高倍速光碟機長程傳動馬達的模糊伺服控制及模糊系統應用於混沌系統的判別

Abstract

在本論文中，我們將模糊系統應用在光碟機長程傳動伺服控制及混沌 系統判別上。在長程傳動伺服控制方面，以NEC 六倍速光碟機中的音圈馬 達為控制對象，首先根據音圈馬達的結構，依其物理特性及運動方程式建 立其數學模型，再經系統判別求得系統轉移函數。我們提出一個比例積分 控制加上模糊控制的混合控制法，其藉模糊控制的強健性改善模糊控制的 缺點，也藉比例積分控制的優點改善模糊控制的缺點，其結果不論在電腦 模擬或以實際電路的控制上，都改善了原本單一控制器的缺點，使得光碟 機長程傳動馬達的定位時間更短。 在混沌系統別方面，我們使用模糊 系統當混沌系統的判別工具，提出一個新的動態估測模式，稱為模糊觀測 器，其可用來預測估測誤差。模糊觀測器，其函數和卡曼濾波器的卡曼增 益相同，可以改善傳統判別法在混沌系統判別上誤差大、不具動態特性的 缺點，而且更具強健性，在輸入資料消失時，仍可有良好的效果。 In this paper, the fuzzy systems is applied to the sled motor servo control of aCD-ROM and the chaotic system identification. In sled motor servo control, a voicecoil motor(VCM) of a NEC cXi CD-ROM is used as a control plant. First of all, the structure of the VCM is investigated. Second, the mathematical model is generatedaccording to its physical characteristics and motion eguations. We get the transferfunction in this model by the measurement from the VCM using the technique of systemidentification. We propose a hybrid control scheme that includes a PI control and a fuzzy control. It improves the disadvantages of the PI control by the robustnessproperty of the fuzzy control. It also overcomes the disadvantages of the fuzzy control by the stability of the steady state for PI control. Both of comuper simulation and experimentalresults show that the PI-Fuzzy control improves the disadvantages when only controlleris used. It can also get the fast response time of the CD-ROM sled motor. In chaotic system identification, we use fuzzy system as idertifiers for chaotic systems.It is provided that a new dynamic predictor model, called fuzzy observer, which is used to predict the estimation error. The fuzzy observer, whose function is the same as that the Kalman gains in Kalman filtering, overcomes the disadvantages of big error and non- dynamics in traditional identificationscheme. It is shown that the structure of the fuzzy observer is also robust to the lost of theincoming information.