摘要
提出了一种基于径向基函数(RBF)神经网络的非线性模型预测控制系统,利用RBF神经网络的非线性拟合性,构建一个神经网络预测器(NNP)来预测模型未来时刻的输出值.然后利用神经网络控制器(NNC)实现基于模型的预测控制.仿真结果表明此方法具有较好的控制效果,并且在有扰动和模型失配的情况下,表现了良好的鲁棒性.
In this paper, a RBF neural network based nonlinear model predictive control is proposed, in which the RBF neural network is used to form a NNP (neural network prediction) to predicate the model output in the future. The NNC (neural network control) is used to realize the predictive control. An illustrated simulation example shows that this method is of control effects, and well robustness under the situation of disturbance and model mismatching.
出处
《浙江工业大学学报》
CAS
2007年第2期123-126,共4页
Journal of Zhejiang University of Technology
基金
浙江省自然科学基金资助(Y105397)
关键词
非线性系统
神经网络
预测控制
nonlinear system
neural network
predictive control