APPLICATIONS OF INTELLIGENT CONTROL TECHNIQUES FOR IMPROVED PERFORMANCE IN POWER ELECTRONIC CONVERTERS

Abstract

Intelligent control techniques based upon Expert Systems, Fuzzy Logic and Artificial Neural Networks are recently having a significant impact on power electronics and motor drives. Neural Networks in particular have created a new and advancing frontier in power electronics, which is already a complex and multidisciplinary technology that is going through dynamic evolution in the recent years. Converters are evidently the crux of power electronics application. Thus we must experiment, analyze and realize the utility of such intelligent control techniques in this field. Further a comprehensive analysis with respect to the conventional methods can be established. Neural Networks have proven to be of excellent usage in scenarios involving complex, non-linear and transcendental mathematical models. Providing robust and flexible architectures that can learn in real-time, these networks can be a ready replacement for on-line calculations and storage apparatus. Through this report, we explore how a neural network system can be used in PWM strategies to produce near perfect sinewaves devoid of the harmonics which are sometimes responsible for instabilities. Further we explore how such intelligent control methodologies can be used in applications such as speed control of an induction machine fed by a voltage source inverter. Also through this dissertation, we take steps further from Neural Networks based control to other techniques such as Fuzzy Logic based systems to analyze the utility, advantages and disadvantages of intelligent control techniques for different scenarios. The study ends with a brief description of a successful realization of a hardware prototype.