhance the lifetime of inverters for PV systems. Inverters are also one the least energy efficient components in state of the art PV systems with an energy efficiency in the 75-90% range. With contemporary solar materials that are somewhat costly and typically exhibit efficiencies less than 20%, it is in the interest of the consumer to maximize the power efficiency of the entire PV system. This goal should be motivated by both economical and environmental concerns. Economically speaking, since PV systems are relatively expensive, power efficiency should be maximized so that the investor can get the largest energy return on his investment. Correspondingly and possibly more important, energy efficiency should be increased so that the net CO2 emissions prevented by the system is as large as possible. As the efficiency and costs of alternative energy sources is optimized, consumers and the business community may choose to use clean energy not only because it is the right thing to do for the environment, but because it will make financial sense.
REFERENCES [1] S. Johnston, P. Gostelow, E. Jones, R. Fourikis, Engineering & Society: An AustralianPerspective,HarperEducational, Australia, 1995. [2] H. J. Moller, Semiconductors for Solar Cells, Artech House, Inc, Norwood, MA 1993. [3] P. Midya et al, Dynamic Maximum Power Point Tracker for Photovoltaic Applications, 27th Annual IEEE PESC, IEEE Computer Soc. Press, New York, USA, 1996, vol.2, pp. 1710-16. [4] C. Hua and C. Shen, Study of Maximum Power Tracking Techniques and Control of DC/DC Converters for Photovoltaic Power System, 29th Annual IEEE PESC, IEEE Computer Soc. Press, New York, USA, 1998, pp. 86-93. [5] P. Midya et al, Dynamic Maximum Power Point Tracker for Photovoltaic Applications, 27th Annual IEEE PESC, IEEE Computer Soc. Press, New York, USA, 1998, pp.679-683. [6] Shengyi Liu, Maximum Power Point Tracker Models, http://vt.engr.se.edu, January 2001. [7] D. Linden, Handbook Of Batteries, McGraw-Hill, New York, 1995 [8] Robert W. Erickson, Fundamentals of Power Electronics, Chapman & Hall, 115 Fifth Avenue, New York, NY 10003, 1997.
Fig.9 I-V and power curves of a typical solar cell.
7. CONCLUSION When the PV array is used as a source of power supply to charge a 12V lead acid battery, it is necessary to use the MPPT to get the maximum power point from the PV array. For this thesis project, the MPPT is implemented by using a Boost-Converter, which is designed to operate under continuous conduction mode and a microcontroller to control the PWM signals to the Boost-Converter and also to monitoring the state of charge of the battery. The Incremental Conduction Algorithm is used as the control algorithm for the MPPT. Experimental results have shown that the MMPT has the conversion efficiency of 87.28% and tracking accuracy of 97.6%. Current research within the field of power electronics is being carried out with the aim of enhancing energy efficiency of PV systems and
260
National Institute of Technology, Rourkela 769008, Orissa. [9] N. Mohan, T.M. Undeland, W.P. Robbins, Power Electronics: Coverters Application and Design, John Wiley and Sons Inc., USA, 1995, pp. 162-183 [10] A. R. Hambley, Electronics: A top-down Approach to Computer-Aided Circuit Design, Prentice Hall Inc., NJ, USA, 1994. [11] Simon S. Ang., Power Switching Converters, Marvel Dekker, New York, 1995 [12]K.Ktse. S.h Chung A Novel Maximum Power Point Tracker For PV panels using Switching frequency Modulation IEEE Trans on power electronics vol 17 no 6 Nov 2002 [13] Hussein, K. H., Muta, I., Hoshino, T. and Osakada, M.Maximum Photovoltaic Power Tracking: an Algorithm for Rapidly Changing Atmospheric Condition,IEE Proc.-Gener. Transm. Distrib, Vol. 142, pp. 59 64 (1995). [14] Matsui, M., Kitano, T., Xu, D. H. and Yang, Z. Q., A New Maximum Photovoltaic Power Tracking Control Scheme Based on Power Equilibrium at DC Link, Proceedings of the IEEE Industrial Electronics Society 25th Annual Conference, pp. 804 809 (1999). [15] Bodur, M. and Ermis, M., Maximum Power Point Tracking for Low Power hotovoltaic Solar Panels, Proceedings of the IEEE Electro Technical Conference, M., A Fuzzy Logic Based Photovoltaic Peak Power Tracking Controller, Proceedings of the IEEE International Symposium on Industrial Electronics, pp 300 325 (1998). [16] Mahmoud, A. M. A., Mashaly, H. M., Kandil, S. A Khashab, H. E. and Nashed, M. N. F., Fuzzy Logic Implementation for photovoltaic Maximum Power Tracking, Proceedings of the IEEE International Workshop on Robot and Human Interactive Communication, Osaka, Japan, 27 29 (2000).
261
DIRECT ACTION FUZZY PID CONTROLLER STRUCTURE

Sarika V Tade1, Amol A kalage2, Arun D Sonar3
Abstract- The aim of this paper is to present a fuzzy logic controller(FLC)for servo system used for position control and compare the time specification performance between conventional controller and FLC.Two types of controller namely Proportional integral derivative(PID)and FLC will be used to control the output response. This project consists of hardware equipment and software design. The hardware parts involve servo system and data acquisition system with a personal computer. The software part includes programming and simulation using Matlab software. Finally, the software will be integrated with hardware to produce a desired response. In this work, the simulation and implementation of FLC for servomotor has been carried out under different working conditions. KeywordsPID, fuzzy logic, controlsystem,servomotor,Matlab. position
1.0 INTRODUCTION Servomotors are widely used in many automatic systems, including drive for printers, tape recorders, robotic manipulators, machine tools, rolling machines etc. Proportional-integral derivative(PID)controllers usually control these
motors. Such controllers will be effective enough if the speed and accuracy requirements of control systems are not critical under varying environments of the systems. The main purpose of this research is to design a FLC for servomotor used for position control. The application of FLCs to servo systems produces results superior to classical controllers and their robustness qualities are acknowledged. For instance, it is seen that, if there is a change in system parameters or load disturbances, the response of system due to proportional-integralderivative (PID) controller is considerably affected and PID controller needs retuning . However, FLCs preserve the desired response over wide range of system parameters and load disturbances. In this work, the simulation and implementation of FLC for DC servomotor, have been carried out under different working conditions.. It also involves understanding on how to interface personal computer with position control system through data Acquisition System (DAS). This paper can be divided into two parts; namely hardware and software. The hardware part involves interfacing servo System and Data acquisition system with a personal computer the software part includes programming and simulation using matlab software
______________________________________________________________________
1
Lecturer in Electrical Engineering Dept. at SND College of Engineering, Yeola (M.S.) sarika.tade@rediffmail.com 2 Lecturer in Electrical Engineering Dept. at SND College of Engineering, Yeola (M.S.) amolkalage@rediffmail.com 3 Asst. Prof. in Instrumentation Engineering Dept. at D Y Patil College of Engineering, Pune (M.S.) arundsonar21@yahoomail.com
262
A. PID CONTROLLER PID stands for Proportional-IntegralDerivative. When the control problem is to regulate the process output around a set point, it is natural to consider error as an input, even to a fuzzy controller, and it follows that the integral of the error and the derivative of the error may be useful inputs as well. In a fuzzified PID controller, however, it is difficult to tell the effect of each gain factor on the rise time, overshoot, and settling time, since it is most often nonlinear and has more tuning gains than a PID controller. A systematic tuning procedure would make it easier to install fuzzy controllers, and it might pave the way for auto-tuning of fuzzy controllers.
From there it is easier to go to fuzzy control. Each step will be investigated in the following. Controller Kp Ti Td ----------------------------------------------------P 0.5Ku PI 0.45Ku Tu/1.2 PID 0.6Ku Tu/2 Tu/8 Table1: Ziegler- Nichols rules If n and l are zero, then Kp should be high in order to ensure that the process output x is close to the reference Ref Furthermore, if l is nonzero, a high value will make the system less sensitive to changes in the load l . But if n is nonzero kp should be moderate, otherwise the system will be too sensitive to noise. If the process dynamics are considered, the closed loop system will normally be unstable if Kp is high. Obviously the setting of Kp is a balance between the control objectives: stability, noise sensitivity, and load regulation. A PID controller may be tuned using the ZieglerNichols frequency response method. Procedure (a) Increase the proportional gain until the system oscillates; that gain is the ultimate gain Ku. (b) Read the time between peaks Tu at this setting. (c) Table 1 gives approximate values for the controller gains. FUZZY LOGIC CONTROLLER (FLC) Fuzzy Logic Controllers are based on the fuzzy set and fuzzy logic theory originally advocated by Lotfi A. Zadeh. Also, FLC have common feature of not requiring a detailed mathematical model and lead to much faster and accurate controllers for servo systems . B.
Fig1: Proportional control with load l and noise n. PID controllers may be tuned in a variety of ways, including hand-tuning, Ziegler-Nichols tuning, loop shaping, analytical methods, by optimization, pole placement, or auto tuning. Furthermore, fuzzy controllers show similarities with PID controllers under certain assumptions. But there is still a gap, it seems, between the PID tuning methods and a design strategy for fuzzy controllers of the PID type. Before implementing the fuzzy controller: 1. Tune a PID controller 2. Replace it with an equivalent linear fuzzy controller 3. Make the fuzzy controller nonlinear 4. Fine-tune it It seems sensible to start the controller design with a crisp PID controller, maybe evenjust a P controller, and get the system stabilised.
263
Due to its heuristic nature, fuzzy logic control is much closer in spirit to human thinking and natural language than traditional logic systems. The FLC provides an algorithm, which can convert the linguistic control strategy, based on expert knowledge into an automatic control strategy. In particular, the methodology of FLC appears very useful when the plants are too complex for analysis by conventional quantitative techniques or when the available sources of information are interpreted qualitatively, inexactly or with uncertainty. The use of FLC significantly changes the approach to control of drives. A conventional controller adjusts the system control parameters on the basis of a set of the differential equations, which describes the model of drive system. In a fuzzy controller, the adjustments are made by a fuzzy rule based expert system, which is a logical model of human behaviour of the plant operator. An FLC usually gives better results than those of conventional controllers, in terms of the response time, settling time and particularly in robustness. The robustness of FLC is commendable feature in motor drive applications, where, the system parameters are widely varying during plant operation. Due to nonlinear structure of the FLC, the main design problem lies in the determination of the consistent & complete rule set and the shape of membership functions. However, FLCs design is made easier by friendly. The control structure of the servomotor is shown in Fig.2. 2 .servo system
Following set of equations represents the system dynamics (1) (2) (3) (4) (5) (6) (7) Where position of the motor shaft Angular speed of the motor shaft Reference position ea armature voltage back emf armature current armature resistance armature inductance KA Gain of input signal amplifier back emf constant Kb KT torque constant
264
TM motor torque TL load torque B Viscous damping coefficient of the motor referred J Moment of inertia of the motor and load referred to the motor n gear ratio Kp potentiometer sensitivity in Volts/rad The general configuration of FLC, based on state evaluation fuzzy control rules has been considered . We select an error signal (e) and its rate of change (de) as linguistic variables to be used in the premise of control rules. These are defined as (8) (9) where r&c, are reference and actual shaft positions respectively and KT is tacho-generator constant in V/rad/s. We choose seven fuzzy sets that are specified on the domains of e and de: PL M, ,ZE,NS,NM,NLwhere P,N,ZE,S,M,L correspond, respectively, to positive negative, zero, small, medium and large. Thus, for instance, PS stands for positive small and NM stands negative medium, etc. In the same manner, the seven fuzzy sets are defined the domain of definition of the input u.The membership functions that are chosen for both input and output variables are triangular. A total number of 19 control rules are formed to give robust performance to the controller. The rule base is shown in Table 2. We can interpret it linguistically in the following form i f e A and de B, then u C where A, B, and C may be any of the linguistic terms like PM, NS, etc. It is to be noted that Z*N in this table implies a superposition of both ZE and NB. Further note that lack of specification of the table element does not imply that there is no control action for these cells. In fact, based on overlapping nature of the control rules, the control action is still
inferred fiom rules surrounding the unspecified region. The control action is then defuzzified as u Where Wi and ui denote the weight of the implication and the inference outcome of each control rule ,respectively. e\de NL NM NS ZE PS PM PL NL NL NL NL NL NL NL NL NM NM NM NM NM NM NM NM NS ZE NL NM NS NS ZE ZN PS PM PM PL PS PM PL NM PM PL PM PL ZN PM PL PS PM PL PM PL PM PL PM PM PL
Table2 Rule base for fuzzy logic controller

