Process Monitoring: Distinguishing Defect Shapes by Strain Field Signatures. (September 2022)

• Record Type: Journal Article
• Authors: Basu, Saurabh; McComb, Christopher C.; Rifat, Mustafa; Sun, Hongtao; Kumara, Soundar
• Is Part Of: Manufacturing letters. Volume 33(2022)Supplement
• Page Start: 808
• Page End: 816
• Publication Date: 2022-09
• Subjects: Monitoring -- Defect -- Strain fields; Manufacturing industries -- Periodicals; Production engineering -- Periodicals; Manufacturing industries; Periodicals; 670
• DOI: 10.1016/j.mfglet.2022.07.100
• Languages: English
• ISSNs: 2213-8463

Abstract

Abstract: In this research, a theoretic physics-based framework for identification of defects via analysis of strain fields is presented. This framework comprises identification of self-similarity of strain fields followed by their dimensionality reduction using kernel based principal component analysis. The efficacy of this framework is tested qualitatively, by visual analysis, and quantitatively, using numerical classification algorithms. We see high (>95%) accuracy of classification via cross-validation studies using support vector machine algorithm. These results suggest that strain field can provide a viable approach for constructing highly robust in line defect detection system in modern manufacturing environments.