Loading...
Adaptive High Gain PI Controller for Speed Regulation of Induction Motor Drive System
AlBarri, Sarah Adnan
AlBarri, Sarah Adnan
Date
2019-12
Authors
Advisor
Type
Thesis
Degree
Description
A Master of Science thesis in Electrical Engineering by Sarah Adnan AlBarri entitled, “Adaptive High Gain PI Controller for Speed Regulation of Induction Motor Drive System”, submitted in December 2019. Thesis advisor is Dr. Habibur Rehman and thesis co-advisor is Dr Shayok Mukhopadhyay. Soft copy is available (Thesis, Approval Signatures, Completion Certificate, and AUS Archives Consent Form).
Abstract
The performance of a speed regulator of an indirect field oriented (IFO) induction motor with a fixed gain proportional-integral (PI) controller degrades specifically under external disturbances and parameter variations. To date, most of the manufacturing industry prefers the use of PI controllers due to their implementation simplicity and predictable PI tuning behavior. One of the proposed solutions for tuning PI gains is by using high gain adaptation, which automatically adjusts the gains based on the motor speed tracking. Despite various proposed techniques in which high gain adaptive PI tuning is utilized, instability occurs when this controller is tested on an induction motor. The high gain adaptive PI gains keep increasing boundlessly under external disturbances and encoder noise, thus making the system unstable. To overcome this instability, sigma, deadzone, and epsilon modifications are applied to the high gain adaptive law. In this work, the performance of the high gain adaptively tuned PI controller with sigma, deadzone, and epsilon modification is evaluated. The epsilon high gain controller is found to perform better than sigma and deadzone modifications. Therefore, detailed experimental validation of the epsilon modified high gain controller is performed. The tests applied are step and square wave reference tracking, disturbance rejection, detuning and field weakening. Moreover, the effect of the initial conditions of the adaptive PI gains is investigated. The results are compared with the conventional fixed gains PI controller. The epsilon modification high gain adaptively tuned PI controller is also investigated on the electric vehicle traction system. The effects of the adaptively tuned PI gains on the battery state of charge is noted using Coulomb’s counting method. The evaluation results show that epsilon modified high-gain adaptive PI controller provides a better speed tracking performance than the fixed gain PI controller, while requiring almost the same torque commanded current.