3. THE DC MOTOR CHARACTER~STICS UNDER CLOSED-LOOP
The motor simulation results obtained through Fuzzy Logic Toolbox and SIMULINK in MATLAB are discussed in this section. The dc motor chosen for this study has a voltage range of 30 V and a maximum current of 3 A. The motor parameters are given in Table 2. With these parameters of the motor and system dynamics expressed by equations (1)-(7) the transfer function of the system is given by
=344.585/
From Table 2 it is observed that some of the cells contain the superposition of more than one consequent for same combination of antecedents. For example we observe that in cell (3,5) the antecedents are (error is NS) and (error-rate is PS) If we choose the consequent of
265
(control input is ZE), the system response is shown consequmt of (control input is NB), the system response is shown in Fig. 5 (b). It can be seen that the system response one case is underdamped while in the outer case it is overdamped. Instead if we choose both these consequent and superimpose the two rules, the system response is shown in Fig. 5 (c). It can not be seen that the response is has overshoot nor has a large settling time. `
Table 3 DC motor parameters Potentiometer sensitivity( Kp ) Signal amplifier gain ( KA) Back emf constant ( K, ) Armature resistance ( R, Armature inductance ( L , ) Motor torque constant ( K , ) Combined moment of inertia motor sh& & load referred to the motor shaft side (J) Viscous damping coefficient of the motor referred to the motor shaft side (B)
in Fig. 5 (a). Similarly, if we choose the (b). It can be seen that the system response one
5.093 V/rad 1 15x10-3 V/rad/s 4.67 170mH 15x10-3Nm/A 42.6x106 Kg-m2 47.3x10-6 Nm/rad/s
fig-3 Improved response obtained from FLC by superposing consequent parts.
To compare the effects of the FLC on the system, we also design a PID controller. For the PID controller a proportional gain , integral gain and derivative gain are from Ziegler-Nichols method. For the above system parameters the typical responses for PID controller and fuzzy controller are shown in Fig. 6. It is evident from this figure that the fuzzy logic controller has smaller
overshoot and faster settling time compared to the PID controller.
From Table 2 it is observed that some of the cells contain the superposition of more than one consequent for same combination of antecedents. For example we observe that in cell (3,5) the antecedents are (error is NS) and (error-rate is PS) If we choose the consequent of (control input is ZE), the system response is shown in Fig. 3 (a). Similarly, if we choose the consequmt of (control input is NB), the system response is shown in Fig. 3
266
National Institute of Technology, Rourkela 769008, Orissa. fig-4 Response of servo motor a)without controller b)with PID controller c)with fuzzy controller.
4 CONCLUSIONS It is seen that the fuzzy controller preserves the desired response, even in the presence of load disturbance and varying control environments. This ensures the controllers robustness, the choice of rules and membership functions have considerable effect on fizzy controller performance, e.g., rise time, settling time, overshoot etc.It is observed that using the superposition of different consequent active at particular region of domain, for the same combinations of antecedents, performance of FLC is improved considerably in terms of settings time and overshoot. The combination of two sets of rules with same antecedents but different consequent reduces the settling time and overshoot. Further it is possible to find some other combinations of antecedents and consequent to reduce rise time to give best performance to servo motor, i.e., to rotate the shaft of the motor to a set point with a faster response without overshoot..
REFERENCES [1]T. J. Ross, Fuzzy Logic with Engineering Applications, [2]McGraw-Hill, Inc., Chap. 13, pp 469-5 15, 1995 [3]L. J. Huang and M. Tomiruka, A Self - Paced FuzzyTracking Controller for Two - Dimensional Motion Control, W E Trans. on Sys. Man and Cybem., Vol. 20, NO. 5, pp. 1115-1 123, Sept./Oct. 1990. [4] W. Li, A Method for Design of a Hybrid NeuroFuzzyControl System Based on Behaviour odelling, F ETrans. on Fuzzy Systems, Vol. 5, NO. 1, pp 128-137,Feb. 1997. [5]Bolognani and M. Zigliotto, Fuzzy Logic Control of as witched Reluctance Motor Drive, IEEE Trans. on Indl.Applications, Vol. 32, NO. 5, pp. 1063-1068, Sept./Oct.I996 [6]Mohdfuaad Rahmat, performance comparison between PID and FLC in position control system of DC servomotor journal teknologi, 45(D), 2006
267
Hardware Implementation of 3-phase Inverter by SHE Technique using PIC16F877A Microcontroller

*Lect. Brijesh Kumar, Prof. Abhijit Dasgupta Electrical Engineering Department, KIIT-University Abstract- A 3-Phase Inverter can be constructed by using SHE technique and microcontroller. If the neutral is isolated, 3rd harmonic eliminates automatically. Lower order harmonic can be eliminated by firing the inverter at calculated Notch angles and remaining higher order harmonic can be filtered out. By solving the notched square wave output by Fourier transform, and solving the various values of theta angles by iteration method, the notches angles are determined. And by programming the microcontroller and generating pulses at the desired interval to the inverter gate pulses, inverter can be constructed. The fundamental voltage, when compared with other harmonic elimination technique like SPWM, having same switching frequency is very good. 1. INTRODUCTION SELECTIVE harmonic elimination (SHE) is a long-established method of generating pulse-width modulation (PWM) with low base band distortion. Originally, it was useful mainly for inverters with naturally low switching frequency due to high power level or slow switching devices. Conventional sine-triangle PWM essentially eliminates baseband harmonics for frequency ratios of about 10:1 or greater, so it is arguable that SHE is unnecessary. However, recently SHE has received new attention for several reasons. First, digital implementation has become common. Second, it has been shown that there are many solutions to the SHE problem that were previously unknown. Each solution has different frequency content above the baseband, which provides options for flattening the high-frequency spectrum for noise suppression or optimizing efficiency. 268 Third, some applications, despite the availability of high-speed switches, have low switching-to-fundamental ratios. One example is high-speed motor drives, useful for reducing mass in applications like electric vehicles. SHE is normally a two-step digital process. First, the switching angles are calculated offline, for several depths of modulation, by solving many nonlinear equations simultaneously. Second, these angles are stored in a look-up table to be read in real time. Much prior work has focused on the first step because of its computational difficulty. 2. THEORY The output fundamental voltage V phase has the maximum amplitude (4Vd/pi) at square wave, but by creating three notches, the undesirable lower order harmonics of a square wave can be eliminated and the fundamental voltage can be controlled. Taking the Fourier transform of the notched square wave, following three equations are obtained. b1=1-2Cos(1)+2Cos(2) 2Cos(3); b5=1-2Cos(5*1)+2Cos(5*2) 2Cos(5*3); b7=1-2Cos(7*1)+2Cos(7*2) 2Cos(7*3); b1=4Vs/pi; Maximum Amplitude of the fundamental sin wave. Let the output be calculated in per unit. So, in-order to obtain the variable fundamental output voltage i.e. b1=A_pu where (A_pu = 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1) and the lower order harmonic 5th and 7th are equated to zero.
3. OBSERVATION By controlling the notches angle it is possible to control the output voltage from 0.1pu to 1pu, by minimizing the lower order harmonics. However the 5th and 7th harmonics cannot be maintained at zero. And if the 5th and 7th harmonics are set nearly equal to zero, output voltage can be varied from 0.87pu to 0.95pu only.
Tabulation
269
alpha1=6.886; alpha2=12.739; alpha3=88.75
270
4. CONCLUSIONS 1. The notch angles obtained by the programming are supposed to eliminate the lower order harmonics. This has been verified by simulation and hardware implementation. 2.Since the experiments were carried out at low voltage the protection part of the IGBT required at higher voltages was not studied. This can be done in the future. 3.The SHE was simulated for all the sets calculated. However hardware was tested for one set only. It can be seen from simulation that fundamental output voltage varies with different sets of notch angles. A
microcontroller program which can fetch the required set of notch angles as per the requirement of the fundamental voltage needs to be developed. a) As has been observed SHE suppresses the LOH but THD is not really reduced as higher order harmonics shoots up. This requires design of filters. This can be studied in future. b) The solution of non linear equation was tried by two methods. All such solutions suppresses the required LOH(3rd,5th,7th etc.), further work needs to be done in the methods of solution which will optimize the total harmonic distortion.
271
FLOWCHART
START
Enter the value of Input DC Voltage=Vs Calculating output voltage in Per Unit
Assume Max I/p Voltage=M_pu; Max O/p Voltage =A_pu th1=0 to pi/2 ( 100 values) th2=0; th3=0; (Both 100 values)
th53=(1/5)*acos(real(cos(5*th2)+0.5-cos(5*th1))); th72=(1/7)*acos(real(cos(7*th1)+cos(7*th3)-0.5)); th11=(1/1)*acos(real(cos(1*th2)+0.5-cos(1*th3)-
b1=(M_pu)*(1-2*cos(th1)+2*cos(th2)-2*cos(th3)); b5=(M_pu/5)*(1-2*cos(5*th1)+2*cos(5*th2)2*cos(5*th3));
Is b1=A_pu; b5 &b7 less then 0.1
Get the value of th1 th2 and th3
END
272
National Institute of Technology, Rourkela 769008, Orissa. REFERENCES 1. H. S. Patel and R. G. Hoft, Generalized harmonic elimination and voltage control in Thyristor inverters: Part I Harmonic elimination, IEEE Trans. Ind. Applicat., vol. IE-9, pp. 310317, May/June 1973. 2. Generalized harmonic elimination and voltage control in Thyristor inverters: Part IIVoltage control technique, IEEE Trans. Ind. Applicat., vol. IE-10, pp. 666673, Sept./Oct. 1974. 3. P. Enjeti and J. F. Lindsay, Solving nonlinear equations of harmonic elimination PWM in power control, Electron. Lett., vol. 23, no. 12, pp. 656657, June 1987. 4. P. N. Enjeti, P. D. Ziogas, and J. F. Lindsay, Programmed PWM techniques to eliminate harmonics: A critical evaluation, IEEE Trans. Ind. Applicat., vol. 26, pp. 302316, Mar./Apr. 1990. IEEE Journals:[1] P.Enjeti,J.F.Lindsay, Solving non-linear equations of Harmonic elimination PWM in Power Control, Electron. Lett., vol. 23, no. 12, pp.656-657-44, June 1987. [2]. K. Taniguchi, Y. Ogino and H. Irie, PWM technique for Power MOSFET Inverter, IEEE Trans.PowerElectron. ,vol. 3, pp. 328-334, July 1988. [3] J.R.Wells, X.Geng,P.L.Chapman, P.T.Krein, Modulation Based Harmonic Elimination, IEEE Trans.PowerElectron. , article accepted for inclusion in future issues. [4] S.R.Bowes,P.R.Clark, Simple Microprocessor Implementation of New Regular-Sampled Harmonic Elimination PWM Techniques, IEEE Trans. Ind. Applicat., vol. 28, no. 1, pp. 8995, Jan./Feb. 1992. [5] T.Yiliang, C.Wenjin, X.Xiaorong, H.Yingduo, M.Wong, 80C196MC Microcontroller-based Inverter-Motor Control And IR2130 Six-Output IGBT Driver, IEEE Trans. Ind. Applicat., pp. 655657, May. 1992. [6] V.G.Agelidis,A.Balouktsis,I.Balouktsis, On Applying a Minimization Technique to the Harmonic Elimination PWM Control: The Bipolar Waveform, Electron. Lett., vol. 2, no. 2, pp.41-44, June 2004. [7] J.N.Chiasson,L.M.Tolbert,K.J.McKenzie,Z.Du, A complete Solution to the Harmonic Elimination Problem, IEEE Trans.PowerElectron. , vol. 19, No. 2, pp. 491-499, March 2004. [8] J.W.Chen, T.J.Liang, S.Wang, A Noval Design and Implementation of Programmed PWM to Eliminate Harmonics,No-0-7803-9252-3/05/$20.00 2005 IEEE. [9] J.R.Wells, B.M. Nee, P.L. Chapman, P.T.Krein, Selective Harmonic Control: A GEnerat Problem Formulation and Selected Solutions, IEEE Trans.PowerElectron., vol. 20, No. 6, pp. 1337-1345, Nov. 2005. AUTHORS Lect. Brijesh Kumar: ( Email: brijesh_kiit@yahoo.co.in) Working as Lecturer in Dept. of Electrical Engineering, KIIT University Bhubaneswar-751024, India since 2004. Received BE in electrical Engineering from IGIT Sarang ( Utkal University) in 2001. and M.Tech in Power Electronic And Derives in Electrical Dept. KIIT-University. His area of interest is power electronics, electrical machines, electrical measurements and microcontroller. Professor Abhijit Dasgupta: (Email:abhijit_dasgupta@vsnl.com) Belongs to the Electrical Engineering Department,KIIT-University,Bhubaneswar751024, India. Prof. Dasgupta received BE in electrical Engineering from Regional Engineering collage, Durgapur, India, in 1977, and M.Tech in Power electronics from IIT, Kanpur in 1980. He has worked in industry before joining academics. His areas of interest are power electronic circuits, SMPS, Microcontroller applications in drives.
273
PIEZOELECTRIC TRANSFORMERS: THE KEY TO THE FUTURE INTEGRATED MOSFET & IGBT GATE DRIVERS
Matada Mahesh*
Abstract In the coming era of ubiquitous computing, nanotechnology, increasingly compact or miniaturized power sources and preferably mobile & satellite systems are expected. In this paper, the application of piezoelectric transformer as a recent complementary gate driver for power metaloxide semiconductor field-effect transistors and insulated gate bipolar transistors is presented. This type of transformer has a high integration capability. A new design method has been mentioned based on an analytical Mason model in order to optimize the efficiency, the available power at the transformer secondary ends, and the total volume. This design method takes into account mechanical losses and heating of the piezoelectric material; it can be extended to predict the characteristics of the PT: gain, transmitted power, efficiency, and heating of piezoelectric materials according to load resistance. 1. INTRODUCTION the tendency in power electronics is to integrate the components on a single substrate in order to reduce the equipment volume and particularly its thickness. The electrical insulation constraint of gate drive circuits for modern power electronic switches becomes, thus, very strong. Commonly, magnetic core-based transformers are used to achieve this insulation. Their manufacturing requires a winding process, which not only increases the manufacturing cost, but also prohibits the full automation of the
URRENTLY,
mounting process. This is advantage has prompted research efforts [1][2] onplanar electromagnetic transformer and inductor wound on a printed circuit board (PCB). However, the manufacturing of these structures is complex, and sometimes the galvanic insulation is achieved by air, which limits the dielectric rigidity. Moreover, the winding behaves like an antenna and radiates electromagnetic (EM) fields, which induces electromagnetic interference (EMI) problems.The solution proposed in this paper consists of using piezoelectric transformers (PTs) to achieve a very efficient and integrated electrical insulation. PTs have several inherent advantages over conventional magnetic transformers: low profile, low cost, low EMI, no winding, high efficiency, high power density, high operating frequency, and they suited for automated manufacturing. The use of a high rigidity dielectric material means a high degree of insulation. The dielectric breakdown field can be greater than several kV/mm. Moreover, a PT transfers electric energy via an electromechanical coupling (i.e., acoustic wave) between the primary for step-up or step-down voltage conversion. In a conventional magnetic core transformer, this function is done by the magnetic field, coupling the primary to the secondary windings. Consequently, a piezoelectric transformer is advantageous regarding the EMI. In a first approach, already presented in [3], we have studied the possibility to use a PT to achieve the electric isolation in a gate drive circuit for power electronic switches such as metal oxide semiconductor field-effect transistors (MOSFETs) and insulated gate bipolar transistors (IGBTs). Feasibility was proved in the case of a singleswitch structure where the gate drive circuits are referred to the ground of the power structure. The purpose of this complementary gate driver is to get 274
Research Scholar, NIT Rourkela.
enhanced switching performances for both turn-on and turn-off operations in an inverter-leg structure. Schematic structure of the gate driver is presented in Fig. 1. The piezoelectric transformer is supplied at a constant frequency which is its mechanical resonance frequency. The driving signal is transmitted by pulsed square wave modulation.
secondary electrodes. Because PT is based on the transmission of an acoustic wave, it must work close to the mechanical resonance of the structure. Several resonance frequencies exist as explained hereunder. The choice of the resonance mode is essential regarding the performances of the PT.
Signal is demodulated in the secondary part of the driver by a demodulator circuit (full wave rectifier) which drives the power transistor grid. In this paper, the design method of a PT gate driver is presented. Based on an analytical Mason model, this design method gives the minimal geometrical size of a multilayer PT. This analytical method is applied to predict the characteristics of the PT: gain, transmitted power, efficiency, and heating according to load resistance. Results obtained from successful implementations of the PT in a complementary gate drive circuits in a lowprofile power converter are included.
2. PIEZOELECTRIC TRANSFORMER A. Principle
The optimal structure of the piezoelectric transformer for the gate driver is the multilayer one operating in thickness mode, as presented in Fig. 2. The functioning principle is based on a double electromechanical conversion of energy (reverses and direct piezo-electric effect). If we impose an alternating voltage on primary electrodes, an alternating vibration of the structure is generated which induces an alternating voltage at the
An acoustic wave, propagating from one end along the thickness of the structure toward the other end will be reflected and will travel back toward its origin. If other waves have already been created, the first wave will interfere with them as it travels through. At particular frequencies (called resonance frequencies or modes), this interference produces standing waves. At specific points of the PTs thickness, named nodes, the medium is always at rest and at antinodes the wave amplitude is maximum. Therefore, the only standing waves that can exist are obtained when the structures thickness is a whole number of the half-wave length. In the first mode called /2, the antinodes of the stress wave T (i.e., maximum stress) is situated in the insulation layer. For the second mode called , a maximum stress point exists in the middles of the two piezoceramic parts; a node is situated in the insulation layer. Since the available energy is approximately proportional to the stress in the piezoelectric material, a maximum stress has to be induced in active parts for maximum efficiency [5]. In the case of the /2 mode the maximum energy is not used effectively. Moreover, in the second mode, the minimum stress is situated at the insulation layer. As a result, energy is produced efficiently. Consequently, we will choose operation in this mode. Fig. 3 shows the vibration displacement and stress distributions for the
275
thickness-extensional resonance mode.
vibration
in
second
The piezoelectric material used for the primary and secondary discs is lead titanate polarized along the thickness. This material has the advantage to exhibit a very high electromechanical coupling factor in the thickness resonance mode. The coupling is less significant in radial mode, which makes it possible to decrease the parasitic modes that may exist superposed to the main mode. The second advantage of this material is that the Curie temperature is 490 C compare to the PZT which is in the order of 328365 C. The insulation between the primary and the secondary is realized by a thickness layer of alumina, its dielectric rigidity is 35 Kv mm-1. The three layers are connected by epoxy adhesive.
The accepted equivalent circuit model for PT has been developed in [7]. This model is an electrical circuit, representative of properties exhibited by two layers of piezoceramic physically coupled together. Fig. 4 shows the simplified equivalent circuit model common to all piezoelectric transformers. The RLC series circuit represents the motional branch; it describes the mechanical oscillations of the material. The input capacitance C1and output capacitance C2describe the dielectric behavior of the piezoelectric layers of the transformer. The coupling between the electrical and mechanical branches is represented by the equivalent transformer (ratio : 1).
B. Transformer Design Equations
Owing to the equivalent circuit model described above, heating considerations and material properties, the PTs dimensions, and its electrical performances can be calculated. Table I contains the definitions of the various material coefficients. Equations (1)(6) [7] show the relationships between each material characteristic and geometrical dimension: input and output capacitances C1and C2 , the coupling ratio , the mechanical branch capacitance Cm and RLC series circuit, where C is the total equivalent capacitance
The last equation required in the design process is the heating equation. It defines the relationship between the transformer heating and the losses, is the PT efficiency
276
The gain V2/V1 expression can be easily calculated by
Where R is the PT resonance pulsation (i.e., giving the maximum gain) expressed as Table II contains the definitions of the variables used in (7). Transmitted power and efficiency are depending on the PTs load. In [5],[6] and [7], it is proposed to introduce variables representative of the quality of energy conversion. These new variables are shown in Table III. Q is the electric quality factor, it is inversely proportional to the load resistance RL, Qm is the mechanical quality factor, which characterizes the mechanical losses and c is a capacitance ratio, which characterizes the ratio between mechanical energy and electric energy that can be converted to secondary. It can be regarded as being the inverse of square of the secondary effective electromechanical coupling coefficient keff. Thus, the value of c ratio can never be lower than 1. Equations (8)(9) show the relationships between transmitted power, efficiency and these variables
3. DESIGN GUIDELINES
This part will depict the general design process of a two layers PT corresponding to our application. However, the designing concepts developed in this procedure can be extended to many others applications of piezoelectric transformers. Assumptions: The first assumption consists of exploiting the Mason model which does not take into account the insulating layer inserted between the two active layers. The second assumption consists of considering the piezoelectric material properties as perfectly known. However, this cannot be precisely done, because losses in the material are difficultly predictable. Consequently, the mechanical quality factor has to be estimated. Our previous studies showed that manufacturer values would have to be divided by a factor close to five in order to take
277
into account all extra mechanical losses (e.g., assembling, gluing, etc.). Given: Physical properties of the piezoelectric material. Transmitted power P2at the secondary P2 = PDEM+PMOS, is composed of the power required by the demodulation circuit (noted PDEM) and the power required by the MOSFET/IGBT grid PMOS =VGQG F (where VG is the supply voltage of grid, QG is the gate charge, and F is the switching frequency). PT operating frequency fR PT voltage gain. Temperature rise . To be evaluated: The geometrical dimensions: thickness e1 and e2, area A. The general design steps. 1) On the basis of the specifications of the operating frequency, We calculate the total PT thickness eTOTAL = e1+e2.
heating constraint according to the electric quality factor is plotted in Fig. 5. The two values of the electric quality factor Q corresponding to the optimal load operational points are obtained by resolution of
3) The determination of the area A and the thickness ratio consists to express power P2(8) and V2/V1 gain (10) according to A and , and to solve these two equations. In (8) and (10), we replace resonance pulsation R by the one contained in (11) and the quality factor Q by the solution of (13). In these new expressions of P2 and V2/V1, c is a function of . The dissipating area S and losses resistance Rare functions of the area A. This system has solutions only if the initial conditions are realizable, i.e., if the authorized heating is not too small for the desired power or if the gain is not too great. The thickness ratio value of primary and secondary layers must lie between 0.8 and 1.2. Actually, the transformer would not be able to work properly: a too thin layer could not put into vibration a thick layer. Equation (13) gives two solutions. The resolution of this problem must be done for the two values. Finally, the solution which gives the thicknesses ratio in the range defined previously (closest to 1) must be retained. 4. CONCLUSION In this paper, we have demonstrated the possibility of using piezoelectric transformers to realize an IGBT or MOSFET inverter-leg driver. First, we have highlighted the process to choose a PT structure well suited to gate driver applications. It was shown that the multilayered structure was preferred and that a suitable piezoelectric material is lead titanate. The PT must be supplied at its resonance frequency and must work at its loaded operating point. The choice
2) On the basis of losses considerations (7) and using the expression of efficiency (9), we calculate the PT load operation point with respect to the accepted temperature rise and to keep the minimal size. A graphical representation of the
278
of the PT second resonance mode has been carried out to minimize constraints in piezoelectric assembly layers.
REFERENCES: [1] S. J. Hui and S. C. Tang, Some electromagnetic aspects of coreless PCB transformers, IEEE Trans. Power Electron, vol.14, no. 4, pp. 805810, Jul.2000. [2] Optimal operation of coreless (PCB) transformerisolated gate drive circuits with wide switching frequency range, IEEE Trans. Power Electron., vol. 14, no. 3, pp. 506 514, May 1999. [3] D. Vasic, F. Costa, and E. Sarraute, A new MOSFET&IGBT gate drive insulated by a piezoelectric transformer, in Proc. IEEE PESC01 Conf., 2001, pp. 14791484. [4] G. Ivensky, I. Zafrany, and S. Ben-Yaakov, Generic operational characteristicsof piezoelectric transformers, in Proc. IEEE PESC00 Conf., 2000, pp. 16571662. [5], Generic operational characteristics of piezoelectric transformer,IEEE Trans. Power Electron., vol. 17, no.6, pp. 10491057, Nov. 2002. [6] Dejan Vasic, Franois Costa, Emmanuel Sarraute Piezoelectric Transformer for Integrated MOSFET and IGBT Gate Driver IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 21, NO. 1,Jan 2006 [7] T. Zaitsu, T. Inoue, M. Shoyama, T. Ninimiya, F. C. Lee, and G. C. Hua, Piezoelectric transformer operating in thickness extensional vibration and its application to switching converter, in Proc. IEEE PESC94 Conf., 1994, pp. 585589.
279
FPGA BASED DTC CONTROL OF INDUCTION MOTOR DRIVE

