纱线卷绕系统恒张力先进控制策略  

Research on advanced control strategy of constant tension in yarn winding system

作　　者：王延年[1] 李鹏程 廉继红[1] 范昊 王炳炎 WANG Yannian;LI Pengcheng;LIAN Jihong;FAN Hao;WANG Bingyan(School of Electronics and Information,Xi’an Polytechnic University,Xi’an 710048,China)

机构地区：[1]西安工程大学电子信息学院,陕西西安710048

出　　处：《纺织高校基础科学学报》2023年第1期49-56,共8页Basic Sciences Journal of Textile Universities

基　　金：陕西省重点研发计划项目(2021GY-076);西安工程大学(柯桥)研究生创新学院研究生联合培养项目(19KQYB02)。

摘　　要：针对纱线卷绕系统内外部扰动较多,系统具有时变性和不确定性,纱线张力值波动较大,控制精度不高导致纱线断头等问题,提出了学习速率自适应与蚁群算法联合改进的BP神经网络PID控制策略。通过学习速率自适应提高网络学习速率,并利用蚁群算法调整网络初始参数。分别测试PID算法、BP神经网络PID算法与改进BP神经网络PID算法。结果表明:基于改进BP神经网络的无刷直流电机PID控制器鲁棒性更强,性能良好,纱线卷绕系统张力波动较小,断头率降低。Aiming at the problems of yarn breakage caused by more internal and external disturbances, time-varying and uncertainty of the system, large fluctuation of yarn tension value and low control accuracy, a BP neural network PID control strategy combined with learning rate adaptation and ant colony algorithm was proposed. The learning rate was improved by adaptive learning rate, and the initial parameters of the network were adjusted by ant colony algorithm PID algorithm, BP neural network PID algorithm and improved BP neural network PID algorithm were tested respectively. The result shows that PID controller of Brushless DC motor based on Improved BP neural network have stronger robustness and good performance, less tension fluctuation of yarn winding system and lower end breakage rate.

关 键 词：纱线卷绕系统 无刷直流电机 BP神经网络 自适应学习速率 蚁群算法 

分 类 号：TS104.7[轻工技术与工程—纺织工程] TP301.6[轻工技术与工程—纺织科学与工程]