基于模糊神经网络PID的高频焊接机温度控制的研究  

Research on temperature control of high frequency welding machine based on fuzzy neural network PID

作　　者：刘鹏程 马旭[1] 徐智宏 邢少松 温泉 王浩丞 LIU Pengcheng;MA Xu;XU Zhihong;XING Shaosong;WEN Quan;WANG Haocheng(School of Electrical Engineering and Automation,Tianjin University of Technology,Tianjin 300384,China)

机构地区：[1]天津理工大学电气工程与自动化学院,天津300384

出　　处：《天津理工大学学报》2024年第6期15-20,共6页Journal of Tianjin University of Technology

基　　金：天津理工大学研究生教育教学研究与改革项目(YBXM2205)。

摘　　要：在高频感应焊接过程中经常会出现塌焊现象,其主要原因是由于焊接时温度控制不当所导致,使焊接时的温度超过了焊件熔化温度固相线,造成焊件被穿透或凹陷。目前工业上常用的高频感应焊接温度控制方法仍使用传统比例、微分、积分(proportion integration differentiation,PID)控制器,但传统PID对非线性或具有大延迟的系统控制效果较差,容易出现较大的超调,使得焊接时的温度超过焊件融化温度固相线,造成塌焊。文中将模糊控制与神经网络相结合后对PID算法进行优化,提高了控制性能,保证了焊接时温度的准确性。仿真结果表明:基于模糊神经网络的PID控制器与传统PID相比,温度的控制精度显著提高,最大超调仅为既定温度的3.6%,同比下降了5.5%,同时调节时间缩短了3.9 s,有效提高了高频感应焊接时温度=控制的精度,避免了塌焊的产生。In the process of high-frequency induction welding,the phenomenon of welding collapse often occurs,which is mainly caused by improper temperature control during welding,so that the welding temperature exceeds the solid phase line of the welding welding temperature,resulting in penetration or depression of the welding parts.At present,the conventional PID control algorithm is still used to control the temperature of high-frequency induction welding in the industry.However,the traditional PID control algorithm has poor control effect on the nonlinear system or the system with large delay,which is easy to appear large overkill,so that the welding temperature exceeds the solid phase line of the melting temperature of the welding parts,resulting in collapse welding.In this paper,the fuzzy control and the neural network are combined to optimize the PID algorithm,improve the control performance and ensure the accuracy of welding temperature.The simulation results show that compared with the traditional PID controller,the temperature control accuracy of the fuzzy neural network-based PID controller is significantly improved.The maximum overshoot is only 3.6%of the established temperature,which decreases by 5.5%compared with the same period.At the same time,the adjustment time is shortened by 3.9 s,which effectively improves the precision of temperature control in high-frequency induction welding and avoids the occurrence of welding collapse.

关 键 词：高频感应焊接 熔化温度固相线 神经网络 超调量 模糊控制 

分 类 号：TG40[金属学及工艺—焊接]