Mr. A. A. Kalage Govt. College of Engineering, Amravati. amolkalage@rediffmail.com Prof. V. M. Jape Govt. College of Engineering, Amravati. vasantmjape@yahoo.com Mrs. S. V. Tade SND College of Engineering & RC Babhulgaon, Yeola. Sarika.tade@rediffmail.com
Abstract Modern AC drives require a fast digital realization of many mathematical operations concerning control and estimators algorithms, which are time consuming. Therefore developing of custombuilt digital interfaces as well as digital data processing blocks and sometimes even integration of ADC converters into single integrated circuit is necessary. Due to the fact that developing an ASIC chip is expensive and laborious, the FPGA based solution should be used on the design stage of the algorithm. In this paper the application of FPGA in Direct Torque control induction motor drive is presented. KeywordsInduction motor, DTC, FPGA 1.INTRODUCTION The rapid development of very large scale integration (VLSI ) technology in recent years created an opportunity for the development of complex and compact high performance controllers efficient control of the induction motor drive which is a complex system having components of different nature: power converters built from linear and nonlinear electric elements, electrical machines and control system containing linear and nonlinear elements. Effective modeling and simulation of such systems require software tool that can handle all these functions in an integrated environment Now a days FPGA are replacing DSP in many applications due to some advantages In an FPGA multiple operations can be executed in parallel so that algorithm can run much faster which are required by control system. And if the algorithm is to be implemented on prototyping platform it is 280
desirable to have simulation software that can represent exactly the hardware used. The available simulation software for electronic circuits or dynamic circuits can be classified into two main categories: circuit simulation programs such as EMTP and equation solver program such as MATLAB/Simulink. These programs are not designed specifically for power electronic systems; therefore it is necessary to develop a model to fulfill the needs. Which can give good results but it requires considerable efforts. In this paper a method for modeling and simulation of Induction motor drives using MATLAB/Simulink is presented. As field programmable gate array (FPGA) is used to run the algorithm, a software Xilinx system generator, a toolbox of MATLAB/Simulink can be used. It will simulate the hardware as well as generate the VHDL code needed for the implementation in FPGA. It can automatically convert the model into VHDL code. In this paper the application of FPGA for the direct torque control of Induction motor drives is described and simulation results are presented. 2. DESCRIPTION OF POWER SYSTEM BLOCKSET Matlab Simulink is a systems simulator and unable to direct simulate electrical circuits so to simulate electrical circuits power system blocksets are used which is having libraries of electrical blocks and analysis tools which are used to convert electrical circuits into Simulink diagrams. The electrical blocks are electrical models such as electrical machines, current and voltage sources, and different electric elements, power electronic switches, connectors, and sensors for measurement purpose. When the simulation starts
Simulink use the Pm Blockset and transfers the electrical circuit into a statespace representation with the initial conditions of state variables. The actual simulation starts after this initial conversion, This allows the use of a wide variety of fixed step and variable step algorithms available in Simulink. As variable time step algorithms are faster than fixed time step method because the number of steps are less so these algorithms are used for small- and medium-size systems, And for large systems containing a more number of states and/or power switches, a fixed time step algorithm is used. A Simulink scopes can be used to display the Simulation results or these results can be sent to workspace during the simulation. The variety of MATLAB functions and toolboxes are present for processing and plotting of waveforms from stored data. 3.DESCRPTION OF XILINX SYSTEM GENERATOR Matlab/Simulink consists of Xilinx System Generator Blockset, which is developed by Xilinx. Due to development of Xilinx System Generator Blockset the use of high level programming tool for FPGA users become very easy. The Xilinx System Generator Blockset consists of a library of FPGA blocks, required for buildings of model. It also consists a VHDL code generator which automatically generates VHDL code from the created models. The Xilinx System Generator Blockset can work only with Xilinx FPGAs. For implementation of VHDL code in an FPGA an additional conversions are required. Initially the generated VHDL codes are transformed into a standard schematic circuit and then the schematic circuit is transformed by an implementation tool into a configuration file that can be downloaded to the FPGA. The main advantage of using the Xilinx System Generator Blockset to simulate the control algorithm is the possibility to generate code that can be used to program an FPGA directly from 281
the simulation model. Furthermore, the implemented code is guaranteed to perform exactly as in the simulation because the Xilinx System Generator Blockset represent exactly the FPGA hardware. This reduces the implementation time because the algorithm needs to be modeled and simulated only once. The Xilinx System Generator Blockset library contains many blocks that are classified as follows: 1. Basic elements: Contains simple signal processing functions (delay, multiplexers, registers, ..). 2. Math: Contains common mathematical functions (add, multiplication, relational, negate, . . . ). 3. MATLAB I/O: Contains blocks to connect the SG section to the Simulink model and determine the quantization error created by the use of fixedpoint values. 4. Memory: Contains memory elements found in FPGA (RAM, ROM, FIFO, . . . ) 5. DSP: Contains DSP like math functions (FIR, FFT, . .) Many of these blocks are replicas of standard Simulink blocks. So it is easy to build a model the same way than with standard Simulink blocks. To connect the SG algorithm to the remainder of the Simulink model, two specials blocks are used. Gateway In transforms an incoming floating point signal to a fixed-point one with a fixed time step. Gateway Out is used to return a value to the normal Simulink environment. These two blocks represent the boundaries of the SG model. Everything inside them can be implemented. A special block (Xilinx System Generator) is used to generate the implementation VHDL code from the model. The conversion is automatic. The only thing needed to be specified is the FPGA model that will be used. The generated code can be then used to incorporate the control algorithm in an FPGA.
4.MODELLING AND SIMULATION USING MATLAB/SIMULINK The formation of a Simulink model for the power circuit and the control algorithm requires few steps which can be separated in two main parts. 1. Design and testing of the system in MATLAB environment. 2. Generation and implementation of the VHDL code in the FPGA. The detailed step are shown in the flow diagram in Fig. 1.The first step is to determine the parameters of the power electronic section and use them to create a model using the PSB blocks. Also the hardware section of the prototyping board like the ADC can be simulated with a mathematical model. In this way, the control algorithm will be simulated as an exact replica of the real system. The next step is the building of the control algorithm with the use of only Xilinx System Generator blocks. It is then connected to the simulated power system by Gateway In and Gateway Out blocks. Fig. 2 shows the interconnections between the control and the power circuit. The next step is the simulation of the complete system in Simulink and the VHDL code generation which
Fig. 1. Induction Motor drive controller design and implementation process.
is done with the Xilinx System Generator block. The process is totally automatic except the choice of the FPGA model required. Normally, the FPGA contains some other functions in addition prototyping platform. A standard interface or a VHDL software can be used to join all these to the algorithm. Therefore the algorithm must be connected to the remainder of the VHDL code. This code is created only once and comes with the
282
codes. The last step is the synthesis and implementation of the VHDL code. This is done with Xilinx software which requires a few simple steps. After this, a configuration file is obtained permitting programming the FPGA. If some modifications of the system are required, only some steps need to be done again so the process will be much faster. The Matlab/Simulink model is shown in Fig. 3. 5. A DTC CONTROLLER FOR AN INDUCTION MOTOR DRIVE A Direct Torque control algorithm of Induction Motor Drive has been modeled and implemented on an FPGA-based prototyping platform. This algorithm has been chosen because it has wide range of application and offers a good level of complexity. The formation of algorithm model is very simple because almost all the blocks required
Fig. 2. Interconnection between power and control circuit.
Fig 3. Simulink diagram representing DTC Induction Motor Drive.
are present in the Xilinx System Generator library. A simulation of the original algorithm 283
side by side with the Xilinx System Generator is done to check and compare the performance
6.CONCLUSION A new approach to simulate complete Direct Torque control of Induction Motor drive systems using Matlab/Simulink (Power System Blockset, Xilinx System Generator) has been presented. This approach gives many advantages when the control algorithm is to be implemented for an FPGA-based system. As the modeling of the algorithm has to require doing only once it saves a lot of time also the simulation done using SG blocks represents exactly the hardware used to run the algorithm. This reduces the time to implement the algorithm because there are no modifications in the hardware. The implementation of the DTC algorithm has proved that the use of the SG in combination with the PSB is powerful and easy. Implementation time was reduced and the
control algorithm was working as expected on the hardware platform on the first try REFERENCES
1.Modeling and Simulation of Electrical Drives using MATLAB/Simulink and Power System Blockset By Hoang Le-Huy. IEcon01: The 27th Annual Conference of the IEEE Industrial Electronics Society. 2.FPGA Implementation of DTC Control Method for the Induction Motor Drive by Czeslaw T. Kowalski, Jacek Lis, Teresa OrlowskaKowalska, Wroclaw University EUROCON 2007 The International Conference on Computer as a Tool Warsaw, Sept. 9-12,2007 3.A Modified Direct Torque Control for Induction Motor Sensorless Drive by Cristian Lascu, Ion Boldea ,IEEE Transactions on industry applications, Vol. 36, No. 1 Jan/Feb 2000. 4.FPGA Design Methodology for Industrial Control SystemsA Review by Eric Monmasson and Marcian N. Cirstea ,IEEE Transactions on industrial Electronics, Vol. 54, August 2007.
284

