Machine Learning
Machine learning uses algorithms to analyze large amounts of data, learn from that data, and then adapt or perform tasks without instructions. Due to the complexity of corrosion processes, machine learning has become an important tool to help model, predict, and even detect corrosion. While there are different types of algorithms, such as artificial neural networks, random forest regression analysis, multiple linear regression, particle swarm optimization, and support vector machines, machine learning techniques enable us to better predict, mitigate, or even leverage corrosion.
“Machine Learning Approaches to Model Galvanic Corrosion of Coated Al Alloy Systems”
Several machine learning models were developed in this study to predict, analyze, and quantify galvanic corrosion arising between relatively noble fasteners and coated aluminum alloy panels.
“Automated Corrosion Detection Using Crowdsourced Training for Deep Learning”
Images and votes were collected online to help train and test a new artificial intelligence tool that evaluates images for corrosion.
“Predicting Microbiologically Influenced Concrete Corrosion in Self-Cleansing Sewers Using Meta-Learning Techniques”
Comparing standalone models with integrative models, this study sought to improve the accuracy in predicting microbiologically influenced concrete corrosion in sewers.
“Construction and Optimization of Corrosion Map in a Broad Region of Acidic Soil via Machine Learning”
A corrosion map of carbon steel in a broad region of acidic soils in China was constructed and optimized using field experiments and machine learning to aid in future materials selection, equipment maintenance, and risk assessment of buried infrastructure.
“Future Frontiers in Corrosion Science and Engineering, Part III: The Next ‘Leap Ahead’ in Corrosion Control May Be Enabled by Data Analytics and Artificial Intelligence”
This editorial provides a perspective on where the future of corrosion research is headed, including the use of machine learning and other tools to help predict corrosion performance and failures.
