摘要
提出了一种使用神经元网络的非线性系统间接自校正预测控制器．这一控制方法是基于两个神经元网络，一个用于动态过程的建模，另一个用于获取控制信号．控制器的优化目标基于过程输出的向前多步预测．两个网络的在线学习均采用了非线性最小二乘法．
The use of neural networks for indirect self tuning control of nonlinear plants has been proposed. The control is based upon two neural networks. One is used for modelling of dynamic plant, the other is used for obtaining control signals. The optimizing criteria for the control is based upon the multi step ahead predictions of the plant output. Nonlinear least square is utilized in the learning of the two neural networks.
出处
《南开大学学报（自然科学版）》
CAS
CSCD
北大核心
1999年第3期53-57,154,共6页
Acta Scientiarum Naturalium Universitatis Nankaiensis
基金
国家863CIMS应用基础研究基金
天津市自然科学基金
关键词
非线性系统
自校正控制
预测控制
神经网络
neural network
nonlinear system
self tuning control
predictive control
least square