INSTANTANEOUS ADAPTIVE FUZZY CONTROL BASED REACTIVE VOLT-AMPERE COMPENSATOR AND HARMONIC SUPRESSOR UNDER VARIABLE NON-LINEAR LOAD CONDITIONS
C.S.Perumalla * and P.C.Panda ** *PG Scholar, **Professor, Dept. of Electrical Engg., NIT Rourkela. Abstract-The simulation study of a fuzzy logic controlled, three-phase shunt active power filter to improve power quality by compensating harmonics and reactive power required by a nonlinear load is presented. The compensation process is based on sensing line currents only, an approach different from conventional methods. The performance of the fuzzy logic controller is compared with a conventional PI controller. The dynamic behavior of the fuzzy controller is found to be better than the conventional PI controller. PWM pattern generation, to control the switchings of the converter, is based on carrierless hysteresis based current control. Various simulation results are presented under steady state and transient conditions. Index Terms Voltage Source Converter, Active Power Filter (APF), Fuzzy Logic controller (FLC), Hysteresis controller, Non linear Load, Power Quality. List of Symbols : : Instantaneous and peak vs(t), Vm value of source voltage is(t), if(t), il(t) : Instantaneous source, filter and load currents pl(t), pf(t), pr(t), ph(t) : Instantaneous load, fundamental, reactive and harmonic currents
1. INTRODUCTION In recent years, the applications of power electronics have grown tremendously. These power electronic system offer highly nonlinear characteristics. An increase in such nonlinearity causes various undesirable features such as increased harmonics and reactive power components of current from AC mains, low system efficiency and a poor power factor, disturbance to other consumers and interference in nearby communication networks. i.e., the rise of issue of Power quality (PQ). In order to overcome these problems, the active power filters have been researched and developed. In recent years, shunt active power filters based on current controlled PWM converters have been widely investigated and recognized as a viable solution [1], [2]. The percentage processing of electrical energy by Power electronics is increasing due mainly to the fast growth of its capability. A race is currently taking place between increasing Power electronic population and sensitivity. Power electronics has three faces as, one that introduces valuable industrial and domestic equipment; a second one that creates problems; and, finally, a third one that helps to solve those problems. However, most of active power filters are based on sensing harmonics [3], [4] and reactive voltampere requirements of the nonlinear load [5], [6]
285
and require complex control. New scheme proposed using a simple synthetic sinusoid generation technique by sensing the load current [4]. This scheme is further modified by sensing line currents only [7], [8] which is simple and easy to implement. The conventional PI controller was used for the generation of a reference current template. The PI controller requires precise linear mathematical models, which are difficult to obtain and fails to perform satisfactorily under parameter variations, nonlinearity, load disturbance, etc. Recently, fuzzy logic controllers (FLCs), which are based on linguistic description, have generated a good deal of interest especially in control applications [9] as these FLCs can overcome the above mentioned setbacks of PI controllers. This paper presents a fuzzy logic controlled shunt active power filter for the harmonics and reactive power compensation of a nonlinear variable load. The control scheme is based on sensing line currents only; an approach different from conventional ones, which are based on sensing harmonics and reactive volt-ampere requirements of the nonlinear load. The three-phase currents/voltages are detected using only two current/voltage sensors. The performance of fuzzy controlled active power filter has been compared with the conventional PI controller under constant load conditions. 2. COMPENSATION PRINCIPLE The active power filter is controlled to draw/supply the a compensating current if from/to the load to cancel out the current harmonics on AC side and reactive power flow from/to the source there by making the source current in phase with source voltage. Fig.1 shows the basic compensation principle of the active power filter.
Source current Is Equivalent of Source load current Il If Non linear load Filter current
is the required compensating current is determined

Fig.1. Basic Compensation Principle
A. Significance of DC capacitor The DC side capacitor serves two main purposes: 1. It maintains a DC voltage with small ripple in steady state, 2. It serves as an energy storage element to supply the real power difference between load and source during the transient period. When the load condition changes the real power balance between the mains and the load will be disturbed. This real power difference is to be compensated by the DC capacitor. This changes the DC capacitor voltage away from the reference voltage. In order to keep satisfactory operation or the active filter, the peak value of the reference source current must be adjusted to proportionally change the real power drawn from the source. This real power charged/discharged by the capacitor compensates the real power consumed by the load. If the DC capacitor voltage is recovered and attains the reference voltage, the real power supplied by the source is supposed to be equal to that consumed by the load again [10]. In this fashion the peak value or the reference source current can be obtained by regulating the average voltage of the DC capacitor. A smaller DC capacitor voltage than the reference voltage means that the real power supplied by the source is not enough to supply the load demand. Therefore, the source current (i.e. the real power drawn from the source) needs to be increased, while a larger DC capacitor voltage than the reference voltage tries to decrease the reference source current. B. Estimation of reference source current The peak value of the reference current can be estimated by controlling the DC side capacitor voltage. Ideal compensation requires the mains current to be sinusoidal and in phase with the source voltage, irrespective of the load current nature. The desired source currents, after compensation, can be given as
i sa * = I sp sin t isb * = I sp sin ( t 120 )
Shunt active power filter
286

isc * = I sp sin ( t + 120 )
Where I sp ( = I1 cos 1 + I sl ) is the amplitude of the desired source current, while the phase angle can be obtained from the source voltages. From Fig.1 the instantaneous currents can be written as
(1)
From (6), the source current supplied by the source, after compensation, is
is ( t ) = p f ( t ) / v s ( t ) = I 1 cos 1 sin t = I sm sin t

Where
i s ( t ) = il ( t ) i f ( t )
. (2)
Source voltage is given by v s (t ) = V m s in w t
Ism = I 1 cos1
.. (3)
If a nonlinear load is applied, then the load current will have a fundamental component and harmonic components. So the load current can be represented as
il (t ) = In sin(nt + n)
There are also some switching losses in the PWM converter, and capacitor leakage losses, hence the utility must supply a small overhead for the capacitor leakage and converter switching losses in addition to the real power of the load. The total peak current supplied by the source, is
I sp = I sm + I sl
. (8)
il ( t ) = I 1 sin( t + 1)
. (4) The instantaneous load power can be given as
n=1
+ I n sin( n t + n )
n=2
If the active filter provides the total reactive and harmonic power, then is (t) will be in phase with the utility voltage and purely sinusoidal. At this time, the active filter must provide the following compensation current:
i f (t ) = il (t ) is (t )
p l ( t ) = v s ( t ) * il ( t )
pl (t) = vmI1sin2 t *cos1 +vmI1sint *cost *sin1 +vm sint *In sin(nt +n)
n=2
..
(5)
(6) From (5), the real (fundamental) power drawn by the load is given by
p l (t ) = p f (t ) + p r (t ) + p h (t ) .
p f ( t ) = v m I 1 sin 2 t * cos 1 = v s (t ) * is (t )
.. (7)
Hence, for accurate and instantaneous compensation of reactive and harmonic power it is necessary to estimate is (t), i.e. the fundamental component of the load current the reference current. Hence, the waveform and phases of the source currents are known (from (1)), and only the magnitudes of the source currents need to be determined. This peak value of the reference current has been estimated by regulating the DC side capacitor voltage of the PWM converter [10], [11]. This capacitor voltage is compared with a reference value and the error is processed in a fuzzy controller. The output of the fuzzy controller has been considered as the amplitude of the desired source current, and the reference currents are estimated by multiplying this peak value with the unit sine vectors in phase with the source voltages. 3. MODELLING OF THE SYSYTEM
287
The complete active filter system is composed mainly of 3. A voltage source PWM converter, and 4. A fuzzy controller. All these components are modeled separately, integrated and then solved to simulate the system. A. Modelling of Nonlinear Load A three-phase diode rectifier with R-L load is considered as a nonlinear load. Due to the presence of source inductance, six overlapping and six lionoverlapping conduction intervals occur in a cycle. During a non-overlapping interval only two devices will conduct, while during an overlapping interval three devices of the bridge will conduct simultaneously. The dynamic equations during nonoverlap and overlap intervals are given in (8) and (9), respectively:
1. A three-phase source, 2. A nonlinear load, in terms of the DC bus capacitor voltage vdc switching functions stating the on/off status of the as devices of each leg TA, TB, and TC
fa
v dc (2T 3
TC)
v fb = v fc =
v dc ( T A + 2T B T C ) 3 v dc ( T A T B + 2T C ) 3
p i d = (V 0 ( 2 R s + R L ) i d 2 v d ) /( 2 L s + L )
.. (9)
..... (11) The three phase currents ica, icb,icc flowing through the filter impedance (Rc , Lc) are obtained by solving the following differential equations[12,14,15]: 1 p i fa = R f i fa + ( v sa v fa ) L f 1 p i fb = R f i fb + v sb v fb Lf
))
pi d = (V 0 (1.5 R s + R L ) i d 2 v d ) /(1.5 L s + L )
1 pi fc = R f i fc + ( v sc v fc ) L f
..... (10) Where, Rs and Ls are the elements of the source impedance, vd is the drop across each device, RL, and L are the elements of load impedance, id is the load current flowing through the diode pairs and p is the differential operator (d/dt). Vo is the AC side line voltage segment vac, vbc, vba, vcb, vab ,vcc during non-overlap intervals, and vbc+vac/2,vba+vbc/2, vca+vba/2, vcb+vca/2, vab+vcb/2 and vac+vab/2 during overlap intervals, based on diode pair conduction. The phase currents isa, isb, isc are obtained by id considering the respective diode pair combination [13, 14]. B. Modelling of PWM Converter The PWM converter has been modelled as having a three phase AC voltage applied through filter impedance (Rf, Lf) on its input, and a DC bus capacitor on its output. The three phase voltages vfa., vfc, vfc reflected on the input side can be expressed
. (12) The DC side capacitor current can be obtained in terms of phase currents ifa, ifb,ifc and the switching status (1 for on and 0 for off of the devices TA, TB, and TC ).
idc = i faTA + i fbTB + i fcTC

