首页|期刊导航|现代电子技术|基于模糊神经网络的逆变点焊电源控制优化

基于模糊神经网络的逆变点焊电源控制优化OA

Inverter spot welding power supply control optimization based on fuzzy neural network

中文摘要英文摘要

针对传统PID控制在交流逆变点焊电源中动态响应迟缓、鲁棒性不足的问题,提出一种软硬件协同优化的模糊自适应PID控制策略,以实现交流逆变点焊电源动态性能与抗干扰能力的提升,并适用于复杂多变的焊接工况.硬件层面,基于Multisim设计主电路,采用600 V IGBT、50 A/1 000 V整流桥及高精度霍尔电流传感器(CSM050B),优化IGBT驱动电路(含光电耦合器TLP521、驱动器IR2110PBF)与电流采样系统;软件层面,构建二输入三输出模糊控制器,通过实时误差及变化率在线调整PID参数,并采用Simulink构建模糊控制算法模型,结合模糊神经网络实现参数二次优化.结果表明:相较于传统PID控制,所提控制策略的系统超调量降低14.5%,稳态误差减少25%,在50~250 Hz频率内电压波形对称;相较于传统工频及同类逆变电源,IGBT开通/判断电压稳定在15 V/-7.5 V,能量转换效率提升12%,次级整流损耗降低约10%.由此验证了软硬件协同优化能够有效抑制电网波动与负载扰动,显著增强系统动态性能与鲁棒性.

In allusion to the problems of slow dynamic response and insufficient robustness of traditional PID control in AC inverter spot welding power supplies,a fuzzy adaptive PID control strategy with collaborative optimization of hardware and software is proposed to improve the dynamic performance and anti-interference ability of AC inverter spot welding power and adapt to complex and variable welding conditions.At the hardware level,the main circuit is designed based on Multisim,the 600 V IGBT,50 A/1 000 V rectifier bridge and high-precision hall current sensor(CSM050B)are used to optimize the IGBT drive circuit(including optocoupler TLP521 and driver IR2110PBF)and current sampling system.At the software level,a two-input three-output fuzzy controller is constructed to adjust PID parameters online based on real-time error and error change rate,and Simulink is used to construct the model of the fuzzy control algorithm model,so as to realize secondary optimization of parameters by combing with the fuzzy neural network.The results show that,in comparison with traditional PID control,the system overshoot of the proposed strategy is reduced by 14.5%,the steady-state error is reduced by 25%,and the voltage waveform is symmetrical within the frequency range of 50~250 Hz.In comparison with with traditional power-frequency power supplies and similar inverter power supplies,the IGBT driving voltage is stabilized at 15 V/-7.5 V,the energy conversion efficiency is increased by 12%,and the secondary rectification loss is reduced by 9%.It indicates that collaborative optimization of hardware and software can effectively suppress power grid fluctuations and load disturbances,and significantly enhance the dynamic performance and robustness of the system.

张宇浩;牛园园;左广宇;窦银科;马春燕;付骏宇

太原理工大学 电气与动力工程学院,山西 太原 030024太原理工大学 电气与动力工程学院,山西 太原 030024太原理工大学 电气与动力工程学院,山西 太原 030024太原理工大学 电气与动力工程学院,山西 太原 030024太原理工大学 电气与动力工程学院,山西 太原 030024太原理工大学 电气与动力工程学院,山西 太原 030024

信息技术与安全科学

交流逆变点焊电源模糊神经网络模糊自适应PID控制IGBT驱动电路电流采样系统能量转换效率

AC inverter spot welding power supplyfuzzy neural networkfuzzy adaptive PID controlIGBT drive circuitcurrent sampling systemenergy conversion efficiency

《现代电子技术》 2026 (16)

1-6,6

国家自然科学基金青年基金项目:北极海冰融池跨季节演化特征与热力学过程研究(42306260)企业委任横向科研项目(技术开发类):双机头贴片系统及控制算法研发(RH2400001443)2023年太原理工大学大学生创新创业训练计划项目(RC2300004310)山西省水利技术推广与应用项目:抗冰型水库水质全天候自动监测浮标关键技术研发及应用(2025GM22)极地生态与气候变化教育部重点实验室开放课题:北极海冰内部力学行为跨季节演化观测方法与影响因素研究(SO02025-04)

10.16652/j.issn.1004-373X.2026.16.001

评论