Fuzzy sliding mode control and genetic sliding mode control for position of vactor controlled induction motors

Eventhough scalar control methods for induction motor control have cheap price, they have poor characteristics and restrict inverter-motor performance. These restrictions disappear when field orientation is used. Generally, a disadvantage of field orientation method is the necessity of speed or flux sensor on the motor. Vector control techniques is classified in two groups as indirect and direct methods. In this thesis, a mathematical d-q model for squirrel cage induction motor has been used. The synchronously rotating vector components of stator current isq and W controls independently the torque and rotor flux, respectively. Simulation program has been prepared basing on the model. Sliding Mode Control (SMC) system used as controller is also known as variable structure control system. The design of variable structure controller generally consists of two steps which are hitting and sliding phases. Firstly, the system is directed towards a switching surface by a feedback control law, then sliding mode occurs. When the system states enter the sliding mode, the dynamic of the system are determined by the choice of sliding surface. Variable structure systems such as motor control, robotic manipulators and indefinite systems have an important place in the control of modern non-linear system. However, there are some difficulties such as existing many switches between the control bounds, which can not be realized by real controllers. A thin boundary layer neighbouring the switching surface is occurred to overcome this problem. Hence, the discontinuity is eliminated. The boundary layer can be achieved by a saturation function. However, introducing boundary layer, steady- state errors may occur. The mentioned problem in sliding mode control theory can be compensated with fuzzy logic principles. Fuzzy-sliding mode control combines fuzzy logic and variable structure system theories. The main property of fuzzy logic is being based on classical control theories and the experiences of human-being. Genetic algorithm (GA) is based on an analogy to the genetic code in our own DNA (deoxyribonucleid acid) structure, where its coded chromosome is composed of many genes. In GAs, there are three basic operators which are reproduction, crossover and mutation. In this study, firstly, boundary layer sliding mode controller and then fuzzy-sliding mode controller and genetic sliding mode controller are applied to the indirect vector controlled induction motor. Successful results are obtained. Simulations used in this thesis are performed by using MATLAB and results are compared with each other. Keywords: Induction motor, sliding mode control, fuzzy logic, genetic algorithms.