1 pVdc = ( i faTA + i fbTB + i fcTC ) cdc
.. (14)
(13)
From this, the model equation of the DC side capacitor voltage can be written as
C. Estimation of Reference Current Templates The peak value of the reference current Imax is estimated using the fuzzy controller by controlling
288

the DC side capacitor voltage in the closed loop. The output of the fuzzy control algorithm is the reference current Imax (n) at the nth sampling time is determined by adding the previous reference current Imax (n-1) to the calculated change in reference current:
change in reference current Imax (n).The peak at the nth sampling instant are used as inputs for the fuzzy processing. The control scheme is shown in fig2.The output of the fuzzy controller after a limit is considered as the amplitude of the reference current Imax. This current Imax takes care of the active power demand of load and the losses in the system The switching signals for the PWM converter are obtained by comparing the actual source currents (isa, isb, and isc) with the reference current templates (isa*, isb*, and isc*) in the hysteresis current controller. Switching signals so obtained, after proper amplification and isolation, are given to switching devices of the PWM converter. A. Basic fuzzy algorithm In a fuzzy logic controller, the control action is determined from the evaluation of a set of simple linguistic rules. The development of the rules requires a thorough understanding of the process to be controlled, but it does not require
imax ( n) = Imax ( n 1) + Imax ( n)
..(15) In classical control theory this is an integrating effect, which increases the system type and improves the steady state error. 4. PROPOSED FUZZY CONTROL SCHEME In order to implement the control algorithm of a shunt active power filter in closed loop, the DC side capacitor voltage is sensed and then compared with a reference value. The obtained error e = V dc, ref - V dc, act Change of error signal and ce(n) = e(n)-e(n- 1)
289
Fig.2. Schematic Diagram of Closed Loop Fuzzy Logic Controlled Shunt APF
e(n)
Imax (n)
Fuzzification
Vdc Vdcref ce (n)
Defuzzification
imax( n) =Imax( n1 )
+ Imax( n)
FuzzyInference Engine
Fuzzy Logic Controller To analogue multiplier through DAC
Fig.3. Internal Structure of Fuzzy controller
a mathematical model of the system. The internal structure of the fuzzy controller is shown in Fig. 3. Here, the error e and change of error ce are used numerical variables from the real system. To convert these numerical variables into linguistic variables, seven fuzzy levels or sets are chosen [8]. The fuzzy controller is characterized as follows: (i) Seven fizzy sets for each input and output [8]. (ii) Triangular membership functions for simplicity. (iii) Fuzzification using continuous universe of discourse. (iv) Implication using Mamdani's 'min' operator. (v) Defuzzification using the 'height' method.
Fig.4. Input, Output Normalized Membership Functions
B. Rule base: The elements of this table are determined based on the theory that in the transient state, large errors need coarse control, which requires coarse input/output variables; in the steady state, small errors need fine control, which requires fine input/output variables. Based on this, the elements of the rule table are obtained from an understanding of the filter behavior
Table .1.Control Rule Base
ce NB e NB NM NB ZE PS PM PB NB NB NB NB NM NS ZE NM NB NB NB NM NS ZE PS NS NB NB NM NS ZE PS PM ZE NB NM NS ZE PS PM PB PS NM NS ZE PS PM PB PB PM NS ZE PS PM PB PB PB PB ZE PS PM PB PB PB PB
The performance of active power filter has been analysed by solving by solving the above set of differential equations using fourth order Runga Kutta method [16]. 6. SIMULATIONS AND RESULTS Fig.5 shows the response of PI controller. The filter is assumed to be switched on at 50 ms.The
290

figures a,b,c,d and e represents load current, dc capacitor voltage, source current, filter current and source voltage respectively.The responses with fuzzy controller are shown in fig(6).
40 30
S reu nn osc r ti A u r e
20 10 0 -10 -20 -30 -40 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seconds
Fig.5(c)
20 15
ft r u nn i e r ti A l cr e
10 5 0 -5 -10 -15 -20 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seconds
7. CONCLUSION A fuzzy logic controlled shunt active power filter has been studied to improve the power quality by compensating harmonics and reactive power requirement of the nonlinear load. The performance of a fuzzy logic controlled shunt active power filter has been studied and compared with the conventional P1 controller. The fuzzy controller has a better transient response compared to a conventional PI controller, and the steady state performance of the fuzzy controller is comparable to the PI controller.
APPENDIX: Source voltage Vs=100 V (peak), system freq =50 Hz, system impendence, (Rs,Ls)=0.1;0.15mH,Filter impedance R , L = 0.1;0.66mH, Load impedance (R ,L)= 6.7,20mH, 220V
25 20 15
Fig.5 (d)
100
Sre la i V ocvt g n u o e
50
-50
-100 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seconds
Fig.5 (e) Fig.5 Performance of APF with PI controller

30
L c en o un A a r t dr i
20 10 0 -10 -20 -30 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seco n d
Fig.6 (a)
23 0 22 0
Da i r o eV C a Vg C t po l c t i an
21 0
20 0
19 0
18 0
17 0
16 0 0
0.0 2
0 .04
0.06
0.0 8
0 .1
0.1 2
0.14
Time in Seco n d
Fig6 (b)
40 30
S ern A o Ct u u i r c r n e
20 10 0 -1 0 -2 0 -3 0 -4 0 0
0.0 2
0 .04
T ime in Seco n d
0.06
0.0 8
0 .1
0.1 2
0.14
dcref
fe rn A i ru t l cen t r i
DC link capacitance2000F, Reference DC link voltage(V
Fig.6(c)
20 10 0 -10 -20 0
0.02
0.04
Time
0.06
in
second
0.08
0.1
0.12
0.14
La cre ti A o u r n d r n
10 5 0 -5 -10 -15 -20 -25 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Fig.6 (d)
Sc o e V o el g u va i r t n
100 50 0 -50 -100 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seconds
Fig. 5 (a)
230
Time
in
second
Dc at r o gi V C p io la n a c vt e
220 210 200 190 180 170 160 0
Fig.6 (e) Fig.6. Performance of APF with FLC

0.02 0.04 0.06 0.08 0.1 0.12 0.14
Time in seco nd
Fig.5 (b)
1.
2.
3.
8. REFERENCES W. M. Grady, M. J. Samotyj, and A. H. Noyola, Survey of active power line conditioning methodologies, IEEE Transactions on Power Delivery, vol. 5, no. 3, Jul. 1990, pp.15361542 B. Singh, A. Chandra, and K. Al-Haddad, A review of active filters for power quality improvement, IEEE Transactions on Industrial Electronics, vol.46, no 5, Oct 1999, pp1-12. L.A.Morgan, J.W.Dixon & R.R.Wallace, A three phase active power filter operating with fixed switching frequency for reactive power and current harmonics compensation, IEEE Transactions on Industrial Electronics, vol.42, no.4, August 1995, pp 402-408.
4.
5.
6.
7.
R. M. Duke and S. D. Round, The steady state performance of a controlled current active filter, IEEE Transactions on Power Electronics, vol. 8, Apr. 1993, pp. 140146. E.H.Watanbe, R.M.Stephan & M.Aredes, New concepts of instantaneous active and reactive powers in electrical systems with generic loads, IEEE Transactions on Power Delivery, vol.8, no.2, April 1993, pp.697-703. V.Soares, P.Verdelho & G.D. Marques, An instantaneous active and reactive current component method of active filter, IEEE Transactions on Power Electronics, vol.15, no.4, July 2000. B. Singh, A. Chandra, and K. Al-Haddad, Computeraided modeling and simulation of active power filters, Electrical Machines and Power Systems, vol. 27, 1999, pp. 12271241.
291

8.
9.
10.
11.
12.
13.
14. 15. 16. 17. 18.
K. Chatterjee, B. G. Fernandes, and G. K. Dubey, An instantaneous reactive volt-ampere compensator and harmonic suppressor system, IEEE Transactions on Power Electronics, vol. 14, no. 2, Mar. 1999, pp. 381 392. V. S. C. Raviraj and P. C. Sen, Comparative study of proportional-integral, sliding mode, and fuzzy logic controllers for power converters, IEEE Transactions on Industrial Applications, vol. 33, no. 2, Mar./Apr. 1997, pp. 518524. Shailendra Kumar Jain, Pramod Agarwal and H.O. Gupta, Simulation and Experimental Investigations on a shunt Active Power Filter for Harmonics and Reactive Power Compensation, IETE Technical Review, Vol. 20, NO. 6, November December 2003, pp 481- 492. S. K. Jain, P. Agrawal, and H. O. Gupta, Fuzzy logic controlled shunt active power filter for power quality improvement, Proceedings of Institute of Electrical Engineers, Electrical Power Applications, vol. 149, no. 5, 2002. Narain g. Hingorani, Laszlo Gyugyi, Understanding FACTS, IEEE Press, Standard Publishers Distributors, 2001. Mohan, N., Undeland,.T.M, and Robbins,.W.P, Power electronics :converters, applications and design, Singapore,John Wiley and sons, 2003. M.H.Rashid, Modern Power Electronics, Printice Hall- India, 2002. P.S. Bimbhra , Power electronics, Dhanpat rai Publications, 2005. Pipes and Harvill, Applied Mathematics for Engineers and Physians, B. K. Bose, Modern Power Electronics and AC Drives, Singapore, Pearson Education,2004. D.Pradeep Kumar, Investigations on shunt active power filter for power quality improvement PG thesis, NIT, Rourkela, 2007.
292
A FUZZY LOGIC BASED CONTROLLER FOR AN INDIRECT VECTOR CONTROLLED THREE-PHASE INDUCTION MOTOR
A.Amarendra Prof.J.K.Satapathy Department of Electrical Engineering, NIT Rourkela. indirect vector based control of induction motor Abstract - This paper presents the theory, has been investigated. The proposed fuzzy logic design and simulation of a fuzzy logic based controller (FLC) has been successfully simulated controller used for an indirect vector on a simulink model with the help of fuzzy logic controlled three-phase induction motor. The toolbox. The performance of the FLC has been analysis, design and simulation of the fuzzy successfully compared with the conventional PI logic controller for IVCIM drive system are controller. It is found that the proposed FLC is carried out based on the fuzzy set theory. The insensitive to load variation and sudden changes FLC algorithm has been simulated on in the speed command. Simulink in Matlab. The performance of the 2. INDIRECT VECTOR CONTROL OF proposed FLC has been investigated and INDUCTION MOTOR compared to the results obtained from the The indirect vector control method is conventional PI controller based drive at essentially the same as the direct vector control, different operating conditions. The simulation results demonstrate that the performance of except that the rotor angle e is generated in an the FLC is better than that for the indirect manner (estimation) using the measured conventional PI controller. speed r and the slip speed sl. To implement the indirect vector control strategy, it is necessary Keywords -Induction motor drives, Speed to take dynamic equation into consideration and control, Fuzzy Logic controllers, Matlab / the following equations (1-5); Simulink software. 1. INTRODUCTION Induction motors have a simple and rugged structure; moreover, they are economical and immune to heavy overloads. However the use of induction motors also has its disadvantages, mainly the controllability, due to its complex mathematical model and its nonlinear behavior [1]. The vector control or field oriented control (FOC) theory is the base of a special control method for induction motor drives. With this theory induction motors can be controlled like a separately excited dc motor. This method enables the control of field and torque of the induction machine independently (decoupling) by manipulating the corresponding field oriented quantities [1, 2]. In this paper, the configuration and design of the fuzzy logic controller for
e = e dt = ( + sl )dt = r + sl (1)
r
For decoupling control, the stator flux component of current ids should be aligned on the de axis, and the torque component of current iqs should be on the qe axis, that leads to qr = 0 and dr =r then: (2) As well, the slip frequency can be calculated as: (3)
It is found that the ideal decoupling can be achieved if the above slip angular speed command is used for making field-orientation. For a
293
dr = 0 Substituting dt in equation (2) yields the rotor flux set as r = Lm ids (4) The motor developed torque is directly related to
constant rotor flux r and
Where ce denotes the change of e, r (ts) is the actual rotor speed,
r* (t s ) is the reference speed and e (ts 1) is the value of error at a previous sampling time. The output variable of the FLC is the change in torque, T, which is integrated (in discrete sense) to get the reference torque T*(ts) as shown in the equation:
Te*(ts) = T*(ts 1) + T* B. Fuzzification (8)
(5) 3. FUZZY LOGIC SPEED CONTROLLER PRINCIPLE AND DESIGN

Basically, the fuzzy logic controller consists of four blocks as shown in Figure 1, Fuzzification, fuzzy inference engine, Knowledge base and a defuzzification block.
Fig. 1: Complete Fuzzy Logic Controller Used for Vector Induction Motor Control
In this stage the crisp variables of the inputs e(ts) and ce(ts) are converted into fuzzy variables that can be identified by the levels of membership in the fuzzy set. Each fuzzy variable is a member of the subsets with a degree of membership varying between 0 (nonmember) to 1 (full member). The fuzzy sets are defined as Z=Zero, PS=Positive Small, PM=Positive Medium, PB=Positive Big, NS=Negative Small, NM=Negative Medium, NB=Negative Big PVS=Positive Very Small, NVS=Negative V. Small, NVB= Negative very Big, and PVB= Positive very Big. The universe of discourse of all the variables, covering the whole region, is expressed in per unit values. All the MFs have asymmetrical shape with more crowding near the origin (steady state). This permits higher precision at steady state [1-3].
C. Knowledge base and inference stage
A. Input/output variables The design starts with assigning the mapped variables inputs/output of the FLC in Figure 2. The I/O must be clearly defined. In this case the first input variable is the speed error e and the second is the change in speed error ce=e* at a sampling time ts. The two input variables e(ts) and ce(ts) are calculated at every sampling time as: (6) e(t s ) = r* (t s ) r (t s ) ce(ts) = e(ts) e(ts 1) (7)
Knowledge base involves defining the rules represented as IF-THEN rules statements governing the relationship between inputs and output variables in terms of membership functions. In this stage the input variables e(ts) and ce(ts) are processed by the inference engine that executes 7x7 rules represented in rule table. Inferencing stage includes also, application of fuzzy operator (AND, OR), implication and aggregation. An example of rule statements describing the expert system is: IF e(pu) (speed error in per unit) = Z (zero) AND
294
ce(pu) = PS (positive small) THEN du(pu) = PS (positive)

