本論文提出一個利用比例積分微分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，以改善內藏式永磁同步馬達應用在變頻壓縮機驅動系統之估測性能。此外本論文將提出兩種無感測技術，其一為高頻方波電壓注入法結合利用比例積分微分估測器之以凸極式反電動勢為基礎之速度估測法，其二為高頻方波電壓注入法結合利用比例積分微分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。以上兩種無感測控制機制皆是使用高頻方波電壓注入法作為馬達之啟動策略，以達成弦波啟動之目的。當馬達逐漸加速至預設的轉速時，系統會切換到利用比例積分微分估測器之以凸極式反電動勢為基礎之速度估測法或利用比例積分微分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。本文將詳細的分析高頻方波電壓注入法、利用比例積分微分估測之以凸極式反電動勢為基礎之速度估測法。此外，比例積分微分類神經網路的網路架構、線上學習法則、以及收斂性分析將在本文被詳細的討論。最後將以DSP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 proportional-integral-derivative neural network (PIDNN) estimator is proposed in this study to improve the speed estimating performance of the interior permanent magnet synchronous motor (IPMSM) used in inverter-fed compressor drive systems. Two sensorless control schemes are designed for the IPMSM drive system. One is the square wave type voltage injection method combined with the conventional saliency back EMF based speed estimation method using PID estimator, and the other is the square wave type voltage injection method combined with the saliency back EMF based speed estimation method using PIDNN estimator. Both sensorless control schemes use square wave type voltage injection method as the start-up strategy to achieve sinusoidal starting. When the motor speed gradually increases to a preset speed, the sensorless drive will switch to the conventional saliency back EMF based speed estimation method using PID estimator or the saliency back EMF based speed estimation method using PIDNN estimator for medium and high speed control. The theories of the square wave type voltage injection method and the conventional saliency back EMF based speed estimation method are introduced. Moreover, the network structure, the online learning algorithms and the convergence analyses of the PIDNN are discussed in detail. Furthermore, a DSP-based control system is developed to implement the sensorless inverter-fed compressor drive system. Finally, some experimental results are given to verify the feasibility of the proposed control schemes.
