基于BP神经网络的PID控制方法的研究

本文提出了一种基于BP神经网络的PID控制方法,充分利用BP神经网络算法逼近任意连续有界非线性函数的能力,这种PID控制方法能学习和适应严重不确定系统的动态特性。文中采用三层前向网络,动态BP算法,达到了在线实时控制的目的,显示了BP神经网络的PID控制方法很强的鲁棒性,同时也显示了神经网络在解决高度非线性和严重不确定系统方面的潜能。计算机仿真结果表明,基于BP神经网络的PID控制较常规的PID控制具有更好的鲁棒性和自适应性,能取得良好的控制效果。

Research On FID Control Based On BP Neural Networks

A PID control based on BP neural networks, is introduced in this paper, which is used to optimize and adjust the dynamic performance of seriously uncertain system by exploiting the nonlinear mapping capability of neural networks. The dynamic BP algo-rithms of three-layer networks realizes the online real-time control, which displays the robustness of the PID control, and the capa-bility of BP neural networks to deal with nonlinear and uncertain system. The result of computer simulation shows that the PID con-trol based on BP neural networks owns better robustness and self-adaptation than the general one, the great effect of control can be guaranteed.