MODELING AND CONTROL OF A HIGH-SPEED SOLID-ROTOR SYNCHRONOUS RELUCTANCE FLYWHEEL MOTOR/GENERATOR

Open Access
Author:
Park, Jae Do
Graduate Program:
Electrical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
March 08, 2007
Committee Members:
  • Heath Hofmann, Committee Chair
  • Kwang Yun Lee, Committee Member
  • Jeffrey Scott Mayer, Committee Member
  • Charles E Bakis, Committee Member
Keywords:
  • magnetic nonlinearity compensation
  • feedforward control
  • model-based control
  • energy storage system
  • flywheel
  • synchronous reluctance motor
  • rotor current time harmonic analysis
  • LC filter design
Abstract:
This thesis presents a control system for a high-speed solid-rotor synchronous reluctance flywheel motor/generator. The objective of this research is to derive a model for a solid-rotor synchronous reluctance machine and provide a control scheme based on the model which has stable performance at high speed. The control system should be robust with respect to parameter deviation caused by phenomena such as nonlinear magnetics, rotor temperature variation, and inaccurate measurement. This project also includes the development of an LC filter design to improve the thermal performance of the system. A dynamic model for a synchronous reluctance machine with a conducting rotor has been developed and an open-loop current regulator for high-speed operation has been designed based upon this model. The machine dynamic model is similar to an induction machine model, yet includes a magnetic saliency of the rotor. The model is then used to calculate command voltages for a desired current in an open-loop current regulator. Techniques for parameter extraction and discrete-time models for digital implementation are presented. Experimental results consisting of a 120kW discharge of a flywheel energy storage system validates the performance of the controller. The feedforward controller includes machine parameters, and the performance inherently relies on the correctness of these values. However, inductance and resistance parameters will vary due to flux saturation and temperature, respectively. Although a feedforward control scheme is simple, fast and effective, a direct influence which deteriorates control performance can be seen on the controller's output if the parameters are varied. Hence, a feedback compensation method has been investigated to handle the possible deviation of the parameters, and to improve the feedforward controller's robustness. A systematic approach to designing a feedback compensator for the feedforward controller is presented. Also, a stability analysis for the feedback-compensated system has been performed. Improved current tracking performance can be seen in the experimental results. The flux-linkage/current relationships of the machine are one of the major nonlinearities to be handled in a machine controller. A more precise modeling of nonlinear magnetics becomes essential for control purposes and for understanding the limitations imposed by them. The feedforward controller can handle more diverse operating conditions by incorporating a better model of the machine dynamics. Therefore, a more accurate model to represent the nonlinear magnetics for the feedforward controller has been developed. The performance improvement by the modified model has been shown through the experimental results. Although synchronous reluctance machines with solid rotor construction have advantages in certain high-speed applications such as flywheel energy storage systems, the solid rotor allows the flow of eddy currents, which results in heat generation. A three-phase LC filter can reduce rotor losses due to the switching harmonics. The design and control of a high-speed synchronous reluctance drive with a three-phase LC filter has been investigated. A two-phase dynamic model of the drive which incorporates the LC filter dynamics is presented. The model is used to predict rotor losses due to switching harmonics generated by the three-phase inverter using phasor analysis. A feedforward current regulator is utilized, which is modified to include the effects of the LC filter. Experimental results validate the proposed approach.