Robust fault tolerant control of induction motor system

Wang, Zhihuo

January 2018

Thesis or dissertation

© 2018 Zhihuo Wang. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder.

Research into fault tolerant control (FTC, a set of techniques that are developed to increase plant availability and reduce the risk of safety hazards) for induction motors is motivated by practical concerns including the need for enhanced reliability, improved maintenance operations and reduced cost. Its aim is to prevent that simple faults develop into serious failure. Although, the subject of induction motor control is well known, the main topics in the literature are concerned with scalar and vector control and structural stability. However, induction machines experience various fault scenarios and to meet the above requirements FTC strategies based on existing or more advanced control methods become desirable. Some earlier studies on FTC have addressed particular problems of 3-phase sensor current/voltage FTC, torque FTC, etc. However, the development of these methods lacks a more general understanding of the overall problem of FTC for an induction motor based on a true fault classification of possible fault types.

In order to develop a more general approach to FTC for induction motors, i.e. not just designing specific control approaches for individual induction motor fault scenarios, this thesis has carried out a systematic research on induction motor systems considering the various faults that can typically be present, having either “additive” fault or “multiplicative” effects on the system dynamics, according to whether the faults are sensor or actuator (additive fault) types or component or motor faults (multiplicative fault) types.

To achieve the required objectives, an active approach to FTC is used, making use of fault estimation (FE, an approach that determine the magnitude of a fault signal online) and fault compensation. This approach of FTC/FE considers an integration of the electrical and mechanical dynamics, initially using adaptive and/or sliding mode observers, Linear Parameter Varying (LPV, in which nonlinear systems are locally decomposed into several linear systems scheduled by varying parameters) and then using back-stepping control combined with observer/estimation methods for handling certain forms of nonlinearity.

In conclusion, the thesis proposed an integrated research of induction motor FTC/FE with the consideration of different types of faults and different types of uncertainties, and validated the approaches through simulations and experiments.

School of Engineering and Computer Science, The University of Hull
Qualification level
Qualification name
7 MB
QR Code