Вы находитесь на странице: 1из 8

International Journal of Trend in Scientific Research and Development (IJTSRD)

International Open Access Journal | www.ijtsrd.com

ISSN No: 2456 - 6470 | Volume - 2 | Issue – 6 | Sep – Oct 2018

Identifying the Slider-Crank


Crank Mechanism System by the MPSO Method
Chin Wen Chuang
Chuang, Kuan Yu Chen, Yi Hua Su
Department of Electrical Engineering
Engineering, I-Shou University
Kaohsiung, Taiwan

ABSTRACT
This paper mainly proposes an efficient modified However, the dynamic formulations of a slider-crank
slider
particle swarm optimization (MPSO) method, to mechanism with one degree-of of –freedom have more
identify a slider-crank
crank mechanism driven by a ffield- than one independent variable in the past researches
resear
oriented PM synchronous motor, dynamic [3-4].
4]. In this study, the dynamic formulation is
formulations of a slider-crank
crank mechanism have been expressed by only one independent variable of
successfully formulated with only one independent rotation angle. Moreover, its dynamic responses are
variable. The parameters of many industrial machines compared well with the experimental results.
are difficult to obtain if these machines cannot be
taken apart. In system identification, we adopt the Particle swarm optimization (PSO) is a stochastic
MPSO method to find parameters of the slider slider-crank population
on based optimization approach, and was first
mechanism. This new algorithm is added with published by Kennedy and Eberhart in 1995 [5-6]. [5
“distance” term in the traditional PSO’s fitness PSO has been shown to be an efficient, roust and
function to avoid converging to a local optimum. It is simple optimization algorithm. Most studies of the
found that the MPSO method can obtain optimal high high- PSO are empirical with only a few theoretical
quality solution, high calculation efficiency, and its analyses that concentrate
oncentrate on understanding particle
feasibility and effectiveness. Finally, the comparisons trajectories. A large amount of research has been
of numerical simulations and experimental results proposed to improve the performance of PSO. From
prove that the MPSO identification method for the these studies, much effort has been invested to obtain
slider-crank mechanism is feasible. a better understanding of its convergence properties.
These studies es concentrated mostly on a better
Keyword: Slider-crank
crank mechanism, Modified particle understanding of the basic PSO control parameters,
searm optimization. namely the acceleration coefficients, inertia weight,
velocity clamping, and swarm size [7-9]. [7 In this
I. INTRODUCTION paper, a modified PSO algorithm is proposed to
A slider-crank
crank mechanism is widely used in gasoline improve the searching g ability and prevent from being
and diesel engines, and has been studied extensively trapped in a local optimum. The main difference of
in the past three decades. The responses of the system the MPSO from the PSO is its fitness function
found by Viscomi and Ayre [1] are to be dependent considers the “distance” to avoid converging to a local
upon the five parameters as the length, mass, optimum. From these empirical studies it can be
damping, external piston force and frequency. The concluded that the MPSO O is sensitive to control
steady-state
state responses of the flexible connecting rod parameter choices, specifically the inertia weight,
of a slider-crank mechanism with time time-dependent acceleration coefficients and velocity clamping.
boundary effect were obtained by Fung [2]. A slider
slider- However, wrong initialization of these parameters
crank mechanism with constantly rotating speed was may lead to divergent or cyclic behavior.
controlled by Fung et al. [3]. The mathematical model This study successfully demonstrates that the dynamic
of the coupled mechanism of a slider slider-crank formulation can give a wonderful interpretation of a
mechanism was obtained by Lin et al. [4] [4], where the slider-crank
crank mechanism in comparison with the
system is actuated by a field-orientedoriented control experimental results. Furthermore, a new identified
permanent magnet (PM) synchronous servomotor. method using the MPSO is proposed, and it is

