一种基于BP神经网络算法PID控制器的研究与仿真

文中将BP神经网络的原理应用于参数辨识过程,结合传统的PID控制算法,形成一种改进型BP神经网络PID控制算法。该算法利用BP神经网络建立系统参数模型,能够跟踪被控对象的变化,取得较高的辨识精度。针对BP神经网络对权系初始值敏感的缺点,优化BP神经网络的初始权系数。通过BP算法修正BP网络自身权系数,实现PID参数的在线调整。仿真结果显示了该算法收敛速度快、精度高、鲁棒性强、稳定性好,表明了该算法的可行性与有效性。

Research and simulation of PID controller based on BP neural network algorithm

The technology of BP neural network will be used in the process of parameter identification,combining classical PID control algorithm,forms an improved PID control algorithm based on BP neural network.The algorithm uses the BP neural network to build system parameter model,is able to track the changes of the controlled object,achieves a higher recognition accuracy.For the shortcomings that the BP neural network is sensitive to the initial value of the right,optimize the initial weights coefficient of BP neural network.It amends the BP neural network weight coefficients by the BP algorithm,realizes adjustment of the PID parameters online.Simulation results show that these formulae has fast convergence 、high precision,and robustness.Simulation results shows the feasibility and effectiveness of the algorithm.