Prediction of welding residual stress and deformation in electro-gas welding using artificial neural network. (December 2021)

• Record Type: Journal Article
• Authors: Liu, Fangfang; Tao, Congcong; Dong, Zhibo; Jiang, Kun; Zhou, Shouzhen; Zhang, Zhihang; Shen, Chen
• Is Part Of: Materials today communications. Volume 29(2021)
• Publication Date: 2021-12
• Subjects: BP neural network -- Finite element analysis -- Electro-gas welding -- Residual stress and deformation -- Human-computer interface; Materials science -- Periodicals; 620.11
• DOI: 10.1016/j.mtcomm.2021.102786
• Languages: English
• ISSNs: 2352-4928

Abstract

Abstract: Evaluating the welding residual stress and deformation in a reasonable and reliable way is the perquisite of ensuring the quality of weldments. Therefore, in this paper, an artificial neural network model is developed for the prediction of welding residual stress and deformation produced in Electro-gas welding using C++ language. The plate thickness, groove angle, groove clearance, welding current, cooling type and the welding material have been considered as the input parameters, the residual stress and deformation as output parameters in the structure of the model. The comparison between the neural network predictions and finite element analysis results indicates that the established neural network model is sufficiently accurate and efficient in predicting the residual stress and deformation. Additionally, a human-computer interaction interface software based on Qt environment is designed and implemented, aiming to obtain the prediction results in real time.