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 181
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
confirmed that the method can perfectly sea
searches the A machine model of a PM synchronous motor can be
parameters of the slider-crank
crank mechanism driven by a described in a rotor rotating by [10] as follows:
servomotor through the numerical simulations and vq  Rs iq  pq  ws d (10)
experimental results.
vd  Rs id  pd  ws q (11)
II. DYNAMICS FORMULATION OF A
SLIDER-CRANK
CRANK MECHANISM Where
A slider-crank
crank mechanism is a single looped q  Lq iq , (12)
mechanism with a very simple construction shown in d  Ld id  Lmd I fd . (13)
Fig.1(a) and the experimental equipment of a slider
slider-
crank mechanism is shown in Fig. 1(b). It consists of
In the above equations, vd and vq are the d and q axis
three parts: a rigid disk, which is driven by a
servomotor, a connecting rod and a slider. stator voltages, id and iq are the d and q axis stator
currents, Ld and Lq are the d and q axis inductances,
Figure1 (a) shows the physical model of a slider
slider-crank
d and q are the d and q axis stator flux linkages,
mechanism, where the mass center and the radius of
the rigid diskare denoted as point “ O ” and length “ r and Rs and ws are the stator resistance and inverter
”, respectively. And “ l ” is denoted as the length of frequency, respectively. In Eq. (13), I fd is the
the connected rod AB . The angle  is between OA equivalent d -axis axis magnetizing current, and Lmd is the
and the X-axis, while the angle  is between the rod
d -axis
axis mutual inductance. The electric torque is
AB and the X-axis.axis. In OXY plane, the geometric
 m  pLmd I fd iq  ( Ld  Lq )id iq ,
3 (14)
positions of gravity centers
ers of rigid disk, connected 2
rod, and slider, respectively, are as follows: and the equation for the motor dynamics is
x1cg  0, y1cg  0 (1)  e   m  Bmr  J mr . (15)
1 1
x 2cg  r cos   l cos  , y 2cg  l sin  (2) In Eq (14), p is the number of pole pairs,  m is the
2 2
x3cg  r cos θ  l cos, y3cg  0. (3) load torque, Bm is the damping coefficient, r is the
rotor speed and J m is the moment of inertia. The basic
The mechanism has a constrained condition as principle in controlling a PM synchronous motor
follows drive is based on field orientation. The flux position in
r sin   l sin  . (4) the d  q coordinates can be determined by the shaft- shaft
position sensor because the magnetic flux generated
g
The angle  can be found from Eq. (4) as from the rotor permanent magnetic is fixed in relation
r  (5) to the rotor shaft position. In Eqs. (13-14),
(13 if id  0 the
  sin 1  sin θ .
l  d -axis flux linkage d is fixed since Lmd and I fd are
constant for a surface-mounted
mounted PM synchronous
In the kinematic analysis, taking the first and second
motor, and the electromagnetic torque  e is then
derivates of the displacement of slider B with respect
to time, the speed and acceleration of slider B are as proportional to iq , which is determined by closed-loop
closed
follows control. The rotor flux is produced in the d  axis
xB   rsin   lsin  (6) only, and the current vector is generated in the q  axis

x  
2  
B  r sin   r cos  l sin   l cos.
2
(7) for the field-oriented
oriented control. As the generated motor
torque is linearly proportional to the q  axis current as
Similarly, the angular velocity  and acceleration  the d  axis rotor flux is constant in Eq. (14), the
are obtained as follows maximum torque per ampere can be achieved.
With the implementation of field-oriented
field control, the
 rcos  PM synchronous motor drive system can be
 (8)
l cos  simplified to a control system block diagram, as
rcos  cos  rcos sin   r2 sin  cos shown in Fig. 2, in which

 . (9)
l cos 
2
 e  K t iq* , (16)

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 182
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456

Kt 
3
PLmd I fd , (17) where  m is the torque applied in the direction of r ,
2
1
and the variables  r and r are the angular speed and
H p (s)  , (18)
J m s  Bm acceleration of the disk, respectively.

where iq* is the torque current command. By According to Ref. [3], the dynamic equation of the
slider-crank
crank mechanism can be concluded to be
Mˆ    Nˆ  ,  Fˆ  
substituting (16) into (15), the following applied  (20)
torque can be obtained as follows:
 m  K t iq  J m r  Bm  r (19)

