摘要
通过应用具有二阶线性收敛速度的拟牛顿法于多层前向网络 ,以作为非线性预测控制中的预测模型 ,结合非线性优化方法 ,实现对于一般意义非线性系统的预测控制。仿真表明文中算法大大提高了网络学习收敛速度 ,使非线性预测控制算法的实时性能有很大改观。
Since the convergence speed of the BP learning algorithm is slow for real\|time predictive control, the data stability appears to be very poor. A new method for nonlinear multi\|step predictive control based on neural networks is proposed. Based on the above neural network, GPC for linear system is extended to nonlinear multi\|step predictive control. The simulation result show the obvious improvement of the suggest method over BP method both in stableness and robustness.
出处
《浙江工业大学学报》
CAS
2000年第2期121-124,共4页
Journal of Zhejiang University of Technology
基金
国家自然科学基金资助项目!(699740 36)
关键词
预测控制
拟牛顿法
神经网络
predictive control
quasi\|Newton method
nonlinear optilization