摘要
将神经网络和PID控制相结合,提出了一种神经网络整定的PID控制策略,并将其应用于交流伺服系统的控制。利用一个两层神经网络在线自适应调整PID控制器的参数,从而使系统的静态和动态性能指标较为理想。径向基函数神经网络用来辨识交流伺服系统的Jacobian信息,其学习算法采用正交最小二乘算法,首先得到径向基函数神经网络的结构,然后用BP算法对该网络的权值进行训练使它逼近给定的函数。实验结果表明,该交流伺服系统具有响应速度快、稳态精度高和鲁棒性强等特点。
A neural network and PID control are integrated to control the AC servo system, which is consisted of a neural network identifier and a neural network controller. The neural network controller is used to adjust the parameters of PID control on-line, accordingly it can make static and dynamic performance index comparatively ideal. The neural network identifier adopt radial basis function neural network （RBFNN）, and orthogonal least squares are adopted as the learning algorithm of the network. The results of experiments show that AC servo system based on proposed control has quick response speed, high steady accuracy and good robustness.
出处
《微计算机信息》
北大核心
2008年第19期313-314,306,共3页
Control & Automation