Where
2
 r 
3
 m 1 l 1
Mˆ  2m3  m2   3 r cos  cos sin 2    m2  m3 r 2 sin 2   m2   r cos   m1r 2  J m
2

 c  c  3  
c 2
  l2 r lr 2 r3  r3  r2 r2 2r
Nˆ  m2 r 2 sin  cos 1  2  cos  4 cos 2   3 cos  sin 2    m2 sin 3   m3 r 2 sin  cos 1  2 sin 2   2 cos 2   cos 
  3c c 3c 2 c  2 c  c c c

r 4 cos 2  sin 2  r 3 2  r3  1
 4
 3 sin  cos    m3 sin 3  2  Bm m2 gr cos  ,
c c  c  2
 r 
Fˆ  K t iq  FB  FE r sin  1  cos ,
 c 
c  l 2  r 2 sin 2  .

The system becomes an initial value problem and can In general, there are three attributes, current position
be directly integrated by using the fourth order x j , current velocity v j and past best position pbest j ,
Runge-Kutta method. for particles in the search space to present their
features. Each particle in the swarm is iteratively
III. IDENTIFICATIONS OF THE SYSTEM updated according to the aforementioned attributes.
A. Particle swarm optimization For example [5-9], 9], the jth particle is represented as xj
Birds (particles) flocking optimizes a certain objective = (xj,1, xj,2,…, xj,g) in the g--dimensional space. The
function in a PSO system. Each agent knows its best best previous position of the jth particle is recorded
value so far (pbest) and its position. This information and represented as pbestj = (pbestj,1, (pbestj,1 pbestj,2,…,
is analogy of personal experiences of each agent. pbestj,g). The index of best particle among all
Moreover, each agent knows the best value so fa far in particles in the group is represented by the gbestg.
the group (gbest) among pbests. This information is The rate of the position change (velocity) for particle j
analogy of knowledge of how the other agents around is represented as vj = (vj,1, vj,2,…, vj,g). The
them have performed. The PSO concept [5 [5-6] consists modified velocity and position of each particle can be
of changing the velocity of each particle toward its calculated using the current velocity and distance
pbest and gbest locations. In the PSO, eac each particle from pbestj,g to gbestj,g as shown in the following
moves to a new position according to new velocity formulas [9]:
and the previous positions of the particle. This is
compared with the best position generated by v (jt, g1)  w  v (jt, )g  c1 Rand ()  ( pbest (jt, )g  x (jt, )g )
previous particles in the fitness function, and the best  c2 Rand * ()  ( gbest (jt, )g  x (jt, )g ) ,
one is kept; so each particle accelerates in the
x (jt, g1)  x (jt, )g  v (jt, g1) . (21)
direction of not only the local best solution but also
the global best position. If a particle discovers a new j  1,2, , n; g  1,2, , m
probable solution, other particles will move closer to where n is the number of particles in a group; m is the
it to explore the region more completely in the number of members in a particle; t is the pointer of
process. iterations (generations); v (jt, )g is the velocity of the
particle j at iteration t, Vgmin  v (jt, )g  Vgmax ; w is the

