Applied Fuzzy Logic Controls for Improving Dynamic Response of Induction Machines

Abstract

This thesis presents a novel approach in control systems for improving the dynamic response of the induction machine. This approach leads to a better and improved control of the torque and current response of the induction machine when compared to the classical proportional-integral (PI) type controller with de-coupling terms. Mismatches in the actual parameters and the estimated parameters of the induction machine occur for several reasons such as: incorrect parameter estimation, changes in stator and rotor inductance due to saturation, stator and rotor resistance varying with temperature, etc. Under the classical approach, the de-coupling errors resulting from the parameter mismatches can become very large at higher machine rotational speeds. Under such conditions, the classical approach results in poor dynamic control of the torque and current response of the induction machine. Therefore, an advanced fuzzy logic controller is presented as a better alternative to the classical controller. The fuzzy logic-based d-q controller, based on its non-linear approach, provides robust control of the torque and current response of the induction machine even in the presence of mismatched parameters. Furthermore, the performance of the fuzzy logic controller is not dependent on the machine rotational speed. Using MATLABSIMULINK tools, the performance of the fuzzy controller is evaluated with mismatched machine parameters at various machine rotational speeds. The results show that the use of the fuzzy logic controller offers a superior control of the torque and current response of the induction machine, independent of the motor rotational speed when compared with the use of the classical controller.