Cross-domain feature selection and diagnosis of oil and gas pipeline defects based on transfer learning. (January 2023)

• Record Type: Journal Article
• Authors: Wu, Linyu; Liang, Wei; Sha, Duolin
• Is Part Of: Engineering failure analysis. Volume 143:Part A(2023)
• Publication Date: 2023-01
• Subjects: Pipeline defects -- Hybrid feature selection -- Transfer learning -- Feature adaptation -- Defect diagnosis; System failures (Engineering) -- Periodicals; Fracture mechanics -- Periodicals; Reliability (Engineering) -- Periodicals; Pannes -- Périodiques; Rupture, Mécanique de la -- Périodiques; Fiabilité -- Périodiques; Fracture mechanics; Reliability (Engineering); System failures (Engineering); Periodicals; Electronic journals; 620.112
• DOI: 10.1016/j.engfailanal.2022.106876
• Languages: English
• ISSNs: 1350-6307

Abstract

Highlights: An intelligent transfer defect diagnosis framework based on the hybrid feature selection method and transfer learning (TL) technology is proposed. A hybrid feature selection method is proposed to select features suitable for TL. The ReliefF and minimum redundancy–maximum relevancy (MRMR) feature selection methods are optimized by cross-domain feature evaluation. Monte Carlo simulation wrapper is used to form different feature subsets and measure the quality of feature subsets. The proposed diagnosis framework achieves high classification accuracy in the transfer tasks of pipeline defect cross-domain diagnosis. Abstract: Developing intelligent pipeline defect diagnosis technology is of great significance to ensure the efficiency and safety of pipeline transportation. However, a challenge in real-world tasks is the inadaptability and instability of these technologies for processing data from different distributions. Therefore, this paper proposes an intelligent transfer defect diagnosis framework that combines a hybrid feature selection method and transfer learning (TL) technology. The hybrid feature selection method consists of two cross-domain filtering feature selection methods and feature subset evaluation. A Monte Carlo simulation wrapper is used to obtain the best transfer feature subset. The obtained optimal feature subset is fed to the feature adaptation-based classification method for defect diagnosis. We used two different datasets as the target domain and source domain and simulated two transfer scenarios to verify the framework. Experimental results demonstrate that the proposed diagnosis framework achieves high accuracy and effectiveness in pipeline defect diagnosis cross-domain transfer tasks. This will facilitate pipeline maintenance engineers in solving the problem of pipeline defect diagnosis and reasonably arranging pipeline maintenance.