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 183
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
inertia weighting factor; c1 , c2 are the acceleration where  is an adaptive decay scale,  max is set as 0.7
constants; Rand(), Rand*() are random numbers and  min is set as0.005 in this paper. RDk is the rank of
between 0 and 1; x (jt, )g is the current position of Dk assigned to the kth individual, where Dk is the
particle j atiteration t; pbestj is the pbest of particle j; distance from the individual to the current best
gbestg is the gbest of the group g. solution vector,, and is given by
Dk  vk  vbest (24)
In the above procedures, the parameter Vgmax
determine the resolution or fitness, with which regions where vk is the vector of the kth individual in the
are searched between the present position and the
combined population, and vbest is the current best
target position. If Vgmax is too high, particles might fly
solution vector.
past good solutions. If Vgmax is too low, particles may
not explore sufficiently beyond local solutions. An adaptive scheme is defined as [11]
   R  (  max   min ) / g max (25)
The constants c1 and c2 represent the weighting of   ( g )    , Fmin ( g )  Fmin ( g  1)
the stochastic acceleration terms that pull each  ( g  1)  
particle toward pbest and gbest positions. Low values  ( g ) , Fmin ( g )  Fmin ( g  1)
allow particles to roam far from the target regions (26)
before being tugged back. On the he other hand, high and
values result in abrupt movement toward or past target  ( g  1)   min ; if  ( g )      min (27)
regions.
where   is the step size; Fmin is the minimum value
Suitable selection of inertia weighting factor w of fitness functions; R is the regulating scale and is
provides a balance between global and local set as1.25 in this paper, and g max is the maximum
explorations, thus requiring less iteration on average
allowable number of iterations.
iterations
to find a sufficiently optimal solution. As originally
developed, w often decreases linearly from about 0.9
Individuals will be ranked in ascending according to
to 0.4 during a run. In general,
eral, the inertia weighting
their fitness scores by a sorting algorithm. The PS
factor w is set according to the following equation [6
[6-
individuals are transcribed along with their fitness for
7]:
the next generation. If the new population does not
w  wmin
w  wmax  max  iter (22) include the current best solution, the best solution
itermax must be replaced with the last individual in the new
where itermax is the maximum number of iterations population. In addition, a gradually decreased decay
(generations), and iter is the current number of scale can satisfy a successive statistic searching
iterations. process by first using the diversification (bigger
(  ) to
explore more regions, and then the intensification
B. Modified particle swarm optimization (smaller  ) to exploit the neighborhood of an elite
The main point of the MPSO differs from the PSO is solution. The current best solution (point A) for a
to consider the “distance” in its fitness function to minimum fitness problem as shown in the Fig. 3 may
avoid converging to a local optimum. Assign a rank not reach the global optimum [11-12],
[11 and there are
(i.e., the number place 1, 2, 3, …, etc.) RE k tothe three electable solutions exist. Generally, solutions
calculated error of each new individual, vk , with slightly better fitness (point C or B) prevailed, so
the solution trapped into the valley prematurely. The
k  1, , PS , PS is the population size. A combined more attractive solution(point G) is relatively far
population with 2  PS individuals is formed. Unlike away from point A, but itt nears the global optimal. To
previously developed statistic methods, the concept of prevent prematurity,, point G with slightly worse
“distance” is added to the fitness function to prevent fitness than C, it needs a higher rank to be selected.
from being trapped in a local minimum. The fitness That is, a higher RDk is awarded toa longer Dk .
score of the kth individual is modified by [11
[11-12]
Fk  RE k    RDk , k  1, ,2  PS . (23)

