線型磁浮軌道系統之機電整合設計與精密運動控制

Mechatronics Design and Precise Motion Control of Linear Magnetic-Levitation Rail System

李政道(Jeng-Dao Lee)

指導教授 ： 魏榮宗

元智大學/電機通訊學院/電機工程學系/博士(2007年)

摘要

一般而言，線型磁浮軌道系統主要可區分為磁力懸浮與推進系統兩部分，由於電磁鐵所產生的吸引力具有非線性特性且懸浮平台的姿態控制問題亦有待克服，是故磁力懸浮系統為目前極為熱門研究主題之一；另一方面，移動平台懸浮時所產生的正向力往往會對推進系統造成相當程度的干擾，導致於線型磁浮軌道系統耦合動態呈現高度非線性且時變特性。有鑑於此，本論文研製線型磁浮軌道系統及分析其整體耦合動態行為(第二章)，並針對一混合式磁浮系統比較各式不需系統參數控制架構之性能(第三章)，再進而發展以步階迴歸設計為基礎之控制架構與智慧型強健模糊類神經網路應用於磁力懸浮系統之穩定平衡控制(第四及五章)，最後並設計適應性模糊類神經網路控制架構(第六章)同時應用於磁力懸浮及推進系統，期許達成線型磁浮軌道系統穩定平衡及循軌控制之目的。首先，根據拉格蘭茲理論，分析包含電磁鐵動態、懸浮平台動態及推進系統動態之整體線型磁浮軌道系統耦合模型。再者，為解決磁浮系統之高度非線性問題，分別發展比例-積分-微分、模糊類神經網路以及適應控制器應用於一混合式磁浮系統，以達到不需系統參數即可成功達成單維度磁浮定位控制目的。磁浮平台控制方面，引進強健參數估測及動態曲面控制技術經由步階迴歸系統化設計磁力懸浮系統之穩定平衡控制策略，改善傳統步階迴歸控制對於系統不確定量之需求、減輕控制力抖動現象及克服因高階微分所可能引起致動器飽合之問題；接著更進一步結合步階迴歸控制系統與模糊類神經網路發展一強健型模糊類神經網路控制系統於磁浮平台之穩定平衡控制。最後，為降低控制系統對於系統參數的需求性、縮短控制策略執行時間以及省略習用輔助控制器以簡化整體控制架構，融合滑動模式控制及模糊類神經網路設計線型磁浮軌道系統之穩定平衡及循軌控制策略。本論文所發展控制法則皆由里亞普諾穩定分析或終值均勻有界定理的推導中獲得，因此即使線型磁浮軌道系統存在不確定量時，整個閉迴路控制系統依然可保證漸進穩定或追蹤誤差收斂至一定範圍內之特性，並以數值模擬及實作結果來驗證本論文所提出控制系統之有效性與強健性。

關鍵字

線型磁浮軌道系統 ； 磁力懸浮系統 ； 推進系統 ； 步階迴歸控制 ； 模糊類神經網路 ； 適應控制 ； 動態曲面控制 ； 滑動模式控制 ； 里亞普諾穩定理論 ； 終值均勻有界定理

並列摘要

In general, a linear magnetic-levitation (maglev) rail system contains two sub-systems including maglev and propulsion systems. The subject of inherently unstable electromagnetic force and nonlinear attitude control of the moving platform in the maglev system is one of interesting research topics at present. On the other hand, the corresponding control performance of the propulsion system is influenced easily by the normal force produced by the maglev system so that the coupled dynamic model of the linear maglev rail system is highly nonlinear and time varying. Therefore, this doctoral dissertation is mainly to design a linear maglev rail system and to analyze the corresponding entire coupled behavior including electromagnetic dynamic, maglev platform dynamic, and propulsion system dynamic. To cope with highly nonlinear and unstable dynamics, three model-free control strategies including a proportional-integral-differential (PID) scheme, a fuzzy-neural-network (FNN) control and an adaptive framework are adopted for the precise positioning of a hybrid maglev system. Moreover, it introduces the techniques of robust parameter estimation and dynamic surface control into backstepping systematic design for the stable balancing control of the maglev system to relax the requirement of system uncertainties, to eliminate the chattering phenomena, and to deal with the explosion terms caused by repeated differentiations in backstepping design procedure and the possible problem of actuator saturations in conventional backstepping control. In addition, a robust fuzzy-neural-network control scheme for the levitated stable balancing control of the linear maglev rail system with nonnegative inputs are designed to mimic the backstepping control law, and to ensure the stability of the controlled system without the requirement of auxiliary compensated controllers despite the existence of uncertainties. In order to alleviate the sensitivity of a control scheme with respect to system parameters, to reduce the execution time of a control strategy, and to leave out traditional auxiliary controllers for simplifying conventional complex control frameworks, it embeds the sliding-mode control into a fuzzy neural network to design a suitable decoupled control methodology for the stable balancing and tracking control of the linear maglev rail system. All the control laws for the linear maglev rail system are derived in sense of Lyapunov stability analysis or uniformly ultimately bounded (UUB) theorem so that the asymptotical stability of the closed-loop control system can be guaranteed or the tracking error state vector will be attracted into a small stable region near origin, even when the uncertainties occur in the maglev or propulsion system. The effectiveness and robustness of the proposed control strategies in this doctoral dissertation are verified by numerical simulations and experimental results.

並列關鍵字

Linear magnetic-levitation (maglev) rail system ； Maglev system ； Propulsion system ； Backstepping control ； Fuzzy neural network ； Adaptive control ； Dynamic surface control ； Sliding-mode control ； Lyapunov stability theorem ； Uniformly ultimately bounded

參考文獻

[30] C. T. Lin, C. F. Juang, and C. P. Li, “Temperature control with a neural fuzzy inference network,” IEEE Trans. Syst., Man, Cybern., Part C, vol. 29, no. 3, pp. 440–451, Aug. 1999.

[1] I. Boldea and S. A. Nasar, “Linear electric actuators and generators,” IEEE Trans. Energy Conversion, vol. 14, no. 3, pp. 712–717, Sept. 1999.

[2] P. Holmer, “Faster than a speeding bullet train,” IEEE Spectrum, vol. 40, no. 8, pp. 30–34, Aug. 2003.

[3] F. J. Lin, R. J. Wai, W. D. Chou, and S. P. Hsu, “Adaptive backstepping control using recurrent neural network for linear induction motor drive,” IEEE Trans. Ind. Electron., vol. 49, no. 1, pp. 134–146, Feb. 2002.

[4] R. J. Wai and W. K. Liu, “Nonlinear control for linear induction motor servo drive,” IEEE Trans. Ind. Electron., vol. 50, no. 5, pp. 920–935, Oct. 2003.