非线性Hammerstein-Wiener模型辨识预测控制  

Nonlinear model predictive control based on identification of Hammerstein-Wiener model

作　　者：满红[1,2] 邵诚[1] 王国峰[3] 

机构地区：[1]大连理工大学先进控制技术研究所,辽宁大连116024 [2]大连交通大学信息工程学院,辽宁大连116052 [3]大连海事大学信息科学技术学院,辽宁大连116026

出　　处：《大连海事大学学报》2011年第2期101-105,共5页Journal of Dalian Maritime University

基　　金：国家重点基础研究发展计划(973)基金资助项目(2007CB714006);国家科技部中小企业创新基金(09C26222123243)

摘　　要：为解决工业过程控制领域中非线性系统的模型辨识与预测控制问题,提出一种基于BP神经网络模型的预测控制策略,采用一种分段最小二乘支持向量机辨识Hammerstein-Wiener模型系数的方法建立非线性预测控制器.利用BP神经网络训练预测控制输入序列和拟牛顿算法求解非线性预测控制律,从而实现了一种基于支持向量机Hammerstein-Wiener辨识模型的非线性预测控制算法.仿真实验验证了该算法的有效性和可行性.To solve the problems about nonlinear systems of model identification and predictive control in the industrial process control,a strategy of model predictive control based on BP neural network was presented which adopted the method to identify Hammerstein-Wiener model coefficient based on least squares support vector machines and then to build nonlinear predictive controller.A nonlinear predictive control algorithm based on least support vector machines Hammerstein-Wiener model was realized by using BP neural network to train predictive input sequences and using Quasi-Newton method to solve nonlinear predictive control rule.Simulation result illustrates that the proposed approach is effective and feasible.

关 键 词：HAMMERSTEIN-WIENER模型 最小二乘支持向量机 BP神经网络 模型预测控制 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]