摘要
助力特性曲线是反映转向轻便性和路感强度的重要特性,对于目标车型进行电动助力转向系统的开发,需要设计符合目标车型状态变化的助力特性曲线,通过确定车速感应系数,设计出了目标车型的曲线型助力特性曲线,并采用基于BP神经网络的PID自适应控制,通过神经网络的自身学习和加权系数调整,实现参数自整定,避免了传统PID参数整定的繁琐。最后针对设计的曲线型助力特性曲线和BP神经网络的控制策略,进行仿真试验。结果表明:BP神经网络控制策略能够实现对曲线型助力特性曲线的目标电流进行实时跟随,而且比传统PID控制策略具有较高的稳定性,提高了系统的鲁棒性,对汽车电动助力转向控制器的开发具有重要意义。
Assist characteristic curve is an important characteristic reflecting the lightness of steering and the strength of road feeling intensity.In order to develop the electric power steering system for the target vehicle,it is necessary to design the assist characteristic curve which conforms to the state change of the target vehicle.By determining the speed correlation coefficient,the curve assist characteristic curve of target vehicle type is designed,and the base is adopted. In the PID adaptive control of BP neural network,the parameter self-tuning is realized by the self-learning of BP neural network and the adjustment of weighting coefficient,which avoids the tedious tuning of traditional PID parameters.Finally,the simulation experiment is carried out for the designed curved assist characteristic curve and the control strategy of BP neural network.The results show that the BP neural network control strategy can realize the real-time tracking of the target current of the curved assist characteristic curve,and compared with the traditional PID control strategy,it has higher stability,and improves the robustness of the system.It is of great significance to the development of electric power steering controller for automobiles.
作者
商显赫
林幕义
陈勇
马彬
SHANG Xian-he;LIN Mu-yi;CHEN Yong;MA Bin(School of Mechanical Electrical Engineering,Beijing Xinxi Keji University,Beijing 100192,China;Beijing Electric Vehicle Collaborative Innovation Center,Beijing 100192,China)
出处
《机械设计与制造》
北大核心
2022年第3期36-40,共5页
Machinery Design & Manufacture
基金
国家自然科学基金青年项目(51608040)
促进高校内涵发展研究生科技创新项目机电学院(第二批)(5121911407)。
关键词
电动助力转向
助力特性曲线
BP神经网络
鲁棒性
Electric Power Steering
Assist Characteristic Curve
BP Neural Network
Robustnessy
