Treffer: Assessing the efficacy of machine learning models in hydroxyapatite nano-powder assisted electro discharge machining of Ti-6Al-4 V Grade-5 alloy.

Ti-6Al-4 V Grade-5 alloy is remarkably used in the medical, aerospace, and automotive industries owing to its characteristics like lightweight – high strength, and admirable corrosion resistance. Experiments have consistently held significant value in uncovering new knowledge and assessing the performance attributes of a process. However, due to high costs and limited trials, there is a shift towards using Machine Learning (ML) to forecast the properties of such materials. This study aims to establish a connection with the performance parameters and surface characterization of EDM of Ti-6Al-4 V Grade-5 alloy using Al<subscript>2</subscript>O<subscript>3</subscript> and SiC Hydroxyapatite (HA) nano-powder mixed deionized water. The tool wear rate (TWR), material removal rate (MRR), and surface roughness (SR) properties characteristics were analyzed by experimenting with the input variables of peak current, pulse on and off time, and powder concentration. Four different ensemble ML methods were used to predict the performance characteristics, co-starring bagging, extra tree (ET), extreme gradient boosting (XGB), and random forest (RF) regressors. The data set is used to train the proposed Python models, and the hyperparameter tuning procedure is used to find the best model for predicting the target values. The observed data shows the accuracy of the proposed model's predictions for tool wear rate values outperforms that of its predictions for MRR and SR values. The XGB model exhibits superior performance compared to the bagging, RF, and ET models, as demonstrated by its higher values of adjusted R² score, root mean square error, and mean absolute error. The study emphasizes the significance of choosing the best min_samples_split value, with min_samples_split = 2 consistently producing superior results for both RF and ET regressors. [ABSTRACT FROM AUTHOR]

Copyright of International Journal on Interactive Design & Manufacturing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)