基于非线性预测模型的单神经元自适应PID板形控制

Single Nerve Cell Adaptive PID Flatness Control Based on Nonlinear Prediction Model

为了解决传统PID板形控制精度低、速度慢、抗干扰能力差等问题,将BP神经网络和单神经元引入到板形的控制中,提出一种基于BP神经网络预测模型的单神经元自适应PID控制的板形控制策略。利用BP神经网络的非线性逼近能力和单神经元的自学习、自适应能力,通过两者的有机结合寻找一个最佳的P、I、D非线性组合控制律,实现对带钢板形缺陷的有效控制。仿真实验结果表明,该控制算法能很好地跟踪板形的目标设定值,提高了系统的控制精度,加快了系统的响应速度,并且具备较强的抗干扰能力。

The strategy of single-nerve-cell adaptive PID flatness control based on BP neural network prediction model is proposed in order to resolve the problem of low precision,slow speed and bad anti-interference ability in conventional PID flatness control.According to the combination of the nonlinear approach ability of BP neural network and the self-learning and adaptive ability of single nerve cell,an optimal nonlinear composite control rule of P,I and D is found to control flatness defects of strips effectively.Simulation results indicate that the control algorithm can track the target set value of flatness,increasing the control precision of the system and accelerating the response speed of the system with strong anti-interference ability.