基于BP神经网络的卷染机温度控制技术  

Temperature control technology of dyeing machine based on BP neural network

作　　者：魏苗苗 刘洲峰[1] WEI Miaomiao;LIU Zhoufeng(College of Electronics and Information,Zhongyuan University of Technology,Zhengzhou,Henan 450007,China)

机构地区：[1]中原工学院电子信息学院

出　　处：《毛纺科技》2019年第7期71-75,共5页Wool Textile Journal

基　　金：河南省自然科学基金重点项目(162300410338);河南省重点科技攻关计划项目(152102210152)

摘　　要：为了解决高温高压卷染机中的温度控制问题,提出了一种改进的温度控制算法。该算法在BP神经网络PID控制算法的基础上引入积分分离结构,当温度控制误差较大时采用PD结构,去除积分成分以提高控制调整速度,当温度控制误差较小时采用完整的PID控制算法结构,保留积分成分以保证控制精度。仿真结果显示:相对于传统的PID控制算法,基于BP神经网络的PID控制算法能够保证在较高温度时控制精度的前提下,提高温度控制稳定速度、缩短温度控制时间,达到更好的控制效果。In order to solve the problem of temperature control in dyeing machines under high temperature and high pressure,an improved temperature control algorithm was proposed.The integral separation structure was introduced into the algorithm based on PID algorithm aided by BP neural network.When the temperature control error was above the threshold value,the integral component of PID controller was removed from the system to improve the control speed.When the temperature control error was below the threshold value,integral component was retained by the system to ensure the control accuracy.The simulation results show that,compared with traditional PID control algorithm and the PID algorithm based on BP neural network,this algorithm can shorten the time of temperature control to achieve better control effect promising the higher temperature control precision.

关 键 词：逆向传播神经网络 卷染机温度控制 比例积分微分算法 积分分离 

分 类 号：TF325.64[冶金工程—冶金机械及自动化]