基于扩展线性矩阵不等式的鲁棒模型预测控制  

Robust Model Predictive Control Method Based on Enhanced LMIs

作　　者：李乐尧[1] 王新民[1] 尹海韬[1] 郑翌[1] 

机构地区：[1]西北工业大学自动化学院,西安710129

出　　处：《科学技术与工程》2014年第10期99-103,共5页Science Technology and Engineering

摘　　要：提出了一种基于扩展线性矩阵不等式的鲁棒模型预测控制器的综合方法。通过引入松弛矩阵变量,实现了Lyapunov矩阵变量与系统矩阵的解耦,提供了额外的设计自由度,同时也降低了Lyapunov变量的保守性;通过将提出的方法应用于某型飞机综合飞行/推进控制系统的纵向控制律设计,并与传统的鲁棒模型预测控制方法进行对比仿真,检验了所提出方法的有效性和优越性。A robust model predictive controller synthesis method based on enhanced Linear Matrix Inequalities is proposed in this paper. A socalled slack variable is adopted in order to render the nonconvex optimization prob lem to a set of convex Linear Matrix Inequalities, to provide extra freedomofdegree in controller design, as well as to reduce conservatism due to quadratic Lyapunov candidate function. The proposed control method is applied to an Integrated Flight/Propulsion control law design of an aircraft, simulation results show that comparing to traditional robust model predictive control, one could obtain better robust performance via proposed controller design method.

关 键 词：鲁棒模型预测控制 扩展线性矩阵不等式 综合飞行 推进系统 

分 类 号：V231.9[航空宇航科学与技术—航空宇航推进理论与工程]