摘要
以机械手臂作为被控对象搭建了神经网络模型参考控制的仿真模型,利用BP神经网络模型参考控制器对被控对象进行控制仿真试验.通过逐渐增加控制器的训练样本数,逐渐改善控制器的控制效果.在无干扰和有干扰的条件下分别对系统进行控制效果测试,结果表明:BP神经网络模型参考控制器可有效提高被控系统的鲁棒性和自适应性.
A neural network model reference control simulation model is built with the robot arm as the controlled object,and the BP neural network model reference controller is used to control the simulation of the controlled object.By gradually increasing the number of training samples of the controller,the control effect of the controller is gradually improved.The control effect of the system is tested under the condition with and without interference.The results show that the BP neural network model reference controller can effectively improve the robustness and adaptability of the controlled system.
作者
聂启鹏
唐明新
杨洋
NIE Qipeng;TANG Mingxin;YANG Yang(School of Electrical and Information Engineering,Dalian Jiaotong University,Dalian116028,China)
出处
《大连交通大学学报》
CAS
2019年第4期104-107,共4页
Journal of Dalian Jiaotong University
关键词
BP神经网络
非线性
系统辨识
模型参考
自适应控制
系统仿真
BP neural network
nonlinear
system identification
model reference
adaptive control
system simulation