带神经网络补偿的Smith预估极点配置自校正控制  

Pole Placement Self Tuning Control with Neural Network Compensation and Smith Predictor

作　　者：靳其兵[1] 李鸿儒[1] 刘子静 顾树生[1] 

机构地区：[1]东北大学信息科学与工程学院

出　　处：《东北大学学报（自然科学版）》1999年第2期133-136,共4页Journal of Northeastern University(Natural Science)

基　　金：辽宁省自然科学基金

摘　　要：用一个常规线性模型对被控对象进行辨识，线性模型辨识的余差用一个神经网络进行补偿，线性模型和神经网络共同构成对象的辨识模型·基于这一模型对大滞后对象提出了带神经网络补偿的Ｓｍｉｔｈ预估极点配置自校正控制和带神经网络补偿的Ｓｍｉｔｈ预估极点配置自校正ＰＩＤ控制·这些方法适用于非线性对象，具有较强的鲁棒性和较好的控制精度·The plant was identified by using a normal linear model,and the deviation identified by the linear model was compensated via a neural network. The identification model consists a linear model and a neural network. Based on the model, a pole placement self tuning control with neural network compensation and Smith predictor is proposed for a plant with large time delays. Furthermore, pole placement self tuning PID control with neural network compensation and Smith predictor is also established. Both of the models are suitable for nonlinear plant, and have stronger robustness and better control precision.

关 键 词：神经网络 极点配置 Smith预估 鲁棒性 PDI控制 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置]