Date of Award

11-8-2023

Publication Type

Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

Keywords

Electric Vehicle;Magnetic Saturation;Parameter Estimation;Permanent Magnet Synchronous Motors;PMSM;VSI Nonlinearity

Supervisor

Narayan Kar

Rights

info:eu-repo/semantics/embargoedAccess

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

PMSMs are widely used in various industrial applications, including EVs, due to their high power density, high efficiency, and precise control characteristics. Accurate parameter identification is crucial for optimal controller design, fault detection, performance optimization, and condition monitoring in PMSMs. This thesis investigates advanced parameter estimation techniques for PMSMs to enhance the accuracy and reliability of the estimation process. A novel noninvasive multiparameter estimation approach is proposed to estimate the electrical parameters of PMSM, including the stator winding resistance, PM flux linkage, and dq-axis inductances. The method incorporates compensation techniques to address VSI nonlinearity and the magnetic saturation effect is modeled through polynomial functions, and the effect of temperature on the stator winding resistance is compensated. A method for decoupling the parameters in the PMSM dq-axis voltage equations is proposed to reduce both the computational burden and cross-coupling effects between parameters. Three objective functions are defined and implemented using the CCPSO algorithm which improves the estimation accuracy and efficiency. Experimental validation is carried out on a 4.25-kW interior PMSM under varying speed and load conditions, and error analysis is performed that compares the estimated and measured parameters, demonstrating high accuracy. A comparative analysis is also conducted with the least squares algorithm, which indicates better estimation results for CCPSO. A parameter-independent model for MTPA control of PMSMs is proposed, eliminating the need for online parameter estimation, and increasing computational efficiency. A gradient descent algorithm-based current angle search approach is also proposed and validated through simulations. This thesis includes parameter sensitivity analysis to enhance parameter reliability. The Taguchi method, a reliable and efficient experimental design technique, is implemented to investigate the sensitivity of PMSM parameters to key factors like torque, speed, and temperature at different levels.

Available for download on Monday, November 04, 2024

Share

COinS