摘要
实际工业过程往往呈现强非线性和时滞特性,在复杂工况下如何建立过程的高精度模型并对其进行有效控制是研究重点.本文研究了一种基于相关分析的非线性系统Wiener模型辨识.Wiener非线性模型由动态线性子系统和静态非线性子系统串联组成,利用时滞状态空间模型和神经模糊模型分别建立动态线性子系统和静态非线性子系统,并利用设计的组合信号实现Wiener非线性模型分离辨识.首先,利用后移算子的性质,将时滞状态空间模型转化为传递函数模型,在高斯信号作用下利用相关分析方法辨识动态线性子系统的参数,解决了Wiener模型中间变量不可测问题.其次,为了改善辨识模型的精度和收敛速度,推导了带遗忘因子的递推随机梯度方法,得到静态非线性子系统的参数估计.将提出的非线性Wiener模型辨识方法运用于连续搅拌反应釜,实验仿真结果表明,本文提出的Wiener模型辨识方法能够有效辨识连续搅拌反应釜系统,并取得较好的控制效果.
The actual industrial process often presents strong nonlinearity and strong coupling in operation.How to control the system quickly and accurately under complex working conditions is a current research focus.In this paper,an identification of Wiener model for nonlinear systems based on correlation analysis is studied.The Wiener nonlinear model consists of a dynamic linear subsystem and a static nonlinear subsystem in series,and the dynamic linear subsystem and the static nonlinear subsystem are established by using time-delay state space model and neural fuzzy model respectively,and the Wiener nonlinear model is separated and identified by using the designed combined signal.Firstly,the time-delay state space model is transformed into a transfer function model by using the properties of the backward operator,and the parameters of the dynamic linear subsystem are identified by correlation analysis method under the action of Gaussian signal,which solves the unmeasurable problem of intermediate variables in Wiener model.Secondly,in order to improve the accuracy and convergence speed of the identification model,a recursive stochastic gradient method with forgetting factor is derived to estimate the parameters of the static nonlinear subsystem.The proposed nonlinear Wiener model identification method is applied to continuous stirred tank reactor.The simulation results show that the proposed Wiener model identification method can effectively identify the continuous stirred tank reactor system and achieve better control effect.
作者
丁振宇
李峰
徐亮亮
DING Zhen-yu;LI Feng;XU Liang-liang(College of Electrical and Information Engineering,Jiangsu University of Technology,Changzhou 213001,China;Changzhou Railway Higher Vocational and Technical School,Changzhou 213011,China)
出处
《陕西科技大学学报》
北大核心
2024年第3期197-202,208,共7页
Journal of Shaanxi University of Science & Technology
基金
国家自然科学基金项目(62003151)
江苏省常州市科技计划项目(CJ20220065,CM20223015)
江苏省研究生科研与实践创新项目(SJCX22_1477)。