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 184
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
Stopping Criteria: Stopping criteria is given in the The main parameters of a slider-crank
slider mechanism and
following order: servomotor used in the numerical simulations and the
1. Maximum allowable number of iterations reached. experiments are as follows
2. Number of iterations reached without improving m1  0.232 kg , m2  0.332 kg ,
the current best solution. m3  0.600 kg , r  0.030 m ,
Fig. 4 shows the flow chart of the proposed algorithm
algorithm. l  0.217 m , FB  0.100 N ,
FE  0.000 N , iq  0.400 A ,
C. Parameter identifications J m  6.700 105 Nms 2 ,
Kt  0.5652 Nm / A ,
How to define the fitness function is the key point of
Bm  1.430 102 Nms / rad.
the MPSO, since the fitness function is a figure of
merit, and could be computed by using any domain
knowledge. In this paper, we adopt the fitness The physical model of the slider-crank
slider mechanism
function as follows [13-14]: driven by a servomotor. In the parameter
n identification, we utilize the MPSO and PSO methods
F ( parameters)   Ei2 (28) to identify the 5 parameters m1 、 m2 、 m3 、 r and l
i 1
simultaneously, and the fitness function is described
Ei  x (i )  x(i ) (29) as Eq. (20). The identified results are given in Table
E 1.
where n is the total number of samples and i is the
*( i )
calculated error of the i th sampling time, x is a Figure 6 shows the displacement errors of the slider-
slider
solution by using the fourth-orderr RungeRunge-Kutta crank mechanism by both the PSO and MPSO
method to solve the dynamic Equations (31) for the methods. Furthermore, their displacement errors are
PM synchronous servomotor drive coupled with a all about ± 0.2  m . Figure 7 shows the convergence
slider-crank
crank mechanism with the parameters identified characteristics in PSO and MPSO methods of the
(i )
from these two methods, and x is the displacement slider-crank
crank mechanism system. It is seen from Fig. 7
measured experimentally at the i th sampling time. that the proposed MPSO method is superior to the
PSO method.
IV. SIMULATIONS AND EXPERIMENTAL
RESULTS V. CONCLUSION
A block diagram of the computer control system for The dynamic formulations of a slider-crankslider
the PM synchronous servomotor drive coupled with a mechanism driven by a field-oriented
field PM
slider-crank mechanism is shown in Fig. 5(a) and the synchronous motor have been successfully formulated
experimental equipments of a slider-crank
crank mechanism with only one independent variable. Furthermore, the
of computer control system is shown in Fig. 5(b). The main objective of this study is to utilize PSO and
control algorithm is implemented using a Pentium MPSO methods to identify a slider-crank mechanism
computer and the control software is LABVIEW. The driven by a servomotor. According to the comparisons
PM synchronous ous servomotor is implemented by between identified results and displacement errors, it
MITSUBISHI HC-KFS43 KFS43 series. The specifications is found that MPSO method has the best matching
are shown as follows: Rated output 400 (W), Rated with the experimental results.
torque 1.3 (Nm), Rated rotation speed 3000 (rpm) and
Rated current 2.3 (A). The servo is implemented by It is concluded that the implementations of MPSO are
MITSUBISHI MR-J2S-40A1.. The control system is different from the PSO in five aspects. Firstly, its
Sine-wave
wave PWM control, which is a current control fitness function considers the distance to avoid
system. In order to measure the angle and angular converging to a local optimum. Secondly, for the
speed of the disk and the position and velocity of the MPSO, vectors with good enough fitness scores
slider B, the interface of the device is implemented by would be used as candidates to create new solutions.
motion control card PCI-7342.7342. It can measure the Thirdly, it has the advantage of the MPSO to conquer
angle of the disk and the position of slider B at the various constraints without using the fitness function
same time. with penalties, and can perform better. Fourthly, the
solution is coded with a decimal representation,
representation and
saves computer memory. Finely, the gradually

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 185
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
decaying parameters can satisfy a successive statistic 10. Krause, P. C., Analysis of Electric Machinery
searching process by first using the diversification (New York: McGraw-Hill)
Hill) , 1986.
(bigger
bigger parameters) to reserve the larger attractive
11. Lin, W. M., Cheng, F. S., and Tsay, M. T.,
region. Then, the intensification (smaller pparameters)
“Nonconvex Economic Dispatch by Integrated
used to search the small neighborhood of an elite
Artificial
icial Intelligence”, IEEE Transactions on
solution.
Power systems, Vol. 16, No. 2, pp. 307-311,
307 2001.
REFERENCES 12. Lin, W. M., Cheng, F. S., and Tsay, M. T., “An
1. Viscomi, B. V., and Arye, R. S., “Nonlinear Improved Tabu Search for Economic Dispatch
Dynamic Response of Elastic Slider
Slider-Crank With Multiple Minima”, IEEE Transactions on
Mechanism” ASME J. of Eng.
g. for Industry, Vol. Power systems, Vol. 17, No. 1, pp. 108-112, 2002.
93, pp. 251-262, 1971. 13. Zhang, X., Huang, Y., Liu, J., Wang, X., and Gao,
2. Fung, R. F., “Dynamic Response of the Flexible F., “A Method Identifying the Parameters of
Connecting Rod of a Slider-Crank
Crank Mechanism Bouc-Wen
Wen Hysteretic Nonlinear Model Based on
with Time-Dependent
Dependent Boundary Effect” Genetic Algorithm”, Intelligent Processing
Computer & Structure, 63, No. 1, pp. 79
79-90, 1997. Systems, pp. 602-605,
605, 1997.
3. Fung, R. F., Lin, F. J., Huang, J. S. and Wang, Y. 14. Ha, J. L., Kung, Y. S., Fung,
Fu R. F., and Hsien, S.
C., “Application of Sliding Mode Control with a C., “A Comparison of Fitness Functions for the
Low-pass
pass Filter to the Constantly Rotating Slider
Slider- Identification of Piezoelectric Hysteretic Actuator
Crank Mechanism” The Japan Society of Based on the Real-Coded
Coded Genetic Algorithm”,
Mechanical Engineers, C, 40, No. 4, pp 717717-722, Sensors and Actuators , Vol.132, pp.643-650,
pp.643
1997. 2006.
4. Lin, F. J., Fung, R. F., and Lin, Y. S., “Adaptive A Y
PM synchronous
Control of Slider-Crank
Crank Mechanism Motion: l r motor
m3 B m2 
Simulation and Experiments”,, International X  
Journal of System Science, 28, No. 12, pp. 1227
1227- FE
O m 1

