Академический Документы
Профессиональный Документы
Культура Документы
2.1 Tools
In the course of 50 years of research, AI has developed a large number of tools to solve the most difficult problems in various fields. Few of them are mentioned below-
1.Neural networks
A neural network, also known as a parallel distributed processing network, is a computing solution which consists of interconnected processing elements called nodes or neurons that work together to produce an output function. It is loosely modeled after cortical structures of the brain.
2.Fuzzy Logic
Logic is used for knowledge representation and problem solving, but it can be applied to other problems as well. Fuzzy logic is a superset of conventional(Boolean) logic that has been extended to handle the concept of partial truth values between "completely true" and "completely false". As its name suggests, it is the logic underlying modes of reasoning which are approximate rather than exact. Fuzzy logic allows the truth of a statement to be represented as a value between 0 and 1, rather than simply True (1) or False (0).
3. Control theory
Control theory, the grandchild of cybernetics, has many important applications, especially in robotics.
4.Languages
AI researchers have developed several specialized languages for research, including LISP and PROLOG.
Incoming Weighted Connections Neuron Output = F ( Inputs ) Outgoing Weighted Connections Fig 3.1 Schematic Diagram of the Neuron General model of the processing unit of ANN can be considered to have the following three elements.
ARTIFICIAL INTELLIGENCE IN POWER STATIONS The Output Layer where the nodes are called output units, which encode possible concepts (or values) to be assigned to the instance under consideration. For example each output unit a class of objects. Other main important concept is the weightage for the connected unit. It can be real or integer numbers. They can be confined to a range and are adjustable during network training. When training is completed, all of them attain fixed values.
Supervised learning or Associative learning is the one in which the network is trained by providing it with input and matching output patterns. These input-output pairs can be provided by an external teacher, or by the system which contains the neural network (self-supervised).
Unsupervised learning or Self-organization is the one in which an (output) unit is trained to respond to clusters of pattern within the input. In this paradigm the system is supposed to discover statistically salient features of the input population. Unlike the supervised learning paradigm, there is no a priori set of categories into which the patterns are to be classified; rather the system must develop its own representation of the input stimuli.
Reinforcement Learning This type of learning may be considered as an intermediate form of the above two types of learning. Here the learning machine does some action on the environment and gets a feedback response from the environment. The learning system grades its action good (rewarding) or bad (punishable) based on the environmental response and accordingly adjusts its parameters. Generally, parameter adjustment is continued until an equilibrium state occurs
ARTIFICIAL INTELLIGENCE IN POWER STATIONS For any crisp set C it is possible to define a characteristic function C: U [0,1] instead from the two-element set {0,1}. The set that is defined on the basis of such an extended membership function is called as fuzzy set. Fuzzy rules are elementary or composed proposals. They result from a conjunction between elementary fuzzy proposals. A fuzzy rule is composed of a premise and a conclusion. The classical structure of a rule is If < premise> then <conclusion>. When the premise is an elementary fuzzy proposal, the rule is described as follows. If <x is A> then < conclusion>. The x is a variable; generally real, defined on a referential called the universe of discourse, given as a capital letter here X. A is a linguistic term, taken in a set of terms noted as TX. Basic concept of fuzzy logic's is fuzzy " If then Rule " or Fuzzy Rule.
mA (x) mB (x)
F: U [0,1]
So, every element u of U has a membership degree F (u) {0,1}. F is completely determined by the set of tuples F = {(u, F (u)) | u U} 8
R(u,v) /(u,v)
This is a binary fuzzy relation on U x V. If U and V are controllable (discrete) universes, then R = R(u,v) /(u,v)
UxV
The integral symbol denoted the set of all tuples on U x V denoted by R(u,v) /(u,v)
-1
0 0 1 0 a b 1 0 1
10
Fig.6.1 Intelligent Substation System Configuration (Image). The whole substation system is combined by optical LAN, and apparatus composition is simplified.
For convenience, power system higher level controls are classified here as: Generation scheduling and automatic control: includes unit commitment, economic dispatch, and automatic generation control; in the past, well established control methods were used but this situation has been changing to deal with the new scenario created by the power industry restructuring; Voltage control: is mostly of the local type but some systems have already gone to a higher coordinated secondary control to allow a more effective use of reactive power sources and increase stability margins; Preventive security control: has the objective to detect insecure operating points and to suggest corrective actions; the grand challenges in this area are on-line Dynamic and Voltage Security Assessment (DSA and VSA); Emergency control: manages the problem of controlling the system after a large disturbance; it is an event driven type of control and includes special protection schemes; Restorative control: its main function is to re-energize the system after a major disturbance followed by a partial or total blackout.
14
Fig 7.1 Gas Combined Switchgear Miniaturization by Digital Correspondence Sensor. 550-kV GCS (gas combined switchgear) GCB: gas circuit breaker CT: current transformer PT: potential transformer Present studies on miniaturizing conventional equipment have so far been aimed at standardizing series.
15
16
hhhhh.
Figure 1Highly flexible and intelligent energy system infrastructures are required to facilitate substantially higher amounts of renewable energy than todays energy systems and thereby lead to the necessary CO reductions as well as ensuring the future security of energy supply in all regions of the world.
