適應性向量控制系統之灰色小腦模型控制器設計

Design of Grey-Cerebellar Model Articulation Controllers for Adaptive Vector Control System

蔡明傑(Ming-Jay Tsai)

指導教授 ： 王順源

國立臺北科技大學/機電學院/機電整合研究所/碩士(2006年)

https://doi.org/10.6841/NTUT.2006.00230

摘要

本論文利用內涵型灰色預測控制器搭配小腦模型控制器整合成一新形態之控制架構，稱之為灰色小腦模型控制器，並將之植入感應電動機向量控制系統中，以改善系統響應。由於小腦模型控制器具有非線性之學習能力、結構簡單和快速學習等特性，因此灰色預測控制器可預測系統未來響應之趨勢與作事前補償之功能，再搭配灰色動態步距調整機制即可有效改善系統動態響應。本研究將投影演算法導入小腦模型控制器學習架構，達成線上學習與即時控制之效果，並藉由權重可信度概念加快其誤差收斂速度。最後，再搭配灰預測控制器提供適量的誤差預測量以補償系統響應，可達到改善動態響應與提昇系統強健性之目的。
在向量控制架構上，本研究採用直接轉子磁場導向控制架構，所以磁通與轉速估測器設計亦是研究中之一大重點。本研究根據適應性控制法則設計虛擬降階型磁通估測器，並利用小腦模型PI控制器作轉速與轉子電阻估測，以完成無轉速量測器控制與解決轉子電阻值漂移造成磁通估測不準確之問題。
經過實驗結果印證，灰色小腦模型控制器應用於感應馬達向量控制系統，在重載(8Nm)情況下啟動且操作轉速範圍分佈在36rpm~2000rpm間之動態響應，明顯優於傳統小腦模型控制器，且操作於馬達參數變動之環境下仍具有一定程度之強健性。

關鍵字

小腦模型控制器 ； 灰色預測控制器 ； 適應性磁通估測器 ； 感應電動機 ； 向量控制

並列摘要

In this thesis, a novel control architecture, Grey-CMAPIC, which is based on the Cerebellar Model Articulation PI Controller (CMAPIC) and the intensity grey decision prediction controller (IGDPC), is proposed. Also, the Grey-CMAPIC is employed to improve the performance of vector control system of induction motor.
In general, the merits of the CMAPIC include rapid learning convergence, simple structure, on-line training, and non-linear learning ability. However, in order to enhance the performance of conventional CMAPIC, this thesis proposes the Grey-CMAPIC which is integrated with the CMAPIC and the grey-prediction compensator. The proposed Grey-CMAPIC experimentally not only improves the transient response but also yields better steady-state performance, especially on the load disturbance tolerance, parameter variation tolerance, and accuracy.
The structure of the vector control system in the thesis adopts the Direct Rotor Flux Orientation Control (DRFOC), and thus the design of flux and speed estimator is one of the main focuses. The adaptive control method is used to design the pseudo-reduced-order flux observer. Moreover, for implementing the sensor-less vector control, this study uses the CMAC to estimate speed and rotor resistance, and consequently the problem of inaccurate flux estimation due to rotor resistance variation is solved.
According to the experimental results, under the speed command from 2% to 100% of rated speed, the load of 8Nm, and the operation range between 36rpm to 2000 rpm, it is seen that the dynamic behavior of induction motor vector control system equipped with the novel Grey-CMAPIC performs well as comparing to the conventional CMAPIC, and the robustness remains in the presence of parameter variations

並列關鍵字

Intensity Grey Prediction Controller ； Cerebellar Model Articulation Controller ； Adaptive Flux Observer ； Induction Motor ； Vector Control

參考文獻

[23] 李谷元，高效能向量控制系統之小腦模型控制器設計，碩士論文，國立台北科技大學電機工程研究所，2005。

[8] 黃健龍，具SVC補償感應發電機系統之灰決策預測控制器設計，碩士論文，國立台北科技大學電機工程研究所，台北，2005。

[1] J. S. Albus, "A new approach to manipulator control：the cerebellar model articulation controller (CMAC)," Journal of Dynamic Systems, Measurement and Control, vol. 97, no. 1, 1974, pp.220-227.

[2] Y. Wong and A. Slideris, "Learning convergence in the cerebellarm model articulation controller," IEEE Transaction on Neural Networks, vol. 3, no. 1, 1992, pp. 115-122.

[3] S. F. Su, T. Tao, and T. H. Hung, "Credit assigned CMAC and its application to online learning robust controllers," IEEE Transactions on Systems, Man, and Cybernetics, vol. 33, no. 2, 2003, pp. 202-213.