Design, implementation and performance of conventional and intelligent control schemes and speed estimators for variable speed induction motor drives
This thesis discusses the implementation and performance of various conventional and artificial-intelligence-based control and estimation techniques for variable speed induction motor drives. The drive types considered are scalar drives, and high-performance vector and direct torque controlled drives. In the first part, various conventional control design techniques are developed and applied to conventionally controlled drives. The performance of these drives is compared to that of fuzzy logic controlled drives, for which a specific design method is developed using universal fuzzy logic controllers. The developed methods are studied using simulations and experimental results. Implementation issues are presented in detail. In the second part, the design and implementation of conventional and artificial-intelligence-based speed estimators for use in scalar, vector and direct torque controlled drives are discussed. The conventional estimators considered comprise both a mathematical-model-based speed estimator and a model-reference-adaptive-system-based speed estimator. The artificial-intelligence-based estimators considered use artificial neural networks which are trained on data collected in simulations. The performance of both types of estimator, in the entire operating range of the drive, is discussed extensively.