摘要
针对液压随动系统实际运行中存在理论设计模型与现场不匹配、运行中的不确定性以及外部负载扰动等问题,设计了一种基于参数估计和RBF神经网络的逼近模型非线性特性的滑模控制器。该控制器由滑模控制器和自适应模型两部分构成,其中自适应模型部分由自适应滑模参数估计器和自适应逼近液压随动系统模型非线性特性组成,自适应逼近液压随动系统由求解RBF神经网络得出。滑模控制器由李雅普诺夫函数求导得到,并且分析了该控制策略的可行性。通过理论分析,液压随动系统是全局渐近稳定的。最后,对液压随动系统进行了仿真,仿真结果验证了所提出算法的有效性,为开发高性能液压随动系统提供一些借鉴。
A sliding mode controller was designed based on parameter estimation and nonlinear characteristics of approximation model of RBF neural network for hydraulic servo systems,which have problems such as mismatches between the theoretical design model and the field and uncertainties and external load disturbances during operation.The controller consists of a sliding mode controller and an adaptive model,in which the adaptive model consists of an adaptive sliding mode parameter estimator and an adaptive approximation of the nonlinear characteristics of the hydraulic servo system model.The adaptive approximation of the hydraulic servo system was obtained by solving the RBF neural network.The sliding mode controller was derived from the Lyapunov function,and the feasibility of this control strategy was analyzed.Through theoretical analysis,the hydraulic servo system designed is globally asymptotically stable.Finally,a hydraulic servo system was simulated,and the simulation results verify the effectiveness of the proposed algorithm,which provides some reference for the development of high performance hydraulic servo system.
作者
王瑞明
马振兴
Wang Ruiming;Ma Zhenxing(College of Physics and Electrical Engineering,Kashgar University,Kashi 844000,China)
出处
《煤矿机械》
2024年第2期79-82,共4页
Coal Mine Machinery
基金
喀什大学2022年引进博士科研启动经费(022022276)。
关键词
RBF神经网络
参数估计
液压随动系统
滑模控制
RBF neural network
parameter estimation
hydraulic servo system
sliding mode control
