Observer-Based Controller Design for Linear Parameters Varying Stochastic Systems

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST106-2221-E019-006

Translated Name/計畫中文名

針對線性參數時變隨機系統之觀測器與控制器設計

Project Coordinator/計畫主持人

Cheung-Chieh Ku

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Marine Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=12275779

Year

2017

Start date/計畫起

01-08-2017

Expected Completion/計畫迄

31-07-2018

Bugetid/研究經費

656千元

ResearchField/研究領域

電子電機工程

"一般而言，不確定系統常被用於描述具有建模誤差與元件變化之真實動態系統，由於線性參數時變系統可廣泛的應用於描述任何形式的不確定系統，且可完整地呈現其參數時變之特性，目前線性參數時變系統乃近年來常用於描述不確定系統的方法；此外，隨著隨機建模方式的問世，針對隨機系統的穩定性分析與控制器設計議題亦被廣泛的討論，在描述真實操作環境下，本計畫將針對線性參數時變隨機系統之穩定特性進行討論。目前討論之線性參數時變隨機系統的控制問題多數屬狀態回授控制範疇，其相關的成果皆基於系統狀態可量測的情況下進行探討，倘若面對無法量測到系統狀態的情況時，將導致目前已發展之控制器設計方法全數失效；對此控制問題，觀測器的設計問題則是一無法避免的議題；目前里阿伯諾夫方程式是最常見且強大的分析方法，但是由於正定矩陣架構而發展出之分離定理與二次計算方法皆具部分保守性，對此，本計畫預計藉由保留任何正定矩陣的元素與發展改良型投影方法，提出一新穎的觀測器與控制器設計方法，藉此可設計出觀測器與控制器並保證線性時變隨機系統的穩定，最後透過數值模擬來驗證本計畫之可行性與實用性。""Generally, Linear Parameter Varying (LPV) system can be widely applied to represent the uncertain systems. Besides, stochastic behavior is common signal causing instability in practical operating environment. Since the development of stochastic modeling approach, the stability analysis of stochastic systems is discussed and investigated. For describing the practical operating conditions, the stability issue of LPV stochastic systems is discussed in this project. The existing control problem of LPV stochastic system was generally discussed via state-feedback scheme based on measured states. For invalid measured states, the most of existing control schemes cannot be applied to design controller. Therefore, the issue of designing observer is unavoidable. For the issue, Lyapunov function is a powerful and useful tool to analyze stability of systems. According to stricture of positive definite matrix in Lyapunov function, the existing decomposition theory and two-step calculating process possess some potential conservatism in solving the observer and controller design problem. For the reason, this project focuses on developing a novel design scheme to design observer and controller through holding all elements of positive definite matrix and proposing a project lemma. By using the designed observer and controller, the stability of LPV stochastic systems can be guaranteed. Finally, some numerical simulations are proposed to demonstrate the usefulness and effectiveness of the proposed design method in this project."

Keyword(s)

觀測器設計
線性參數時變系統
隨機系統
線性矩陣不等式
里阿伯諾方程式
改良型投影方法
Observer Feedback Control
LPV
Stochastic System
LMI
Lyapunov Function
Extended Project Lemma

DSpace CRIS

Observer-Based Controller Design for Linear Parameters Varying Stochastic Systems

Details

Description