摘要
为了更好地发挥RBF和PIDNN神经网络的优势,通过对伺服电机模型辨识和控制问题的分析,对RBF和PIDNN网络的应用效果进行了仿真实验的对比研究。结果表明,RBF神经网络结构复杂,参数难以调整,但具有最佳一致逼近能力,辨识效果优于PIDNN;PIDNN结构简单,比例元、积分元和微分元具有类似PID的控制作用,控制效果优于RBF。
To take the advantages of RBFNN and PIDNN,the simulations and contrast research of their application effect are done by analyzing model inentification and control problems of servo electromotor.The experimet results show that RBF network has complex structure and its parameters are difficult to adjust.But the RBFNN has an ability of optimal coherent approach,so its identification effect is better than the PIDNN.PIDNN has simple PID-function-bearing structure of proportion cell,integral cell and differential...
出处
《控制工程》
CSCD
2008年第S1期113-115,118,共4页
Control Engineering of China
