基于BP神经网络的六角编织机控制策略  

Control strategy of hexagonal braiding machine based on BP neural network

作　　者：付睿云 周炯 吴垠舟 FU Ruiyun;ZHOU Jiong;WU Yinzhou(East Pudong Branch,Shanghai Open University,Shanghai 201200,China;Fudan University Affiliated Pudong Medical Center,Shanghai 201399,China;Hangzhou Xiaoshan Technician College of Intelligent Design and Manufacturing,Hangzhou,Zhejiang 311201,China)

机构地区：[1]上海开放大学浦东东校,上海201200 [2]复旦大学附属浦东医院,上海201399 [3]杭州萧山技师学院智能设计与制造学院,浙江杭州311201

出　　处：《毛纺科技》2025年第1期104-110,共7页Wool Textile Journal

基　　金：上海市浦东新区教育科学研究课题(2024D055)。

摘　　要：为提高六角编织机控制系统的抗干扰能力,研究六角编织机的编织原理,提出一种BP(Back Propagation)神经网络与PID相融的编织控制策略,设计多电动机并行同步控制的方案,借助Simulink工具搭建六角编织机控制系统的仿真模型。仿真结果表明:与传统PID控制相比,BP-PID控制策略峰值时间下降0.032 s,最大超调量缩减0.272%,调整时间缩短0.524 s。并在0.5 s引入干扰,BP-PID算法显出更快的响应速度、更小的超调量和更强抗干扰能力。借助实际的六角编织机,从单电动机卡顿率、单次编织卡顿率和产品合格率3个指标验证了BP-PID控制策略有效性,提高编织物产品合格率。To enhance the anti-interference capability of the hexagonal braiding machine control system,the braiding principles of a hexagonal braiding machine was investigated in this study,and a braiding control strategy that integrates BP(Back Propagation)neural network with PID was proposed.A scheme for multi-motor parallel synchronous control was designed,and a simulation model of the hexagonal braiding machine control system was constructed using Simulink tools.Simulation results indicate that compared to traditional PID control,the BP-PID control strategy reduces the peak time by 0.032 seconds,decreases the maximum overshoot by 0.272%,and shortens the settling time by 0.524 s.Moreover,when disturbances were introduced,the BP-PID algorithm exhibits faster response speed,smaller overshoot,and stronger disturbance rejection capability.The effectiveness of the BP-PID control strategy is validated using real hexagonal braiding machines and increasing the product qualification rate of the woven fabric,based on three metrics:single motor stall rate,single braiding stall rate,and product qualification rate.

关 键 词：六角编织机 步进旋转式 BP神经网络 PID自整定 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]