最佳化之自調式PI控制器

An Optimal Self-Tuning PI Controller

陳彥廷(Yen-Ting Chen)

指導教授 ： 陳金聖

國立臺北科技大學/機電學院/自動化科技研究所/碩士(2009年)

摘要

比例積分控制器為現階段工業界慣用之控制器，在各種不同的控制環境下，如何調變比例與積分兩項參數用以滿足特定之所需條件(如：最大超越量、上升時間、安定時間、穩態誤差等)，一直是控制工程師所追求的一項開放性問題。本篇論文架構在定能量輸入能維持定輸出之系統下，先從能量的觀點重新詮釋比例積分控制器，進而提出一套能滿足特定條件下之自調式標準流程，並使用本文所提出之梯度二分搜尋法最佳化比例增益與積分增益。而後，分別透過兩大典型系統(溫控系統與伺服馬達速度控制系統)，對提出之調整方法，進行模擬驗證：以溫控系統為例，採用耦合效應嚴重與具長延遲時間特性之射出成型機溫控模組進行驗證；而伺服馬達速度控制系統則分別對有無電流控制迴路進行比較，模擬結果顯示，本文所提出之自調式調整法，具有高度可靠性與實現性。最末，再透過一伺服馬達速度控制實驗，證實本篇所提之調整法確實可達成一最佳化之自調式PI控制器。

關鍵字

比例積分控制器 ； 自調式控制 ； 梯度二分法 ； 射出成型機 ； 定溫控制 ； 定速度控制

並列摘要

A Proportional-Integral (PI) controller is an important instrument in industry control application. In control engineering field, to find out the optimal PI parameters by observing system response is very important, especially when system model is difficult to identify exactly. Therefore, this paper proposes a self-tuning methodology with a proposed Gradient Bisection Algorithm (GBA) to achieve finding the optimal PI parameters in fixed-input sustaining fixed-output (FIFO) system. The whole tuning tactic is constructed under desired constraints (such as overshoot, occurrence times of oscillation etc) with the minimum integral absolute error (IAE). Two kinds of typical FIFO system which are thermal system and servo motor velocity control system have simulated in this paper. Besides, a servo motor velocity control experiment has also been realized. The simulation and experiment results verify that our proposed optimal self tuning methodology is a reliable and practicable way to searching the optimal PI gains under desired indexes.

並列關鍵字

Proportional-Integral controller ； self-tuning ； Gradient Bisection Algorithm ； barrel temperature ； injection molding machine

參考文獻

[1] F. Johannaber, Injection Molding Machines: a user’s guide, New York, Hanser, 1994.

[2] H. Havlicsek, A. Alleyne, “Nonlinear control of an electrohydraulic injection molding machine via iterative adaptive learning”, IEEE/ASME Transactions on Mechatronics, Vol. 4, No3, 1999, pp.312-323.

[3] M. S. Wang, C. M. Chang, “DSP-based adaptive fuzzy velocity/pressure control of injection molding machines”, Journal of Chinese Institute of Engineer, Vol. 30, No.5, 2007, pp. 819-827.

[5] K. Ibrahim, “IMC based automatic tuning method for PID controllers in a Smith predictor configuration”, Computers and Chemical Engineering, Vol. 28, No3, 2004, pp. 281-290.

[6] R. Vilanova, “IMC based Robust PID design: Tuning guidelines and automatic tuning”, Journal of Process Control, Vol. 18, 2008, pp. 61-70.