1238, 1997. Fb X b l R
Gear box
5. Kennedy, J., Eberhart, R. C., “Particle Swarm
Optimization”, Proceedings of the IEEE Fig. 1(a) The physical model of a slider-crank
slider
International Joint Conference on Neural mechanism.
Networks, IEEE Press, pp. 1942-1948,
1948, 1995.
6. Eberhart, R. C., Kennedy, J., “A New Optimizer
Using Particle Swarm Theory”, Proceedings of the
Sixth International Symposium on Mic
Micro machine
and Human Science, Nagoya, Japan, pp.39
pp.39-43,
1995.
7. Kennedy, J., “The Particle Swarm: Social
Adaptation of Knowledge”, Proceedings of the
IEEE International Conference on Evolutionary
Computation, Indianapolis, USA, pp. 303
303-308,
1997. Fig. 1(b) The experimental equipments of a slider-
slider
crank mechanism
8. Naka, S., Genji,
nji, T., Yura, T., Fukuyama, Y., “A Fig. 1 The slider-crank
crank mechanism
Hybrid Particle Swarm Optimization for
Distribution State Estimation”, IEEE Transactions
on Power systems Vol. 18, No.1, pp.60
pp.60-68, 2003.
9. Shi, Y., Eberhart, R. C., “Empirical Study of
Particle Swarm Optimization”, Proceed
Proceedings of the
IEEE Congress on Evolutionary Computation,
IEEE Press, pp.1945-1950, 1999.

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 186
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
Fig.2 Simplified control block diagram

(b)
Fig.5
5 The experimental setup. (a) Computer control
system block diagram; (b) The experimental
Fig.3
3 The concept of distances. equipments of a slider-crank
crank mechanism of
Computer control system.
Step 1 Data Collection

Step 2 Initialization Production of initial population

- particle : including m1, m2, m3, r, l


- parameter of PSO : c1 , c2 , w, R, ρmin, ρmax

Step 3 Evaluation Calculate Error and Distance


- Rank and Sort the Error and Distance
- Evaluate the Fitness Function
- Find the gbest and pbest

Step 4 Position Change Calculate velocity and position

- Generate the new particle


- update all parameter mentioned in Step2

Step 5 Termination Converge in optimum

Fig. 6 The comparisons in displacement errors


reach max. iteration number ? No Step 3
between the PSO and MPSO methods.
Yes

End

Fig.4 The flow chart of the MPSO.

PM
Command Synchronous
Current Motor


Voltage

Fig. 7 Comparison of convergence characteristics


in PSO and MPSO methods of the slider-crank
slider
(a) mechanism system.

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 187
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
Table 1. The identified parameters of the numerical simulations.
Parameter m1 kg  m2 kg  m3 kg  r m l m 
Feasible domain 0.000 ~ 1.000 0.000 ~ 1.000 0.000 ~ 1.000 0.000 ~ 0.100 0.000 ~ 1.000
The actual value 0.232 0.332 0.600 0.030 0.217
The identified value of the
0.311 0.304 0.752 0.024 0.325
PSO method
The identified value of the
0.234 0.331 0.603 0.030 0.216
MPSO method

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 188