点击次数：

发表时间：2020-01-01

发表刊物：Chuan Bo Li Xue/Journal of Ship Mechanics

所属单位：运载工程与力学学部

卷号：24

期号：10

页面范围：1294-1301

ISSN号：1007-7294

摘要：Welding deformation has always been a key problem in shipbuilding, which is generated by residual stress of welding. In order to predict the distribution of residual stresses, the support vector machines algorithm (SVM) is used in this paper for the high-strength steel sheets butt welding. Seven vectors which include sheet length, sheet width, sheet thickness, welding electric current, electric arc voltage, welding speed and welding type are selected as the inputs of SVM to predict the width of longitudinal residual tensile stress region, the width of high stress, and the maximum stress. The results show that the prediction accuracy of SVM prediction model is very high and the model has great values in practical applications and popularization. © 2020, Editorial Board of Journal of Ship Mechanics. All right reserved.

备注：新增回溯数据