Machine Learning for Predicting Material Hardness

August 1, 2020
Image: Nikhil Khatavkar et al.

Machine learning (ML) has accelerated the development of novel materials for various applications. In a new study, IISc researchers led by Abhishek Singh at the Materials Research Centre have developed highly accurate ML models to predict the Vickers hardness – an important material property – of nickel and cobalt-based superalloys. Superalloys are used widely in aerospace, marine, chemical and petrochemical industries.

To develop the ML model, a database was initially generated comprising the microstructures, compositions and Vickers hardness of several cobalt and nickel-based superalloys. Scanning electron microscopy images were processed to obtain binary microstructures, which were then used to calculate statistically derived parameters called 2-point correlations. Principal component analysis (PCA) was performed on these correlations to select the most dominant ones. These PCA-derived correlations along with composition of the superalloys were used as descriptors for building the ML models.

The approach developed in this study can be applied for any material property.