增量式PID神经网络控制器和仿真

摘要

According to the characteristics of the controllable three-levels damping vibration damper, the PID neural network controller based on the BP neural network of M ×Q ×3 structure is designed by using the neural network theory. According to the operation state of the controlled system, the PID controller parameters can be output through the neural network by the self-learning and the adjustment of the weighting coefficients of the neural network. A better control effect can be gained in this way. The effectiveness of this control algorithm is verified by the simulation analysis of three typical operating conditions for the passive suspension and the controlled suspension system.%针对三档阻尼可控减振器的特点,利用神经网络理论,设计一种基于MxQ×3结构BP神经网络的PID神经网络控制器.该控制器可根据被控系统的运行状态,通过神经网络的自学习、加权系数调整,使神经网络输出PID控制器参数,从而达到较好的控制效果.对被动悬架与可控悬架系统进行三种典型工况的仿真分析,验证本文所提控制算法的有效性.