Machine learning insights into surface processing's impact on tool metal tribology

The processing of their surfaces can greatly influence the tribological behavior of tools used in different machining processes. In the study, we proposed the potential of a machine learning (ML) approach known as multi-objective gravitational search optimized adaptive naive bayes algorithm (MGSO-ANBA) for forecasting and enhancing tools' tribological performance in response to a wide range of surface processing variables. The first step is to gather extensive tribological data under different surface processing settings using the NEU metal dataset. A Box Cox transformation is used for preprocessing to improve the dataset's quality by correcting for abnormality and reducing variation in tribological measures. To describe the interactions between the tool and the metal, acoustic emission (AE) is used extensively. The dynamics of the tribological processes are clarified with the help of the extracted characteristics. By acquiring insights into the effect of surface processing on tool metal tribology, our suggested technique MGSO-ANBA can be utilized for classification purposes to increase understanding and optimize performance across various industrial applications. Significant tribological results, such as Surface roughness (96%), Wear rate (94%) and Coefficient of friction (97%), can be predicted with great accuracy using the suggested model.