Corrosion is a major form of materials degradation, causing substantial economic losses estimated at ~4% of global GNP. Its complex electrochemical mechanisms and interactions among environmental, climatic, and materials-related factors make reliable corrosion prediction highly challenging.
Our research integrates computational chemistry, reaction mechanistic modelling, and machine learning with large datasets from environmental monitoring and field corrosion testing to investigate corrosion mechanisms, identify key controlling factors, and assess the effects of climate change on atmospheric corrosion. By integrating machine learning with the concept of Damage Accumulation, we aim to improve long-term corrosion prediction under climate change conditions. Using extensive field and environmental data, we are developing a digital corrosion map of New Zealand for predicting corrosion rates and supporting infrastructure durability and corrosion management.
The presentation will also introduce self-healing, smart, and environmentally sustainable coatings designed to provide active corrosion protection and replace toxic Cr⁺⁶-containing treatments. Together, these computational and materials-engineering approaches offer new strategies for predicting and mitigating atmospheric and marine corrosion and improving the sustainability of engineering infrastructure.
 

AI CORROSION