D. Defuzzification stage
This stage introduces different inference methods that can be used to produce the fuzzy set value for the output fuzzy variable T.
Fig. 2: Indirect Vector Control Based Induction Motor Block Diagram with Fuzzy Logic Control
Fig. 3: FIS system (FLC) layout, memberships functions
Fig.4: Fuzzy Speed Controller in vector-controlled drive System
In this paper, the center of gravity (COA) or centroids method is used to calculate the final fuzzy value T*(ts). Defuzzification using COA method means that the crisp output of T*(ts) is obtained by using the center of gravity, in which the crisp du(pu)0 or T(ts) variable is taken to be
295
the geometric center of the output fuzzy variable value out( T) area, where out( T) is formed by taking the union of all the contributions of rules with the degree of fulfillment greater than 0. Then the COA expression with a discretized universe of discourse can be written as follows:
Then by integration Te * is obtained as shown by equation (8). This Torque value is used to calculate iqs *, which in turn used to command the induction motor via 2 -3 block.
The overall system is shown in Figure 2 (the speci -fication parameters of the motor are given in the appendix) and the constructed Fuzzy Logic Controller in MATLAB/SIMULINK environment [4] is shown in Figure 3. The block diagram of the drive representing the IVCIM with Fuzzy Logic Control is shown in Figure 4. 4. SIMULATION RESULTS Various simulation tests were carried out on both the PI controller and the FL controller on the indirect vector control of induction motor (IVCIM). Time response and steady state error were compared. Figures 5 and 6 show the PI and FLC response speed at no load. FLC performed better with respect to rise time and steady state errors. Figure 7 shows the speed track performance test,when a sudden change in speed reference is applied in the form of a look-up table. Figures 8 and 9 examine the load disturbance rejection capabilities of each controller when using a load torque step from 0 to 200 N.m applied at 0.6 seconds. The FL controller at that moment returns quickly to the command speed within (0.1sec) with a maximum drop in speed of 0.7rad/sec. whereas the PI controller was affected by the change on load torque and sustained a steady state error.
Fig.5 Speed Response comparison at no-load
Fig 6 Enlarged response comparison
296
It is found that the FLC is more robust and follow the difficult ramp without delay. It was also found that the FLC did not show significant changes in its response due to load variation, whereas the PI controller was sensitive to changes in load conditions. 5. CONCLUSION An indirect vector controlled induction motor drive system has been introduced. The drive system was simulated with both a fuzzy logic controller and a conventional PI, and their performances were compared. Simulation results showed that the fuzzy logic controller is more robust during load changes and eliminates the transients during sudden changes in speed. Overall simulation results showed that the fuzzy controller has higher performance than the PI controller. Appendix: Motor specification Machine Type: 3-phase Induction Motor Rotor Type: Squirrel Cage Stator and Rotor: Y-connection to an internal neutral point Reference Frame: Stationary 50 hp, 1500rpm, (120 rad/sec), 460V 50Hz 4poles Rs = 0.087 Rr = 0.228 Ls= 0.8mH Lr = 0.8mH Lm = 34.7mH Jn = 1.662Kg.m2 f = 0.1N.m.s
REFERENCES [1] B. K. Bose. Modern Power Electronics and AC Drives, Prentice-Hill PTR Companies, Inc. Upper Saddle River, NJ 07458, 2002. [2] Hoang Le-Huy, Minh Ta-Cao and J.L. Silva Fuzzy Logic Based Controller for Induction Motor Drives Canadian Conference on Electrical and Computer Engineering, 1996., Volume: 2 , 2629 May 1996 [3] M.N. Uddin, T. S. Radwan and M. A. Rahman Performance of Novel Fuzzy Logic Based Indirect Vector Control for Induction Motor Drive Proceedings of IEEE 2000 ref, 0-78036401-5. [4] http://www.mathworks.com (The official site for MATLAB &SIMULINK as well the Fuzzy Logic.)
Fig 7 Speed tracking response comparison
Fig 8 Speed Response Comparison during Sudden Load Change
Fig 9 Enlarged results of Fig. 8.
297
FUZZY CONTROL OF QUASI RESONANT BUCK CONVERTER

A.Rameshkumar 1, Dr.S.Arumugam2
1 2
Research scholar, Govt. College of Technology, Coimbatore-13, India, arkr1966@yahoo.com. Chief Executive Officer, Nandha Engineering College, Pichandapalayam, Erode 638 052, arumugamdote@yahoo.co.in. control of AC and DC drives, feedback control of converters, non linearity compensation, on and off line diagnostics, etc., due to its capability of fast computation with high precision and rule based algorithm. Therefore, it is a paradigm for the alternative design methodology which naturally provides the ability to deal with the highly non linear, timevariant and ill defined systems where the mathematical models are difficult to be obtained or control variables are too hard to measure or where human reasoning, perception or decision making are inextricably involved [3]. Design of fuzzy logic or rule based non-linear controller is easier than other advanced control methods in that its control function is described by using fuzzy sets and if then rules rather than cumbersome mathematical equations or large look up tables; it will greatly reduce the development cost and time and needs less data storage in the form of membership functions and rules in order to simplifies the complexity of design. It can also exhibit increased reliability, robustness in the face of changing circuit parameters, saturation effects, and external disturbances and so on [3]. Therefore, the focus here is strictly on the feasibility of implementing a Fuzzy logic based controller to establish its superior performance on Quasi-Resonant Buck Converter over the conventional Pulse Width Modulated (PWM) converter at various operating points of the converter. Simulation results for two categories of tests namely line and load regulation are depicted and peak overshoot and settling time are used to measure the performance of the system. II. GENERALIZED STATE-SPACE AVERAGING TECHNIQUE Consider a periodically switched network with k different switched modes in each switching cycle, described by the state equation X (t) = Ai x (t) + Bi (t), i = 1, 2 k (1.0) The equation (1.0) for the conventional method is characterized by the Generalized State Space Averaging equation [1] as
Abstract A controller for a power-electronic converter based on Fuzzy logic is implemented. Being free of complex equations and heavy computation, it achieves fast dynamic response and adapts to varying conditions of operation. The topology is modelled using MATLAB software and operated at finite higher switching frequency to evaluate its closed-loop performance in respect of line and load regulation. It is verified by transient characteristics that due to quasi-resonance there is a drastic change in peak overshoot and settling time and the proposed strategy has good rejection ability for supply and load disturbances. Keywords: Quasi Resonant Buck Converter and Fuzzy Logic. I. INTRODUCTION Power converter is the need of the day, may it be Process control automation, telecommunication, energy conservation, medical instrumentation, utility related or aerospace applications and it is imperative to design such converter circuits capable of operating at higher frequencies to achieve high power density [2]. The converter must be made increasingly reliable and efficient and the output voltage regulation of such converter against load and supply voltage fluctuations is an important criterion in designing high density power converters. Crucial to the performance of power converters is the choice of control methods. Traditional frequency domain analog methods predominantly used because of its simplicity and lowimplementation cost are based on an equivalent linear small signal model of the converter and this model has restricted validity especially for systems with strong non-linearity [9]. It also cannot meet the more stringent requirement of todays digital circuit because of inherent disadvantages such as low flexibility, low reliability, temperature drift of the components, susceptibility to electromagnetic interference and very sensitive to the environmental influence such as thermal and aging [3]. The complexity of the system, the practical converter operation because of problems associated with parasitic resistance, stray capacitance and leakage inductance of the components, and increasingly demanding closed loop system performance necessitates the use of more sophisticated controllers. In particular, research has been directed at applying non linear control principles to the regulation and dynamic control of output voltage of the converter [7]. With the aid of advanced microcomputer technology, digital control of power converter becomes feasible but such methods involve a lot of complex equations and calculations. If the control method is based on an artificial intelligence instead of solving equations arithmetically, the required processing time of the controller can be reduced and the sluggish response of the system shall be improved. Among the various techniques of artificial intelligence, the most popular and widely used method in control systems is the Fuzzy logic [10]. It was first introduced in the early 1970s in an attempt to design the controller for systems that are structurally difficult to model due to naturally existing non-linearities and other modeling complexities. Fuzzy logic is a departure from classical Boolean logic as it relies on human capability to understand systems behaviour and is based on qualitative control rules. It is one of the intelligent schemes that convert the linguistic control strategy based on expert knowledge into an automatic strategy or otherwise it implements non-numeric linguistic variables on a continuous range of truth values which allows intermediate values to be defined between conventional binary system [13], [4]. In addition, it has emerged as one of the most active and promising control methods in the power electronic systems such as speed
x = {
i =1
di Ai }x + 1/T

i =1
ti
ti 1
Bi()d
(1.1)
T is the switching period, fs = 1/T is the switching frequency and fo = the highest natural frequency of state matrix Ai. If the input control variable functions namely Bi are bounded and fs is much greater than fo, then, the equation (1.1) can be obtained. III. SMALL SIGNAL MODEL OF THE CONVERTER The Half-wave mode Zero Current Switched Quasi Resonant step down converter in Fig. 1 operated in discontinuous conduction mode use only uni-directional switch and hence it is not able to return excessive tank energy to the source. It offers many distinct advantages [6] such as self commutation, low switching stress and loss, high efficiency and power density, reduced electromagnetic interference and noise, and faster transient response to load and line variations.
Fig. 1: ZCS QRC BUCK CONVERTER The first step in the design of controller for such a system is to obtain its control-to-output transfer function [8]. The
298

transfer function of such converter under the worst case condition namely the minimum line and maximum load condition is the base in the design of the controller [10]. The transfer function is obtained as in equation 2.0 by analyzing the circuit in its four modes of operation using the Generalized State Space Averaging Technique and it has overcome the limitations of the conventional method [1]. (a) Inductor charging mode
Fig. 2: INDUCTOR CHARGING MODE (b) Resonant mode
2. To partition the universe of discourse of the interval spanned by each variable into a number of fuzzy subsets, assigning each a linguistic label (subsets include all the elements in the universe). 3. To assign or to determine membership function for each fuzzy subset. 4. To assign the fuzzy relational between the input fuzzy subsets on the one hand and the output fuzzy subsets on the other hand, thus forming the rule base. 5. To choose appropriate scaling factors for the input and output variables in order to normalize the variables to [-1, 1] interval. 6. To fuzzify the inputs or to classify the input data into suitable linguistic values or sets to the controller. 7. To use fuzzy appropriate reasoning to infer the output contributed from each rule. 8. To aggregate the fuzzy outputs recommend 9. To apply defuzzification technique to form a crisp output. 10. To send the change of control action to control the plant. (a) Identification of inputs and output Error e and change in error voltage ce (k) are the two inputs to the Fuzzy Controller and change in duty cycle is the resulting output. The error is computed by subtracting the actual output voltage Vo from the reference voltage Vg and the derivative input is calculated by subtracting the previous error from the current error as ce(k)=e(k)- e(k-1) at the kth sampling instant. The output of the fuzzy control algorithm is the change of duty cycle [d(k)]. The duty cycle d(k), at the kth sampling time, is determined by adding the previous duty cycle [d(k-1)] to the calculated change in duty cycle as d(k)=d(k-1)+ d(k) [7]. Depending on the magnitude of error and change in error, the switching frequency is varied for regulating the output voltage. (b) Membership functions Three Gaussian membership functions are chosen to model, analyze and simulate the Fuzzy Controller. It has been defined taking into account the conditions of normality and convexity of fuzzy sets; it embodies the mathematical representation of membership in a set and is required to have uniform shapes, parameters and functions for the sake of computational efficiency, efficient use of the computer memory and performance analysis. It also gives the degree of confidence about the result. The membership functions for the input and output are shown in Fig 7, 8 and 9.
Fig. 3: RESONANT MODE (c) Capacitor discharging mode
Fig. 4: CAPACITOR DISCHARGING MODE (d) Free wheeling mode
Fig. 5: FREE WHEELING MODE
(2.0) where Fs = switching frequency, Vo = output voltage, Lo = Filter inductor, Co = Filter (output) capacitor, R = Load resistor. IV. FUZZY CONTROLLER FOR QUASI RESONANT BUCK CONVERTER The general structure of a fuzzy logic controller [5] is represented in Figure.6 and the steps involved in design are:
Fig. 7: MEMBERSHIP FUNCTION FOR ERROR SIGNAL
Fig. 8: MEMBERSHIP FUNCTION FOR CHANGE IN ERROR SIGNAL
Fig. 6: BASIC CONFIGURATION OF FUZZY LOGIC CONTROLLER 1. To select control elements and parameters as scaling factors for input and output signals.
Fig 9: MEMBERSHIP FUNCTION FOR CONTROL SIGNAL
299

(c) Fuzzification Fuzzy sets contain objects that satisfy imprecise properties of membership. It provides a mathematical way to represent vagueness in humanistic systems and is defined for each input and output variables. For ease of computation, seven fuzzy subsets [7] are defined by the library of fuzzy set values for the error and change in error and they are NB (Negative Big), NM (Negative Medium), NS (Negative Small), ZE (Zero), PS (Positive Small), PM (Positive Medium), and PB (Positive Big). (d) Development of rule base Normally, the fuzzy rules are heuristic in nature; they are typically written as antecedent consequent pairs of IF THEN structure and the inputs are combined by AND operator. The antecedent and consequent are the description of process state and control output respectively in terms of a logical combination of fuzzy propositions. 49 rules as shown in Table 1 are formed depending on the number of membership functions to play a key role in the improvement of system performance [9]. 1. If the output of the converter is far from the set point, the change of the duty cycle must be large so as to bring the output to the output to the set point quickly. 2. If the output is approaching the set point, a small change of the duty cycle is necessary. 3. If the output is near the set point and is approaching it rapidly, the duty cycle must be kept constant so as to prevent overshoot. 4. If the set point is reached and the output is still changing, the duty cycle must be changed slightly to prevent the output from moving away. 5. If the set point is reached and the output is steady, the duty cycle remains unchanged. 6. If the output is above the set point, the sign of change of duty cycle must be negative and vice-versa. V. DESIGN DATA In order to compare the performance of Fuzzy controller with PWM converter and Quasi-Resonant Buck Converter, aforesaid controller as in Fig. 11 with the specifications given in Table 2 is considered with the design constraints as mentioned hereunder [1]. 1. The rating of the filter components must be much higher than the rating of the resonating components. 2. The switching frequency must be higher than the natural frequency of the low pass filter at the output and hence the state variables of the filter state can be regarded as constant in each cycle.
Fig. 11: SINGLE INPUT SINGLE OUTPUT FUZZY CONTROLLER FOR QRC BUCK CONVERTER No. 1 2 3 4 5 6 7 8 9 10 11 12 13 PARAMETER Input Voltage Output Voltage Resonant Inductor Resonant Capacitor Filter Inductor Filter Capacitor Load Resistance Characteristic Impedance Switching Frequency Natural Frequency Resonant Frequency Load Current Peak Resonant Current Output Power(max) SYMBOL Vg Vo Lr Cr Lo Co R Zn fs fo fr Io IM PO VALUE 100 120 V 54 V 1.6 H 0.064 F 0.2 mH 20 F 10 100 5 200 kHZ 2.5165 kHz 0.5 MHz 0.54- 5.4 A 20 A 2.916 kW
Fig. 10: RULE BASE IN TERMS OF SURFACE VIEW
14
Table 2: DESIGN PARAMETERS VI. SIMULATION The duty cycle can be varied from -1 to 0 for the half -wave configuration. To compare the performance of fuzzy controller with Quasi Resonant Buck Converter and with PWM Converter, five different operating points spanning the entire operating range of the converter have been selected [10]. 1. Minimum line and maximum load condition 2. Minimum line and light load condition 3. Mid range line and load condition 4. Maximum line and maximum load condition 5. Maximum line and light load condition Table 1: RULES FOR CONTROL SIGNAL The results of digital simulation for duty cycle equal to 0.0 and -0.2 for various conditions of supply and load variations are shown hereunder; duty cycle = 0.0 depicts the result of digital simulation of Fuzzy control of PWM converter and that of -0.2 depicts Quasi-Resonant Buck Converter. It is shown that Fuzzy control of Quasi-Resonant Buck Converter gives better performance of voltage regulation. The output voltage of this converter at 54 V for varying conditions of operation is illustrated in Table 3.
(e) De-fuzzification Conservation of the fuzzy to crisp or non-fuzzy output is defined as De-fuzzification. Mean of Maxima (MOM) method is implemented, where only the highest membership function component in the output is considered.
300

