以DSP為基礎的交流感應馬達適應解耦控制之研究

Abstract

對交流感應馬達磁場導向向量控制而言，轉子時間常數為一相當重要之參數。本文針對三種轉子時間常數的量測方式進行理論的分析和實驗的驗証；第一種方法建立在實驗系統下，利用回感應馬達之線電壓，再利用人為的判斷或數值運算軟體輔助（如MATLAB）而可得到轉子時間常數值。在不增加額外的感測器，建立在一般伺服驅動器下；第二種方法在驅動器起動時或利用模式的選擇，根據給予的扭矩命令和所得速度響應關係而判斷調整得適當之時間常數。最後一種方法，在驅動器內增加電壓回授的電路，利用估測馬達扭矩的方式，可在馬達操作時，補償其轉子時間常數因溫度等環境因素所造成的變化。 此外，本論文建立一個以單晶片數位訊號處理器(DSP)為基礎之控制器，配合智悲型IGBT為基礎之PWM換流器而成為一體積小、富彈性之全數位式伺服驅動器。利用所量測得參數，控制器完成了軟體化之電流控制，間接式轉子磁場導向電流解耦控制以及速度和位置伺服控制。而換流器亦完成了動態煞車(Dynamic Braking)、柔性起動(Soft Star)以及一般驅動器所需要的各種保護電路。系統採用德州儀器公司之TMS320C50及TMS320C14單晶片。由於雙單晶片的應用，使複雜的運算得以實現，而未來更有機會應用先進之控制理論。

For the field orientation control of an induction motor, the rotor time constant is a very important parameter. Theoretical analysis and experimental verification of three rotor time constant measuring methods are presented in this thesis. An experimental system which a line voltage feedback loop is included; the first method derives the rotor time constant by human judgment or assistance of mathematical package software (such as MATLAB) On teh basis of a general ac drive, the second method doesn't require additional sensors. According to the torque command and the speed performance during the initial state of an ac drive or using a mode selection, the rotor time constant is tuned and identified. With a phase voltage feedback, the last method estimates the torque of an induction motor. This method compensates the variation of rotor time constant which is caused by the environment factor such as temperature. Furthermore, this thesis develops a DSP based controller. An intelligent IGBT based PWM inverter accompanies with the controller consist of an integrated and flexible fully digital servo drive. Apply the measured parameter to the drive, the DSP controller achieves: software-based current control, indirect rotor flux orientation current decoupling control, speed and position servo control. The dynamic braking, soft star and protect circuits are all included in the PWM inverter. TMS320C14 and TMS320C50 DSPs which are generated by the TI company are used in the system. Because the use of dual DSPs, the complicated computattion can be achieved and there will be more opportunities for advaanced control theories to be applied to ac servo control in the future.