Date of Completion

12-15-2017

Embargo Period

12-31-2018

Keywords

Induction machines, core-loss model, stray loss, loss minimization control, machine fault, fault diagnosis, fault-tolerant control, maximum-per-ampere control

Major Advisor

Ali M. Bazzi

Associate Advisor

Yang Cao

Associate Advisor

Sung-Yeul Park

Associate Advisor

Krishna R. Pattipati

Field of Study

Electrical Engineering

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

Three-phase induction motor (IM) drive systems are the most important workhorses of many industries worldwide. This dissertation addresses improved modeling of three-phase IM drives and model-based control algorithms for the purpose of designing better IM drive systems. Enhancements of efficiency, availability, as well as performance of IMs, such as maximum torque-per-ampere capability, power density, and torque rating, are of major interest. An advanced power loss model of three-phase IM drives is proposed and comprehensively validated at different speed, load torque, flux and input voltage conditions. This model includes a core-loss model of three-phase IMs, a model of machine mechanical and stray losses, and a model of power electronic losses in inverters. The drive loss model shows more than 90% accuracy and is used to design system-level loss minimization control of a motor drive system, which is integrated with the conventional volts-per-hertz control and indirect field-oriented control as case studies. The designed loss minimization control leads to more than 13% loss reduction than using rated flux for the testing motor drive under certain conditions. The proposed core-loss model is also used to design an improved model-based maximum torque-per-ampere control of IMs by considering core losses. Significant increase of torque-per-ampere capability could be possible for high-speed IMs. A simple model-based time-domain fault diagnosis method of four major IM faults is provided; it is nonintrusive, fast, and has excellent fault sensitivity and robustness to noise and harmonics. A fault-tolerant control scheme for sensor failures in closed-loop IM drives is also studied, where a multi-controller drive is proposed and uses different controllers with minimum hand-off transients when switching between controllers. A finite element analysis model of medium-voltage IMs is explored, where electromagnetic and thermal analyses are co-simulated. The torque rating and power density of the simulated machine could be increased by 14% with proper change of stator winding insulation material. The outcome of this dissertation is an advanced three-phase IM drive that is enhanced using model-based loss minimization control, fault detection and diagnosis of machine faults, fault-tolerant control under sensor failures, and performance-enhancement suggestions.

COinS