內藏式永磁同步電機之可變繞組在線切換研究

Abstract

本文提出一種結合遞迴最小平方算法(Recursive Least Square, RLS)的參數估測技術和轉矩估測方法(Torque Estimation)，應用於可變繞組設計之永磁同步電機的在線繞組切換控制，所提出的方法能夠在電機運行時，有效並快速估測電機參數(包括定子電阻、交直軸電感以及轉子磁通鏈)，並根據電機參數與繞組匝數之間的關係，分別估測出主繞組和高速繞組的轉矩響應。透過轉矩估測結果，可以計算出繞組最佳的切換條件，實現平滑的繞組在線切換控制，並同時達到最大轉矩輸出。此外，以磁場導向控制(Field Oriented Control, FOC)為基礎架構，控制策略結合了定轉矩區的每安培最大轉矩控制(Maximum Torque Per Ampere, MTPA)和定功率區的弱磁控制(Field-Weakening, FW)，以完成整個閉迴路控制。最後，在模擬軟體MATLAB/Simulink中建立可變繞組設計之電機模型、控制策略和繞組切換演算法，以驗證本文所提出的繞組在線切換控制策略的可行性與強健性。

This thesis proposes a novel approach that combines a parameter estimation technique using the recursive least square (RLS) algorithm with Torque Estimation for online winding switch control in variable-winding interior permanent magnet synchronous machine (PMSM). The method introduced in this study can effectively and rapidly estimate motor parameters, including stator resistance, d-q axis inductance, and rotor flux linkage, during motor operation. Based on the relationship between motor parameters and number of winding turns, the torque responses of the primary winding and high-speed winding are estimated separately. Through the results of the torque estimation, optimal switching condition for the windings can be calculated. This facilitates a smooth online winding switch control and simultaneously achieves maximum torque output. Moreover, the control strategy, grounded in the architecture of field oriented control (FOC), maximum torque per ampere (MTPA) is applied in the constant torque region and field-weakening (FW) in the constant power region to finalize the complete closed-loop control. Finally, a variable-winding motor model, control strategy, and winding switching algorithm are established in MATLAB/Simulink. This setup serves to validate the feasibility and robustness of the online winding switch control strategy proposed in this thesis.