Speed control of an artificial neural network modeled field oriented three level inverter fed an induction motor

IV SPEED CONTROL OF AN ARTIFICIAL NEURAL NETWORK MODELED FTELD ORIENTED THREE LEVEL INVERTER FED AN INDUCTION MOTOR (Ph.D. Thesis) Okan BİNGÖL GAZI UNIVERSITY INSTITUTE OF SCIENCE AND TECHNOLOGY June 2005 ABSTRACT In this thesis, the speed control of induction motor was achieved using artificial neural network. An artificial neural network modelling was developed, instead of field oriented control and transform matrix for speed control of an induction motor. Complexity of field oriented controller and the affect of the motor parameter variations were reduced using new modelling developed. A multi layer feed-forward neural network was chosen as an artificial neural network. Modelling and application of the induction motor at different operation conditions were done using two and three levels inverters for artificial neural network modelling. TMS320LF2407A digital signal processor was used for experimental study. Current waveform of three levels inverters was found very similar to sine wave form than the two levels inverters, as proved by analysing the harmonics. Science Code : 626.01.01 Key Words : Artificial neural network, Induction motor, Three-level inverter, Space vector modulation. Page number :160 Adviser :Prof. Dr. Çetin ELMAS