论文部分内容阅读
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。
In order to solve the problems of low precision, slow speed and poor anti-jamming performance of traditional PID control, BP neural network and single neuron are introduced into the control of flatness. A single neuron self-predicting model based on BP neural network is proposed Shape Control Strategy Adapted to PID Control. By using the nonlinear approximation capability of BP neural network and single neuron self-learning and self-adaptive ability, an optimal combined control law of P, I, D nonlinearity is found through the organic combination of the two. Effective control. The simulation results show that the proposed algorithm can track the shape of the target well, improve the control accuracy of the system, speed up the response of the system, and have strong anti-interference ability.