汽车左侧围内板后部总成焊接工作台数字孪生系统研究

Digital Twin System of Welding Table for Rear Assemblyof the Left Inner Plate of Automobile

针对汽车左侧围内板后部总成连接件设计了焊接机器人工作台。首先,基于Visual Components搭建焊接机器人、焊钳和焊装夹具的数字孪生体;然后,通过Python脚本开发TCP协议进行通信,完成数字孪生系统的虚实联动;最后,基于BP神经网络构建焊接质量预测模型,并通过正交实验对其预测效果进行验证。实验结果表明,此焊接机器人工作台的预测准确率达到94.81%,能够满足实际生产要求。

The welding robot table was designed for the rear assembly joint of the left inner panel of automobile.Firstly,based on Visual Components,the digital twin system of the welding robot,welding tongs and welding fixtures was constructed.Then TCP protocol was developed through Python script for communication to complete the virtual and real linkage of the digital twin system.Finally,the welding quality prediction model was constructed based on BP neural network,and the prediction effect was verified through orthogonal experiments.The results of experiment show that its prediction accuracy reaches 94.81%,which is able to meet the requirements of actual production.