本論文研究為發展以人工智慧機器學習為基礎之高性能的同步磁阻馬達驅動系統，提出一種基於定子電阻和定子磁通估測的電流角控制器於新型最大功率因數控制。傳統的最大功率因數控制系統需藉由同步磁阻馬達的凸極比來產生電流角命令，由於凸極比需要離線的使用有限元素分析預先處理且無法自動調整，使得最大功率因數控制成本高又耗時，很難在不同的操作區域中提高性能。 有鑑於此，設計了一種基於遞迴式切比雪夫模糊類神經網路電流角控制器的智慧型最大功率因數搜尋控制於同步磁阻馬達的速度控制。為了能在不同的工作條件下線上搜索同步磁阻馬達的最佳功率因數點，設計了遞迴式切比雪夫模糊類神經電流角控制器以生成補償電流角命令。此外，採用比例積分速度控制器生成定子電流命令，以及本文採用的智慧型最大功率因數搜尋控制產生電流角指令。 本研究以32位元浮點運算數位訊號處理器TMS320F28075實現基於定子電阻和磁通估測的電流角控制器之智慧型最大功率因數搜尋控制於功率為4kW的同步磁阻馬達驅動系統。最後，通過實驗結過驗證了所提出的智慧型最大功率因數搜尋控制，能夠在不同的速度和負載的操作區間有效的線上搜尋最佳功率因數的電流角命令，並驗證其強健性和準確性。 ;To develop a high-performance synchronous reluctance motor (SynRM) drive system, a novel maximum power factor control (MPFC) using stator resistance and flux estimator is proposed. First, a traditional maximum power factor control (TMPFC) system using a saliency ratio to generate a fixed current angle command is described. Because the saliency ratio requires offline repreparation and cannot be adjusted automatically, it is difficult to improve the performance of the MPFC in different operating regions. Therefore, an intelligent maximum power factor searching control (MPFSC) using a recurrent Chebyshev fuzzy neural network (RCFNN) controller is designed for the speed control of a SynRM. In order to search the online optimal power factor (PF) points of the SynRM, the RCFNN current angle controller is designed to generate the compensated current angle command. Moreover, a proportional-integral (PI) speed controller is employed to produce the stator current magnitude command, and the proposed system is adopted to generate the current angle command. Furthermore, the proposed system is implemented in a 32-bit floating-point TMS320F28075 digital signal processor. Finally, from the experimental results, the optimal PF can be effectively searched online at different speed operating commands with varied load torque.
