期刊文献+

基于事件触发GrHDP碳纤维角联织机张力控制方法

Based on Event-triggered GrHDP Carbon Fiber Diagonal Loom Tension Control Method
下载PDF
导出
摘要 针对碳纤维角联织机经纱张力控制中存在的张力控制精度低与通信网络资源受限问题,提出一种基于事件触发目标再现启发式动态规划(event-triggered goal representation heuristic dynamic programing,ET-GrHDP)的张力控制方法。首先,分析碳纤维织机的织造原理,构建张力系统控制模型。其次,在GrHDP算法的基础上,采用小波神经网络(wavelet neural network,WNN)设计评价网络,BP神经网络(back propagation neural network,BPNN)设计执行网路和目标网络,以增强GrHDP近似最优控制的能力,同时引入事件触发机制,实现小波GrHDP张力控制器仅在事件触发时刻更新控制信号。最后,使用MATLAB软件对张力系统进行仿真控制实验。与经典GrHDP张力控制相比,所提控制策略在系统状态跟踪响应时间上缩短近22.9%,送经、卷取线速度均无超调量,张力超调量为1.8%,且抗干扰能力强,提高了张力系统的控制性能。在事件触发机制下,张力控制器降低近84%的计算次数,有效减少了通信资源的浪费。 Aiming at the problems of low tension control accuracy and communication network resources limitation in warp tension control of the carbon fiber diagonal loom,a tension control method based on event-triggered goal representation heuristic dynamic programming(ET-GrHDP)was proposed.Above all,the tension system control model was constructed after analyzing the weaving principle of the carbon fiber loom.In the second place,on the basis of the GrHDP algorithm,the wavelet neural network(WNN)was used to design the critic network,the back propagation neural network was used to devise the action network and the goal network to enhance the ability of GrHDP to approach approximate and optimal control,at the same time event-trigged mechanism was introduced to realize the wavelet GrHDP tension controller to update the control signals at the time of event triggering.Finally,the simulation control experiment on the tension system was performed using MATLAB software.Compared with the classical GrHDP tension control.The presented control strategy shortens the tracking response time of system states by nearly 22.9%.There is no overshoot in the linear speed of the let-off and take-up system,the overshoot of tension is 1.8%,and the anti-interference ability is strong,the control performance of the tension system is improved.Under the event-triggered mechanism,the tension controller decreases the number of calculations by 84%and reduces the waste of communication resources effectively.
作者 李文广 刘薇 朱志华 LI Wen-guang;LIU Wei;ZHU Zhi-hua(School of Mechanical Engineering,Tiangong University,Tianjin 300387,China;Tianjin Key Laboratory of Advanced Mechatronics Equipment Technology,Tianjin 300387,China)
出处 《科学技术与工程》 北大核心 2023年第20期8735-8744,共10页 Science Technology and Engineering
基金 天津市自然科学基金(17JCYBJC19400)。
关键词 碳纤维角联织机 张力控制 事件触发机制 目标再现启发式动态规划 小波神经网络(WNN) carbon fiber diagonal loom tension control event-triggered mechanism goal representation heuristic dynamic programing wavelet neural network(WNN)
  • 相关文献

参考文献9

二级参考文献61

共引文献26

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部