Application of machine learning to stress corrosion cracking risk assessment

• Published in: Egyptian journal of petroleum Vol. 31; no. 4; pp. 11 - 21
• Main Author: Alamri, Aeshah H.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2022; Egyptian Petroleum Research Institute
• Subjects: Forms of corrosion; Industrial internet of things (IIOT); Machine learning (ML); Stress corrosion cracking (SCC); Machine learning (ML); Industrial internet of things (IIOT); Forms of corrosion; Stress corrosion cracking (SCC)
• ISSN: 1110-0621
• DOI: 10.1016/j.ejpe.2022.09.001

•The fundamentals on the use of machine learning in stress corrosion crack (SCC) risk analysis or assessment was reviewed.•Current state of the literature on the use of machine learning in stress corrosion cracking were summarized.•Knowledge gaps and challenges of using machine learning in SCC were discussed.•Future perspectives were highlighted. One of the greatest challenges faced by industries today is corrosion and of which, one of the most vital forms is stress corrosion cracking (SCC). It brings highest forms of risks to the industry. Performing risk assessment of stress corrosion cracking is critical to ensure that industrial equipment failure is avoided by employing proper maintenance techniques. With the advancement of digital technology and the fourth industrial revolution called Industrial Internet of Things (IIOT), coupled with the availability of computing power and data, advanced analytical tools like artificial intelligence and machine learning bring powerful algorithms for performing advanced corrosion risk assessment. A perusal of the literature reveals that a review focused on the use of machine learning in corrosion risk assessment of stress corrosion cracking is scarce. So, a comprehensive and up-to-date review on this subject is timely. In this work review we present an overview on the machine learning application in the risk assessment of stress corrosion cracking. First, the current state of the art is briefly summarized. The fundamentals of machine learning algorithms and stress corrosion cracking were presented. Existing knowledge gaps were identified and discussed while the challenges and the future perspectives on the employ of machine learning in corrosion risks assessment of stress corrosion cracking were outlined.