摘要
比例电磁阀工况复杂,具有非线性、时变性等变化特点,传统控制方法难以对其进行精确控制,存在响应时间长,超调量大等弊端。为了解决当前比例电磁阀控制过程中的难题,为了获得理想的比例电磁阀控制效果,设计了一种基于前馈补偿的比例电磁阀控制方法。首先根据比例电磁阀的工作特点,建立比例电磁阀非线性变化的传递函数,然后采用复合控制器对比例电磁阀稳定性进行控制,实现比例电磁阀控制误差前馈补偿,并引入人工鱼群算法优化神经网络对PID控制器参数进行在线优化,最后在MATLAB 2016平台上与传统比例电磁阀控制方法进行了仿真模拟对比测试。实验结果表明,本文方法可以很好跟踪比例电磁阀的时变特性,改善了比例电磁阀的控制效果,缩短了响应时间,控制实时性更好,减少了超调量,比例电磁阀的整体控制效果要明显优于比对方法,具有更高的实际应用价值。
Proportional solenoid valve conditions are complex,nonlinear and time-varying characteristics,traditional control methods are difficult to obtain accurately control results so they have long response time,overshoot and other defects.In order to solve the problem in proportional solenoid valve control process to obtain the ideal control effect,a proportional solenoid valve control method based on feedforward compensation is proposed.Firstly,transfer function of proportional solenoid valve nonlinear change is proposed according to the working characteristics,secondly,hybrid controller is used to control the stability of proportional solenoid valve to compensate error,and artificial fish swarm algorithm optimized neural network is introduced to online optimize the parameters of PID,at last,the simulation test is carried out on MATLAB 2016 compared with traditional proportional solenoid valve control method.The experimental results show that the proposed method can well track time-varying characteristics of proportional solenoid valve to improve the control effect of proportional solenoid valve,which shortens the response time and has good real-time control to reduce the overshoot,and the overall control effect of proportional solenoid valve is better than the traditional methods,so the proposed method has higher application value.
作者
孙菊妹
SUN Jumei(School of Intelligent Equipment and Information Engineering,Changzhou Vocational Institute of Engineering,Changzhou Jiangsu 213164,China)
出处
《电子器件》
CAS
北大核心
2019年第1期106-110,共5页
Chinese Journal of Electron Devices
关键词
比例阀
人工鱼群算法
非线性变化
PID控制器
神经网络
proportional valve
artificial fish swarm algorithm
nonlinear change
PID controller
neural network