Enhancing Predictive Models for Erosion Rates in Aerospace Alloys: Machine Learning Techniques and Data Augmentation Strategies
Using machine learning to predict the maximum erosion rate and incubation period of alloys used in aerospace like Ti64,12%Cr SS, Al2024, Al7075, TiAl, etc., underwater droplet impacts from different rigs that are available in works of literature. The output from different rigs can give different output values for the same input values because of the difference in rig geometry and experimental conditions. Need to use machine learning to overcome this problem and also use techniques to overcome the small data set problems like data augmentation to improve model accuracy and also after this validation of the final model. We need to use different algorithms like Linear regression, Decision tree-based regression, and Neural network-based model and make comparisons of the model and which one outperforms and validated.