Duty cycle = 0.0 Case: 1 Minimum Line and Maximum load condition
90 80 90 80
Case: 4 Maximum Line and Maximum load condition
O tp t V lta e(V u u o g )
70 60 50 40 30 20 10 0 0 0.002 0.004 0.006 0.008 0.01
O tp t V lta e(V u u o g ) Time (sec)
70 60 50 40 30 20 10 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Time (sec)
9 8 9 8
O tp t C rre t (A u u u n )
7 6 5 4 3 2 1 0 0 0.002 0.004 0.006 0.008 0.01
O tp t C rre t (A u u u n ) Time (sec)
7 6 5 4 3 2 1 0 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Time (sec)
Case: 2 Minimum Line and Light load condition

120 100 80 60 40 20
Case: 5 Maximum Line and Light Load condition

120 100 O tp t V lta e(V u u o g ) 80 60 40 20
0 0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
Time (sec)
0 0
0.005
0.01
0.015 0.02 Time (sec)
0.025
0.03
0.035
1.4 1.2 1.4 1.2 O tp t C rre t (A u u u n ) 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0 0 0.005 0.01 0.015 0.02 Time (sec) 0.025 0.03 0.035 1 0.8 0.6 0.4 0.2
1 0.8 0.6 0.4 0.2 0 0
Time (sec)
Case: 3 Midrange Line and Load condition Duty cycle = -0.2

120 100 O tp t V lta e(V u u o g ) 80 60 60 40 20 0 0
Case: 1 Minimum Line and Maximum load condition

70
50 40 30 20 10 0 0
0.005
0.01
0.015 Time (sec)
0.02
0.025
0.03
0.005
0.01
0.015
Time (sec)
3 2.5 O tp t C rre t (A u u u n ) 2 1.5 1 0.5 0 0 60 50 40 30 20 10 0 0
0.005
0.01
0.015 0.02 Time (sec)
0.025
0.03
0.035
0.01
0.02
0.03
0.04
0.05
Time (sec)
301

Case: 2 Minimum Line and Light load condition Case: 5 Maximum Line and Light Load condition
60 50 O tp t V lta e(V u u o g ) 40 30 20 10 30 20 10 0 0 0 0 0.005 0.01 0.015 Time (sec) 0.02 0.025
60 50 40
0.7 0.6
Time (sec)
0.01
0.02
0.03
0.04
0.05
0.5 0.4 0.3 0.2 0.1 0 0
0.7 0.6
0.5 0.4 0.3 0.2 0.1 0 0
0.005
0.01
0.015
0.02
0.025
Time (sec)
CASE 1
0.01 0.02 0.03 0.04 0.05
INPUT VOLTAGE 100 100 115 120 120
LOAD RESISTOR 10 100 40 10 100
LOAD CURRENT 5.4 0.54 1.35 5.4 0.54
OUTPUT VOLTAGE 54 54 54 54 54
Time (sec)
2 3
Case: 3 Midrange Line and Load condition
4 5
60 50 40 30 20 10 0 0
Table 3: OUTPUT VOLTAGE OF THE QUASIRESONANT BUCK CONVERTER The value of peak overshoot and settling time for duty cycle equal to 0.0 and 0.2 for various conditions of supply and load variations is mentioned in chronological order of case in Table 4.
MAXIMUM OVERSHOOT
0.01 0.02 0.03 0.04 0.05
SETTLING TIME (ms) PWM CONVERTER 2.4094 22.309 6.9026 1.9889 22.157 QR BUCK CONVERTER 0.79962 1.6168 0.75107 0.77355 1.6249
Time (sec)
PWM CONVERTER
QR BUCK CONVERTER 66.221 54.003 53.998 66.2635 53.9996
1.4 1.2
86.122 105.84 101.41 85.521 103.64
1 0.8 0.6 0.4 0.2 0 0
0.01
0.02
0.03
0.04
0.05
Time (sec)
Table 4: TIME DOMAIN SPECIFICATIONS OF THE CONVERTER Case: 4 Maximum Line and Maximum load condition VII. CONCLUSION A novel Fuzzy control scheme is implemented to improve the dynamic performance of the Quasi-Resonant Buck Converter. The Converter is simulated by MATLAB Software to show its feasibility and its transient characteristics are compared with the conventional PWM converter. Fuzzy Controller is able to regulate the output voltage to a desired value effectively for each operating point without steady state oscillations despite variasions in load or input voltage and gives an improved performance compared to PWM Converter. As compared to standard controllers it has low electromagnetic interference as there are no very sharp edges of current and voltage waveforms, improved performance in maximum peak overshoot and settling time to parameter variations and better stability. The results obtained by simulation confirm the validity of the technique used. VIII. REFERENCES [1] Jianping Xu and C.Q.Lee, A Unified Averaging Technique for the Modelling of Quasi-Resonant Converters, IEEE TRANSACTIONS ON POWER ELECTRONICS, Vol.13, No.03, May 1998, pp 556 563.
70 60
50 40 30 20 10 0 0
0.005
0.01
0.015
0.02
0.025
Time (sec)
70 60
50 40 30 20 10 0 0
0.005
0.01
0.015
0.02
0.025
Time (sec)
301

[2] Kwang-Hwa Liu, R.Oruganti, and F.C.Y.Lee, Quasi Resonant Converters-Topologies and Converters,IEEE TRANSACTIONS ON POWER ELECTRONICS, Vol.2, No.01, January 1987, pp 62-71. [3] Y.F.Liu and P.C.Sen, Digital Control of Switching Power Converters, Proceedings of the 2005 IEEE Conference on Control Applications, Toronto, Canada, August 28-31, 2005 , pp 635-640. [4] Timothy J.Ross, Fuzzy Logic with Engineering Applications, Second Edition, John Wiley and Sons, Inc. Singapore 129809. [5] P.Mattavelli, L.Rossetto, G.Spiazzi, P.Tenti, General Purpose Fuzzy Controller for DC-DC Converter, IEEE TRANSACTIONS ON POWER ELECTRONICS, Vol.12, No.01, 1997, pp 79 86. [6] D.W.Hart, Introduction to Power Electronics, Prentice Hall International, Inc. [7] Tarun Gupta, R.R.Boudreaux, R.M.Nelms and John Y. Hung, Implementation of a Fuzzy controller for DC- DC Converters using an inexpensive 8 Bit Microcontroller, IEEE TRANSACTIONS ON POWER ELECTRONICS Vol.44,No.05, October 1987, pp 661-669. [8] I. Batarseh and K. Siri, Generalised Approach to the Small Signal Modelling of DC- DC Resonant Converters, IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS Vol.29,No.03, July 1993, pp 894-909. [9] W.C. So, C.K.Tse, and Y.S.Lee, Development of Fuzzy Logic Controller for DC-DC Converters: Design, Computer simulation and Experimental evaluation, IEEE TRANSACTIONS ON POWER ELECTRONICS, Vol:11, No.:01, January 1996, pp. 2432. [10] K. Viswanathan, D. Srinivasan, R. Oruganti, A Universal Fuzzy controllerfor a Non-Linear Power electronic Convereter, Fuzzy Systems: IEEE02 Conf. Rec., 2002, pp 46-51. [11] V.Vorperian and S. Cuk, Small Signal Analysis of Resonant Converters, Proceedings of the IEEE Power Electronics Specialists Conference, June 6-9, 1983, pp 269-282. [12] C. K. Tse and K. M. Adams, Quasilinear Modelling and Control of DC-DC Converters, IEEE TRANSACTIONS ON POWER ELECTRONICS, Vol:7, No.:03, 1992, pp. 315-323. [13] B. K. Bose, Expert System, Fuzzy Logic and Neural network Applications in Power Electronics and Motion Control, Proceedings of the IEEE, Vol: 82, No.8, 1994, pp 1303-1323. [14] F. Ueno, T. Inoue, I. Oota and M. Sasaki, Regulation of CUK converters using Fuzzy Controllers, INTELEC91, Nov 1991, pp 261-267. [15] L. K. Wong, F. H. F. Leung, P. K. S. Tam and K. W. Chang, Design of an Analog Fuzzy Logic Controller for a PWM Boost Converter, IECON97, Vol: 1, 1997, pp 360-363. [16] S. H. Kuh and G. T. Park, An Adaptive Fuzzy Controller for Power Converters, Fuzzy Systems-IEEE, 1999, Conf. Proc., Vol: 1, 1999, pp 434-439. [17] D.Driankov, Hans Hellendoom and Michael Reinfrank, An Introduction to Fuzzy Control, Narosa Publishing House, Second edition, Delhi 1997. [18] J.G.Kassakian, M.Schledt and G.C.Verghese, Principles of Power Electronics, Addison-Wesley,1991. [19] E.J.Miller, Resonant Switching Power Conversion in IEEE PESC Record, 1976, pp 106-122.
302
POWER QUALITY: A RESEARCH REVIEW PROBLEMS AND SOLUTIONS

