History Of Fuzzy Logic

. The Fuzzy Logic is a form of a systematic reasoning that can be integrated into automation systems with classical human reasoning schemes. Fuzzy theory was first suggested and probed by Prof Zadeh in 1965 [11] Fuzzy systems are apprehension based or rule based systems. The heart of a fuzzy system is a knowledge base inherent of the so-called If-Then rules. After allocating the fuzzy sets and their membership functions, rules must be noted to place an action to be taken for each combination of control variables. These rules will correlate the input and output variable by using IF-Then statements.
In a fuzzy logic system, the inference structure resolve upon the type of rules which are to be supplied for the corresponding inputs by coupling

These crisp measurements can be obtained from an expert knowledge of the system. (2) Fuzzification it is generally the process of altering a real scalar value into a fuzzy value. This can be achieved by the different type of fuzzifies or membership functions. In this phase, the crisp variables eω (k) and deω (k) are transformed in to fuzzy variables e ω (error) and de ω (change in error) respectively. Each fuzzy variable is an associate of the subsets with a level of either membership μ varying between 0 (non-member) and 1 (full-member). The Fuzzification block Computes the values of input variables.(3) Fuzzy Inference engine Fuzzy inference system (FIS) employs a fuzzy set theory for portraying of the inputs with the outputs. Mamdani type of inference system is used in this method. In this stage, the variables error and change in error are processed by an inference engine that accomplished 7 RULES. These rules are confirmed using the knowledge of the system performance by the control engineers. Each rule is manifested in the form of the following example: IF (Error is Negative Large) THEN (Control signal provided will also be a negative large to compensate the Error). (4) Knowledge Based / Rule Based A decision building logic replicating a human decision ideology reduces the

Before starting the simulation in MATLAB/SIMULINK, the FLC is to be constructed, but this may involve the use of various steps, first fuzzy inference system (FIS) Editor is opened and this file is being created using the fuzzy logic toolbox. The construction of a FLC requires the selection of proper membership functions. After the appropriate membership functions are chosen rule base is generated. The set of linguistic rules is the key part of a fuzzy controller. Different linguistic variables used in the design of a rule base for output of the FLC are enlisted in Table 1; the response of the FLC is being acquired by using in SIMULINK/MATLAB. Normally two inputs, one is the speed error having the 5 membership functions, two trapezoidal and three triangular and the other is the change in error having 3 membership functions, two trapezoidal, one triangular. Also have one output change in control having 5 membership functions and all are triangular in shape. When the membership functions and Fuzzy rules are resolute, the Surface viewer is developed showing the correlation among the inputs and outputs, thus a properly controlled control output signal is obtained with the use of FLC. Membership function for inputs is as shown in Fig 6(a) and 6(b) respectively while the membership function for output is shown in Fig 6(c) Mamdani type of inference system with Centroid