基于Volterra级数的一类非线性系统性能评估  

Performance assessment based on volterra series for a class of nonlinear systems

作　　者：王志国[1] 刘飞[1] 

机构地区：[1]江南大学自动化研究所、轻工过程先进控制教育部重点实验室,江苏无锡214122

出　　处：《计算机与应用化学》2012年第3期301-304,共4页Computers and Applied Chemistry

基　　金：国家自然科学基金项目(60974001);高等学校学科创新引智计划资助(B12018)

摘　　要：化工过程多具有非线性特征,针对用线性系统性能评估方法处理非线性系统会存在过估计的情况,研究了一类叠加线性干扰的非线性系统的性能评估问题。通过使用Volterra级数近似非线性环节,把最小方差性能评估问题转化成一类模型辨识问题,并从辨识误差中得到非线性系统的最小方差估计值。通过数值仿真,将新方法所得结果和现有线性性能评估方法进行了比较,验证了设计算法的优越性。Many chemical processes have nonlinear character, to solve the over-estimation of performance assessment for nonlinear systems with linear techniques, this paper studies the performance evaluation problems for a class of nonlinear processes with additive linear disturbance By using volterra series to approximate the nonlinear components, the minimum variance performance assessment is converted into the problems of model identification. Moreover, the minimum variance can be estimated for nonlinear systems from the model errors. Simulation results demonstrate that the proposed algorithm shows better minimum variance performance than the linear assessment methods.

关 键 词：性能评估 非线性系统 最小方差性能下限 VOLTERRA级数 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]