New Machine Learning Model to Combat Corrosion
Scientists of the Max-Planck-Institut für Eisenforschung have developed a machine learning model that enhances the predictive accuracy of corrosion behavior in alloys by up to 15 percent compared to existing frameworks.
The model developed by the team of the Max Planck Institute for Eisenforschung uncovers new, but realistic corrosion-resistant alloy compositions. Its distinct power arises from fusing both numerical and textual data.
While initially designed to combat pitting corrosion in strong alloys, the model can also be applied to various other alloy characteristics. The researchers published their latest results in the journal Science Advances.
“Every alloy has unique properties concerning its corrosion resistance. These properties do not only depend on the alloy composition itself, but also on the alloy’s manufacturing process. Current machine learning models are only able to benefit from numerical data.
However, processing methodologies and experimental testing protocols, which are mostly documented by textual descriptors, are crucial to explain corrosion,” commented Kasturi Narasimha Sasidhar, lead author of the publication and former postdoctoral researcher at the Max-Planck-Institut für Eisenforschung.
The research team used language processing methods, akin to ChatGPT, in combination with machine learning (ML) techniques for numerical data and developed a fully automated natural language processing framework. Furthermore, involving textual data into the ML framework allows to identify enhanced alloy compositions resistant to pitting corrosion.
“We trained the deep-learning model with intrinsic data that contain information about corrosion properties and composition. Now the model is capable of identifying alloy compositions that are critical for corrosion-resistance even if the individual elements were not fed initially into the model”, says Michael Rohwerder, co-author of the publication and head of the group Corrosion at the Max-Planck-Institut für Eisenforschung.
In the recently devised framework, Sasidhar and his team utilized manually gathered data as textual descriptors. Presently, their objective lies in automating the process of data mining and seamlessly integrating it into the existing framework. The incorporation of microscopy images marks another milestone, envisioning the next generation of AI frameworks that converge textual, numerical, and image-based data.
Source
