Active learning-based KNN-Monte Carlo simulation on the probabilistic fracture assessment of cracked structures. (January 2022)

• Record Type: Journal Article
• Authors: Guo, Kaimin; Yan, Han; Huang, Dawei; Yan, Xiaojun
• Is Part Of: International journal of fatigue. Volume 154(2022)
• Publication Date: 2022-01
• Subjects: Probabilistic damage tolerance -- Monte Carlo Simulation -- K-Nearest Neighbors -- Probability of fracture -- Active learning; Materials -- Fatigue -- Periodicals; Materials -- Fatigue; Periodicals; 620.1122
• DOI: 10.1016/j.ijfatigue.2021.106533
• Languages: English
• ISSNs: 0142-1123

Abstract

Highlights: The probabilistic fracture assessment could be treated as a classification problem. Active learning strategy could get accurate POF results using a few samples. The proposed method is 3 or 4 orders of magnitude more efficient than MCS. Abstract: The probability of fracture (POF) assessment of complex cracked structures is a difficult task in the reliability assessment of engineering structures. Due to the complexity of structure, balancing efficiency and accuracy is the top concern in POF calculation. In this study, a novel probabilistic solution method called AKNN-MCS (A ctive learning-based K -N earest N eighbors-M onte C arlo S imulation) is proposed. Combining the active learning strategy and the KNN algorithm, this method could get accurate POF results using a few samples. In detail, POF calculation is treated as a classification problem. A learning function is proposed to select sample points near the limit state surface. Then the selected sample points are added into training data set T . A convergence criterion is defined to decide when to stop the enrichment of T . Thanks to the above active learning strategy, the trained KNN model could have a great generalization ability with only a few training samples required. The proposed method is validated by POF assessment of a finite thickness plate containing a surface semi-elliptical crack and POF assessment of the CT specimen. Results show that AKNN-MCS is three or four orders of magnitude more efficient than MCS for almost identical POF results.