Prediction of electron beam weld quality from weld bead surface using clustering and support vector regression

作者：

Highlights：

• Proposes a quality predictive model for EBW process using weld surface attributes.

• Varying weld surface profiles have good correlation with mechanical properties.

• Develops a predictive model for surface attributes using SVR.

• FCM and DBSCAN cluster the weld joint into poor, fair and good weld.

• Avoids time-consuming trial-error, conventional nondestructive testing methods.

摘要

•Proposes a quality predictive model for EBW process using weld surface attributes.•Varying weld surface profiles have good correlation with mechanical properties.•Develops a predictive model for surface attributes using SVR.•FCM and DBSCAN cluster the weld joint into poor, fair and good weld.•Avoids time-consuming trial-error, conventional nondestructive testing methods.

论文关键词：Electron beam welding,Copper alloy,Weld quality,Clustering,Fuzzy C-means,Support vector regression

论文评审过程：Received 20 December 2021, Revised 11 June 2022, Accepted 22 August 2022, Available online 27 August 2022, Version of Record 29 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118677