基于BP神经网络PID自适应控制的激振系统研究  

Research on Excitation System Based on BP Neural Network PID Adaptive Control

作　　者：肖乾 葛一帆 符远航 常运清 汪寒俊 宾浩翔 XIAO Qian;GE Yifan;FU Yuanhang;CHANG Yunqing;WANG Hanjun;BIN Haoxiang(Key Laboratory of Transport Vehicles and Equipment,Ministry of Education,East China Jiaotong University,Nanchang Jiangxi 330013,China)

机构地区：[1]华东交通大学载运工具与装备教育部重点实验室,江西南昌330013

出　　处：《机床与液压》2025年第1期52-57,共6页Machine Tool & Hydraulics

基　　金：国家自然科学基金地区科学基金项目(52065020)。

摘　　要：针对跨座式单轨车辆滚动振动试验台激振系统的位置控制精度易受参数变化和外部干扰等因素的影响,提出基于BP神经网络PID自适应的控制策略。建立激振系统数学模型,并推导出其开环传递函数。基于Simulink搭建3-5-3结构的BP神经网络PID自适应控制器,并施加阶跃干扰信号以验证系统的抗干扰能力。仿真结果表明:与传统PID和模糊PID控制器相比,BP神经网络PID自适应控制下系统达到稳态所需时间分别快52%和50%,且超调量基本为0;在应对外界干扰时,该控制器能自动调整控制参数,系统以较快速度恢复至稳态,显著增强了系统的抗干扰能力,同时展现出良好的适应性和鲁棒性。The position control accuracy of excitation system of straddle type monorail vehicle rolling vibration test bench is easily affected by parameters change and external interference.To solve the problem,a self-adaptive control strategy based on BP neural network PID was proposed.The mathematical model of the excitation system was established,and its open loop transfer function was derived.BP neural network PID adaptive controller with 3-5-3 structure was built based on Simulink,and step interference signal was added to verify the anti-interference ability of the system.The simulation results show that the time for the system to reach the steady state under the BP neural network PID adaptive control is 52%and 50%faster than the traditional PID controller and the fuzzy PID controller,respectively,and the overshoot is basically 0.When dealing with external interference,the controller can automatically adjust the control parameters,and the system can recover to the steady state at a faster speed,which significantly enhances the anti-interference ability of the system,and shows good adaptability and robustness.

关 键 词：激振系统 BP神经网络 模糊PID 学习速率 

分 类 号：TH137[机械工程—机械制造及自动化]