基于二阶滑模观测器的非线性动态系统故障估计  

Fault Estimation of Nonlinear Dynamical Systems Based on Second Order Sliding Mode Observer

作　　者：胡正高[1,2] 赵国荣[1] 马合宝 周大旺[1] 徐壮[2] 

机构地区：[1]海军航空工程学院控制工程系 [2]海军蚌埠士官学校 [3]91498部队

出　　处：《上海交通大学学报》2015年第11期1618-1624,共7页Journal of Shanghai Jiaotong University

基　　金：国家自然科学基金资助项目(61473306)

摘　　要：研究了一类非线性动态系统的执行器故障估计问题.采用Takagi-Sugeno(T-S)模糊模型对非线性动态系统建模,克服了以往研究需对非线性动态系统作Lipschitz条件假设的局限性.为避免传统滑模观测器在实现故障估计时的抖振问题,设计二阶滑模观测器实现了故障估计.为方便观测器设计,通过坐标变换分离出系统的可测状态;采用Lyapunov泛函与线性矩阵不等式设计观测器,并证明了观测误差动态系统的稳定性,进而得到了执行器故障的渐近估计.所提方法只需观测器与系统的输出信息,实现了故障的在线估计.通过一个液压传动系统的仿真分析验证了所提方法的有效性.Actuator fault estimation of a class of nonlinear dynamical systems was considered.A T-S fuzzy model was employed to model the nonlinear dynamical systems,overcoming the shortcoming in the previous studies which need the Lipschitz assumption for the nonlinear dynamical systems.In order to avoid the chattering effect caused by the traditional sliding mode observer in fault estimation,a second order sliding mode observer was designed to fulfill fault estimation.Measurable states of system were separated by coordinate transformation to assist observer designing.Observers were constructed by the Lyapunov function and linear matrix inequality.Besides,the stability of observer error dynamical systems was proved.Furthermore,actuator fault estimation was obtained.Fault estimation can be achieved on line by the output information of observer and system by the proposed method.Simulation analysis of a nonlinear hydrostatic transmission system was conducted to illustrate the efficiency of the proposed method.

关 键 词：故障估计 非线性动态系统 TAKAGI-SUGENO模糊模型 二阶滑模观测器 LYAPUNOV泛函 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]