A review on AI-driven environmental-assisted stress corrosion cracking properties of conventional and advanced manufactured alloys

Abstract

Stress corrosion cracking (SCC) poses a significant challenge to the integrity and longevity of both conventional and advanced manufactured alloys, impacting critical industries such as aerospace, marine, and nuclear energy. This review explores the use of artificial intelligence (AI) to analyse and predict SCC properties. Traditional alloys such as steel, aluminium, and titanium, as well as advanced materials such as additive-manufactured and high-entropy alloys, exhibit unique SCC behaviours influenced by corrosive environments, mechanical stress, and temperature variations. AI techniques, including machine learning models, offer transformative approaches to understanding and mitigating SCC. Leveraging extensive datasets from experiments and field studies, AI can identify patterns and correlations that traditional methods might miss. Predictive modelling through supervised learning forecasts SCC initiation and propagation, aiding in material design and maintenance. Unsupervised and reinforcement learning enhances alloy composition and processing parameter optimisation for better SCC resistance. In conventional alloys, AI identifies specific SCC conditions, leading to improved protective measures. For advanced alloys, AI-driven insights tailor additive manufacturing processes and design high-entropy alloys with superior corrosion resistance. This review includes case studies demonstrating AI's effectiveness in predicting SCC across various alloy–environment combinations, highlighting successes and challenges. Despite hurdles such as data quality and integration, AI holds immense potential to revolutionise SCC research, enhancing understanding of SCC mechanisms and fostering innovations in alloy design and preventative maintenance.