Links between the intelligent infrastructure and the traditional power system structure are the basis for the future flexible and intelligent energy system
Power plants/CHP
HV transmission
Transformer
LV transmission
End use
Intelligent, two way communication between suppliers and end-users together with distributed generation further enhances the flexibility
Utility Communications
Renewables
Dynamic systems Distribution control
PV
Internet
Data management
Plug-in hybrids
17
Wireless Communication
CHAPTER 8
_________________________________________________________________
Neural Networks have been used in a board range of applications including: pattern classification, pattern recognition, optimization, prediction and automatic control. In spite of different structures and training paradigms, all NN applications are special cases of vector mapping. The following fields has attracted the most attention in the past five years: 1-Load Forecasting 2-Fault Diagnosis/Fault Location 3-Economic Dispatch 4-Security Assessment 5-Contingency Screening
1 PREPROCESSING
Fig 8.3 Fault Diagnosis process Progress in the areas of communication and digital technology has increased the amount of information available at supervisory control and data acquisition (SCADA) systems. Although information is very useful, during events that cause outages, the operator may be overwhelmed by the excessive number of simultaneously operating alarms, which increases the time required for identifying the main outage cause and to start the restoration process. Besides, factors such as stress and inexperience can affect the operators 20
ARTIFICIAL INTELLIGENCE IN POWER STATIONS performance; thus, the availability of a tool to support the real-time decision-making process is welcome. The protection devices are responsible for detecting the occurrence of a fault, and when necessary, they send trip signals to circuit breakers (CBs) in order to isolate the defective part of the system. However, when relays or CBs do not work properly, larger parts of the system may be disconnected. After such events, in order to avoid damages to energy distribution utilities and consumers, it is essential to restore the system as soon as possible. Nevertheless, before starting the restoration, it is necessary to identify the event that caused the sequence of alarms such as protection system failure, defects in communication channels, corrupted data acquisition. The heuristic nature of the reasoning involved in the operators analysis and the absence of an analytical formulation, leads to the use of artificial intelligence techniques. Expert systems, neural networks, fuzzy logic, genetic algorithms (GAs), and Petri nets constitute the principal techniques applied to the fault diagnosis problem.
21
Fig 8.4 Data flow in power System Operation Generally there are two types of security assessments: static security assessment and dynamic security assessment. In both types different operational states are defined as follows: Normal or secure state: In the normal state, all customer demands are met and operating limit is within presented limits. Alert or critical state: In this state the system variables are still within limits and constrain are satisfied, but little disturbance can lead to variable toward instability. Emergency or unsecure state: the power system enters the emergency mode of operation upon violation of security related inequality constraints.
22
23
ARTIFICIAL INTELLIGENCE IN POWER STATIONS (i) A great number of test data (including the preventive test and other tests), some expert experience and some criteria which can be directly numerated should be translated into fuzzy numbers. This process is called numerical uncertainty translation. (ii) Some expert experience and some criteria which are expressed in linguistic language cannot be numerated directly. They also need to be changed into fuzzy numbers. The process is called linguistic uncertainty translation. In fuzzy diagnosis system we take first, the associated membership functions of fuzzy subsets were determined empirically or basically in a trial-and-error manner, while the conventional DGA diagnosis criteria were to be implicitly obeyed. And second step, a great number of previous diagnosis records of dissolved gas were mainly employed as a test purpose rather than as a development basis of the fuzzy diagnosis system. The information inherently contained in the numerical data was not fully utilized in establishing the diagnosis system. And the last step, due to the wide variety of conditions which affect the results of DGA, the diagnosis system should be continuously maintained or modified according to the cases newly obtained. However, manual knowledge acquisition and knowledge base revision out of the large numerical data are quite tedious, often resulting in a lengthy process of generating the fuzzy if-then rules and the membership functions of the corresponding fuzzy subsets. The approach is based on the interpretation of DGA data using Fuzzy Logic (FL). The proposed diagnostic method adopts indicators related to the ratios C2H4/C2H6, C2H2/C2H4 CH4/H2 and to the concentration of specific gases such as hydrogen, carbon monoxide, methane, ethane, ethylene, acetylene. Different combinations of these four codes represent different fault patterns, overheating, arcing, and corona. Fuzzy Analysis an integrated analysis, it can diagnose the fault of transformer effectively and manage the data of DGA in oil of transformer.
26
Fuzzification
Defuzzification
Diagnosis Result
ARTIFICIAL INTELLIGENCE IN POWER STATIONS transferred to the load through a PWM inverter. In order to extract maximum power from the wind energy system and transfer it to the load, a fuzzy logic controller has to be provided to regulate the modulation index of the PWM inverter based on the input signals. By fuzzifying these signals and the use of rules based on these fuzzified signals, the fuzzy control is performed giving the fuzzy output required after Defuzzification. This will provide an optimum utilization of the wind energy.
ARTIFICIAL INTELLIGENCE IN POWER STATIONS can offer better damping effects for generator oscillations over a wider range of operating conditions than conventional regulators. The FLC, based on a set of fuzzy logic operations that are performed on controller inputs, provides a means of converting linguistic control requirements based on expert knowledge into an efficient control strategy. Using unsupervised learning of ANN generates a fuzzy associative matrix.
29