本論文提出一個利用小波模糊類神經網路估測器之以凸極式反電動勢為基礎之速度估測法，結合每安培最大轉矩控制，以改善應用在變頻壓縮機驅動系統上內藏式永磁同步馬達之性能。文中首先說明內藏式永磁同步電動機及其凸極式反電動勢之特性與數學模型，並分析了以凸極式反電動勢為基礎的無感測控制及啟動策略，且同時提出適用於數位訊號處理器之新型每安培最大轉矩控制；其次提出一新型無感測技術，利用小波模糊類神經網路估測器之以凸極式反電動勢為基礎之速度估測法做為馬達控制策略，以達到快速的暫態響應及節能效益。此外小波模糊類神經網路的網路架構、線上學習法則將在本文被詳細的討論，並將以PSIM 搭配C 撰寫之DLL 檔為模擬軟體進行模擬。最後利用微芯公司所生產之數位訊號處理器實現變頻壓縮機驅動系統，並且以實驗結果驗證所提出方法之可行性。A saliency back-EMF based wavelet fuzzy neural network (WFNN) torqueobserver combining with a new-type maximum torque per ampere (MTPA)control is proposed in this thesis to improve the speed estimating performance ofa sensorless interior permanent magnet synchronous motor (IPMSM) used ininverter-fed compressor drive systems. First, the structure, characteristics andmathematical model of the IPMSM and the saliency back-EMF estimator arediscussed, and the start-up strategy based on saliency back EMF is added. Then,a new saliency back EMF based MTPA control suitable for the implementationusing digital signal processor (DSP) is introduced. Moreover, a back-EMF basedspeed estimation method using WFNN torque observer is proposed. Furthermore,the network structure and the online learning algorithms of WFNN are discussedin detail. In addition, a Microchip DSP is adopted to develop the sensorlessinverter-fed compressor drive system. Finally, some experimental results aregiven to verify the feasibility of the proposed control schemes.