Suresh.Y* ,B. Chitti Babu* *Dept. of Electrical Engg., National institute of technology Rourkela. Abstract-The dominant application of electronics today is to process information. The computer industry is the biggest user of semiconductor devices, and consumer electronics. Due to the successful development of semiconductors, electronic system and controls have gained wide acceptance in power, information and computing technology and due to the continuous use of drive systems (rotating machines, controlling thyristors and associated electronic components) in industry and in power stations, and the need to keep such systems running reliably, electronic equipment are becoming an integral part of today's industrial, institutional, and commercial facilities. Unfortunately, the same type of equipment often generates power supply disturbances, which in turns affect other items of equipment, and are more likely to generate the distorting harmonics. These harmonics can cause power to be used inefficiently and can be a source of premature equipment failure that will halt production in industrial processing, will result in loss of life in hospitals, data processing activities in real time such as banking transaction processing may be lost, etc The principal aim of this paper is to investigate the most common power quality problems, the effect of the harmonics on the power quality, the ways of evaluating the amount of harmonic distortion present in a power system which lead to isolate the cause of the problem and finally device a solution for a good power quality 1. INTRODUCTION Understanding power quality can be confusing at best. There have been numerous articles and books concerning power quality [1, 2]. There are two terms known in power systems about the quality of power: Good power quality and poor power quality. Good power quality can be used to describe a power supply that is always available, always within the voltage and frequency tolerances and has a pure noise-free sinusoidal wave shape to all equipment, because most equipment was designed on that basis [3]. Unfortunately, most of the equipment that is manufactured also distorts the voltage [4] on the distribution system, leading to what is known as poor power quality. And thus affecting other equipment that was designed with the expectation of consistent undistorted voltage, and are thus sensitive [5, 6] to power disturbances resulting in reduced performance and will cause equipment mis -operation or premature failure. The cost of power quality problems can be very high and include the cost of downtimes, loss of customer confidence and, in some cases, equipment damage. Indeed, power quality is an important point in the relationship between suppliers and consumers [7] but might become a contractual obligation that stress on improving voltage quality, availability, performance [8] and efficiency and these improvements will have: Benefits for industrial customers (customized and flexible availability) and for suppliers- utilities. 2. CAUSES OF POWER QUALITY PROBLEMS In today's fast-paced environments, a huge amount of money is spent on sate of the art computer controlled- equipment and systems. These systems are often installed in "unfriendly" electrical environments, which cause owners, industrial companies and investors a great deal of frustration and disappointment and in many cases, result in a great loss of time and money, and that lead us to ask a valuable question" What is the problem?" The
303
Proceedings of the 1st APE-2009 on Jan. 24-25, 2009.
answer to this question could have one or more of the following points: _ Computer malfunctions _ Interrupted manufacturing sequences _ catastrophic failures _ Erratic equipment behavior _ High electrical maintenance cost. But whatever the answer is, Chapman [9] has classified them as a power quality problems and the latter is subdivided into two categories: Supply system quality problems Installation and load related problems 2.1 SUPPLY SYSTEM QUALITY PROBLEMS Supply interruption Transient interruption Transients Under voltage/ over voltage Voltage dip/ voltage surge Voltage imbalance Flicker Harmonic distortion According to Douglas et all [10] the above problems can be classified into one of three disturbance categories based upon duration: Transient disturbances include unipolar Transients, oscillatory transients (such as Capacitor switching), localized faults and other events typically lasting less than ten milliseconds. Transients can originate internally within the building or externally on utility power lines. They represent about 12 to 15% of all power line problems. These Disturbances can cause: Damage to electronic lighting systems Shutdown to sensitive equipment Immediate or latent damage to digital Microprocessor controlled equipment. Momentary disturbances are voltages increases Or decreases (sags, swells, and interruptions) Lasting more than 10 milliseconds but less than three seconds. The majority of voltage sags result on utility lines from faults on the Distribution or transmission lines and they Represent about 60% of all power problems. Voltage swells are the least frequent of the Power line problems representing
about 2 to 3% of all power problems occurring to industry studies [10]. These momentary disturbances can cause: Sudden decrease in line loads loosing wiring Re-energizing of power after a utility power Interruption, when power comes back in. Steady-state disturbances are voltage increases or decreases (under voltages, overvoltages, and interruptions). Interruptions and power outages can originate from electrical short circuits in Building wiring or on utility power lines. These interruptions will cause electrical, computer and electronic equipments shut down and losses in operations and revenues. 2.2 .INSTALLATION AND LOAD RELATED PROBLEMS The major problems in this category can be classified [11] in one of the three following groups: Harmonic currents Earth (Ground) leakage currents Voltage dips and transients 2.2.1 HARMONIC CURRENTS These currents will cause wiring, motors and Transformers to overheat. The result may be a breakdown of insulation and a significant reduction of equipment lifespan. Some of the adverse effects of concentrated nonlinear loads upon a facility are - Nuisance tripping of circuit breakers - Overvoltage problems - Metering problems - Overheating of transformers and induction motors - Computer malfunctions - Metering problems - High levels of neutral-to-ground voltage - Power factor rate penalties All non-linear loads generate harmonics. This includes all loads, which use switching to control or convert power, for example: - Switched mode power supplies- computers, office equipment, domestic equipment - Variable speed motor drives - Thyristor controlled heating elements - Dimmer switches
304
- Solid state fluorescent la mp ballasts - Over loading magnetic devices such as motors, lamp ballast and transformers as a result of saturation of the magnetic core material.
2.2.2 EARTH LEAKAGE CURRENTS The principal design consideration for an earthling system is that it must protect people and animals from receiving potentially fatal electric shocks in the event of a fault condition. Now, earth conductors are carrying large leakage current permanently as well as serving as a sink for high frequency noise currents. If for any reason the connection to earth is poor, then the impedance of the primary earth route will be high and earth Leakage currents will seek alternative routes to earth. This may result in current flowing in unexpected places with consequent risk should the system be disconnected. 2.2.3 VOLTAGE DIPS, FLAT TOPPING AND TRANSIENTS Heavy loads such as air conditioning systems, large motors during the starting process, principally cause Dips. However, flat topping is caused by electronic equipment such as the start -up of printers. The combination of surges and dips in the voltage lead to what is known as voltage Flicker and this latter is caused by the operation of large cyclic loads and can reduce the life of motors drives and electrical contacts. 3. OBJECTIVE OF A POWER QUALITY SURVEY The power quality survey is the first, and perhaps most important, step in identifying and solving power problems cited previously. In other words it is thus designed to locate, identify and eliminate the electrical disturbances which disrupt data collection networks, PLCs, variable speed motor drives thyristor controlled heating elements and other sensitive electronic equipment that contain some form of microchip or "logic circuits". There are two types of power quality surveys:
_ The first type is a preventative survey, uses a Number of tests and inspections to locate potential Problems before they cause a production outage. _ The second type is, a troubleshooting survey, it is used to locate and eliminate problems as quickly as possible after a production outage. Whether the investigation involves a simple piece of equipment or the facility's entire electrical system, the survey process typically requires the following steps [10] Planning and preparing the survey Inspecting the site Monitoring the power Analyzing the monitoring and inspection data Applying corrective solutions. From the above steps, the survey should provide the background information and basic methodology and tools required to benchmark the power quality performance and improve the reliability with respect to interruptions. Thus, the process basically involves finding out the What, Where, When, How and Why of the power related Problems at hand. _ Monitor requirements [12, 13]: what are the requirements or what are the specific equipment resources needed, to get the job done. _ Where to monitor: depends on where the Problems are observed or suspected. _ When to monitor: The time when the problem occurs can also provide important clues about the nature of the power problem. If the problem occurs at a certain time of day, the equipment switched on at that time should be suspect. The monitoring period should last at least as long as one business Cycle," which is how long it takes for the process in the facility to repeat itself. _ how long to monitor _ Data collection and analysis systems _ Indices for describing performance _ Results of other benchmarking efforts from around the world _ Other benefits of the bench marking effort.
305
Proceedings of the 1st APE-2009 on Jan. 24-25, 2009.
As part of the planning and preparation process it is necessary to obtain a site history for the facility of equipment being investigated. Asking questions of equipment operators or others familiar with operations is an important part of a site history. 3.1 INSPECTING THE SITE The site examination begins by visually inspecting outside the facility and around the immediate vicinity in order to gains a better perspective of the utility service area. Inspecting the facility helps to identify equipment that might cause interference. It will also surface electrical distribution system problems such as broken or corroded conduits, hot or noisy transformers, poorly fitted electrical panel covers and more. Any inspection should include a physical review of the wiring from the critical load to the electrical service entrance and any loads, which might cause power problems, will be identified 3.2 MONITORING THE POWER To solve a power problem for a single unit of equipment, the monitor should be placed as close to the load as possible. Looking for a power quality problem, need that voltage signal is monitored and finally to find the cause, accurate measurement of power quality issues requires that the monitor accurately [12] measure the voltage and current waveforms.
3.3 ANALYZING THE MONITORING AND INSPECTION DATA
The potential solutions for such problems are dependent on the type of disturbance. However, for the voltage disturbances, such as momentary outage, sags and swells and transient voltages, the most convenient solutions to improve the performance of a sensitive equipment is to install a protective device between the power source and sensitive equipment. There is a wide variety of protective devices [14, 15], each device has a different problem solving function and can be used in a variety of applications. Voltage regulators are installed between the power source and sensitive equipment to control the incoming voltage [16] in order to sustain a constant output voltage, it protects the equipment against overvoltages and under voltages. However, Transient Voltage Surge Suppressors (TVSS) cuts noise and voltage transients only, and it does not regulate voltage to limit surges and sags. 4.2 SOLUTION OF HARMONIC PROBLEMS With current technology, virtually all of today's high performance electronic equipment uses static power rectifiers, which convert alternating current to direct current, and the reverse. It pulls a nonlinear current and the latter induces voltage distortion and when distorted voltage is delivered to equipment designed to expect a sinusoidal voltage, the result is overheating or malfunctions. Harmonic currents are a fact of life and cannot be eliminated unless nonlinear loads are avoided, all industry is increasingly exposed. Although future developments may provide improved electronic systems producing lower levels of harmonics, the problem must be addressed in current and future installations. Since harmonic currents cannot be prevented, installations must be designed to cope with them. Utility companies impose limits on the Harmonic voltage distortion, which a customer's site can impose on the system. Where the utility limits are exceeded, special additional steps must be taken to filter the harmonic content. Active harmonic filters [17] are now becoming available which inject an exactly complimentary harmonic current into the supply to cancel that produced by
306
To identify equipment problems, it is key to analyze data in a systematic manner. First, look for power events that occurred during intervals of equipment malfunction. Second, identify power events that exceed performance parameters for the affected equipment. Third, review power monitor data to identify unusual or severe events. Finally, correlate problems found during the Physical inspection with equipment symptoms. 4.SOLUTIONS 4.1 SOLUTIONS TO SUPPLY SYSTEM QUALITY PROBLEMS
the nonlinear loads. Harmonic filters will provide a solution, although all ramifications of their use may not be anticipated in advance. Filters types include line-reactors, passive harmonic filters, active harmonic filters [18], electronic feedback filters and special transformers that use out -of phase windings to accomplish harmonic reduction [19]. While these units are effective at reducing the harmonic current as seen by the utility, they do not reduce the harmonic current flowing in the cables of the installation. According to Martin [11] practicing the following measures minimizes the effect of harmonics: Take account of harmonic generating loads when planning the installation. Reduce the number of socket outlets on each circuit and increase the number of circuits. Carefully distribute these circuits among the phases to reduce out of balance currents. Increase the cross sectional area (CSA) of feeders. 'Two sizes up" is a good rule of thumb with a sound technical justification- and it saves energy too! Increase the CSA of neutral feeders and distribution panels - twice the phase CSA is recommended. The old practice of using halfsized neutrals is definitely no longer satisfactory! Uses 5 core copper cable- one core for each phase and two for the neutral. Keep circuits, which supply harmonic generators- such as office equipment and variable speed drives separate from those supplying harmonic hating Loads- such as induction motors Maintain records of cable layout and usage. Maintain and upgrade the system carefully. Consult the relevant records before making changes. Routinely monitor neutral and phase currents. Check for excessive heating in transformers, motors and distribution boards. 4.3 SOLUTION OF EARTH LEAKAGE PROBLEMS The primary purpose of grounding electrical systems is to protect personnel and property if a fault (short circuit) were to occur. The second purpose of a grounding system is to provide a controlled, low impedance path for lightning
induced currents to flow to the earth harmlessly. Problems can be avoided if it is recognized that the grounding system in electrical installations is no longer designed solely for fault conditions. Wherever modern electronic equipment is used, the earth acts as a key-working component of the electrical installation. As such, it must be given equal consideration in terms of: _ Impedances _ Connections _ Documentation _ Working practices 5.CONCLUSIONS Semiconductors are the heart of computer industry; unfortunately these electronic components are nonlinear and thus may affect the safe or reliable operation of computers and computer-based equipment. Often more important than the physical effect on the equipment is the loss of productivity resulting from computer equipment failure, mis calculations and downtime. Most of surveys done by different researchers show a high incidence of problems caused by harmonics while problems caused by earth leakage were less common. Although filtering is a solution of reducing the effect on such harmonics on the quality of power, it solves harmonic voltage problems at the frequency to which they are tuned only. And thus changes in the equipment on site will change the harmonic Profile, so rendering the filters ineffective. Thus due to the technology and software now available, monitoring is highly-effective means to detect, solve, and even prevent problems on both utility and customer, it can detect problem conditions throughout the system before they cause equipment malfunctions, and even equipment damage or failure. However, before monitoring a design step is needed and should take into account the electrical environment and define how the electrical installation must meet the needs of the business. 6. REFERENCES
[1] Roger, C. Dugan, Mark F. McGranaghan and H. Wayne Beaty " Electrical power systems quality". Prentice Hall, Feb.2000.
307
Proceedings of the 1st APE-2009 on Jan. 24-25, 2009. [2] Gregory J. Porter and Andy Van Sciver " Power quality solutions: case studies for troubleshooters". The Fairmont Press, Prentice Hall, 1st edition Nov 24, 1998. [3] " Different types of power quality conditioning. http://www.powerquality.com/art0022.art1.htm [4] "Why equipment is a source of power quality problems". http://www.powerquality.com/pqpark/pkpk104.htm [5] " EC and M Practical Guide to quality power for sensitive electronic equipment". DeDad, John and Waggoner, Ray, Intertec Publishing, Overland Park, KS, 1993. [6] " Details of equipment sensitivity "http://www.powerquality.com/pqpark/pqpk1052.ht m [7] N.G. Hingorani, " Introducing custom power " IEEE Spectrum, Jun. 1995, 41-48. [8] D.D. Sabin and A. Sundaram, "Quality enhances reliability". IEEE Spectrum, Feb. 1996. 34-41. [9] David Chapman, " Electrical design - A good practice guide ", CDA Publication 123, Dec.1997. [10] Douglas S. Dorr, Thomas M. Gruzs and James J. Stanislawski. " Interpreting recent power quality surveys to define the electrical environment". http://www.powerquality.com/art0021/art1.ht m [11] Marty Martin "Common power quality problems and best practice solutions " Shangri-la Kuala Lumpur, Malaysia 14. August 1997. [12] Dennis Stewart, " Cover Story: Power monitoring technologyDispelling-metering myths" http://www.powerquality.com/articles.html [13] " Comprehensive monitoring - covering all aspects" http://www.powerquality.com/art0031/art1.htm [14] J.M. Powell, "Power conditioning system and apparatus" U.S. Patent 4, 544, 877, Oct.1, 1985. [15] M.B. Brennen and B. Banerjee, " Low cost, high performance active power line conditioners". Proc. Conf. PQA' 94, Part2, Amsterdam, The Netherlands, Oct. 2427,1994. [16] W. K. Chang, W.M. Grady and M.J. Samotyj" Controlling harmonic voltage and voltage distortion in a power system with multiple active power line conditioners". IEEE 95, WM 257- 6PWRD. [17] Janko Nastran, Rafael Cajhen, Matija Seliger, and Peter Jereb " Active power filter for nonlinear ac loads". IEEE Trans. PE, Vol.9, No.1, Jan. 1994. [18] " Active filter design and specification for control of harmonics in industrial and commercial facilities". http://www.powerquality.com/art0020/art4.htm [19] " Design solution for harmonic load current effects on electrical power distribution equipment ". http://www.powerquality.com/art0028/art1.htm [20] " Wiring and Grounding for power quality ", Electric Power Research Institute, Palo Alto, CA [21] " The how's and why's of isolated grounding ". http://www.powerquality.com/art0043/art1.html
308

Session 4

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Session 4

Загружено:

Авторское право:

Доступные форматы

National Institute of Technology, Rourkela 769008, Orissa.

SiC Power Semiconductor Devices- An impact on Next Generation Power Components

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig. 1. Basic structure and equivalent circuit of Schottky power diodes.

National Institute of Technology, Rourkela 769008, Orissa.

Table I: Physical Properties of Various Semiconductors for Power Devices

Material Si Ga As 3C-SiC 6H-SiC 4H-SiC GaN GaP C

Eg (eV) 1.12 1.4 2.3 2.9 3.2 3.39 2.26 5.6

(cm /V.s) 450 400 45 90 115 350 150 1800

Ec (V/cm) 3 x105 4 x10

(W/cm.K) 1.3 0.54 5 5 5 1.3 1.1 20

2 x106 2.5 x106 3 x106 5x106 107 5.6x105

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig.4 Cross-section of DMOSFET power transistor shows its resistive components

National Institute of Technology, Rourkela 769008, Orissa.

National Institute of Technology, Rourkela 769008, Orissa

AC LINK INVERTER BASED ON MULTIPLE CARRIER PWM FOR FUEL CELL

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

dc/dc converter type

Fig.1 High frequency link dc/ac converter

Carri 0/180 deg Phase shifter 1

0/180 deg Phase shifter 3

Fig. 3 Multiple carrier pulse width modulator circuit

National Institute of Technology, Rourkela 769008, Orissa

TABLE 1. COMBINATIONS CONDITIONS AND RESULTING PWM SEQUENCES

Phase Shifter 2 0 0 0 0 180 0

Phase Shifter 3 180 0 180 0 0 180

Combining method Add Subtract Add Subtract Add Add

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig. 4 (d) three carrier PWM sequence

dc Input Fuel Cell/ Battery

Carrier Modulating Signal

Multiple Carrier PWM Sequence Generation

Selection of Positive/ Negative Converter

Current Polarity Detector

Fig. 5 Block Diagram for circuit operation

National Institute of Technology, Rourkela 769008, Orissa

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

National Institute of Technology, Rourkela 769008, Orissa.

ENHANCING ENERGY EFFICIENCY IN PHOTOVOLTAIC SYSTEM USING POWER CONDITIONING UNIT

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig-1 Block diagram of a battery charging system.

Fig-2 Block diagram of solar system using an inverter.

National Institute of Technology, Rourkela 769008, Orissa.

Fig-3 Ideal switch V, Duty ratio D, and switching period

Fig-4 DC-DC Boost converter

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig-6 Differential Amplifier

G = (R2 / R1) -----------------------

National Institute of Technology, Rourkela 769008, Orissa.

Fig 8. I-V characteristics of solar cell under illumination and darkness.

Fig-7 The equivalent circuit of a solar cell.

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

Fig.9 I-V and power curves of a typical solar cell.

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

DIRECT ACTION FUZZY PID CONTROLLER STRUCTURE

National Institute of Technology, Rourkela 769008, Orissa.

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

National Institute of Technology, Rourkela 769008, Orissa.

Table2 Rule base for fuzzy logic controller

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

fig-3 Improved response obtained from FLC by superposing consequent parts.

Proceedings of the 1st APE-2009 on Jan. 24-25, 2009

th53=(1/5)acos(real(cos(5th2)+0.5-cos(5th1))); th72=(1/7)acos(real(cos(7th1)+cos(7th3)-0.5)); th11=(1/1)acos(real(cos(1th2)+0.5-cos(1*th3)-