Wear and frictional training of alumina base compositions using tribology-strategy based on a machine learning approach

Due to the intricate and dynamic relationships that exist between the material and its working surroundings, tribological reactions in alumina-based composites are a considerable problem. Conventional approaches for comprehending Wear and friction fail to capture the complex interactions among variables influencing these characteristics. To address this, study leverages a machine learning (ML) approach to create a thorough knowledge base regarding the tribological qualities of compositions that use Alumina. To anticipate wear and friction qualities, a novel dynamic grey wolf-driven support vector machine (DGW-SVM) technique is presented in this paper. The compositions are undertaken through Wear and frictional study in which a variety of material elements and production circumstances are methodically investigated. The suggested model incorporates experimental evidence from tribological testing and it is developed to identify trends and relationships in the dataset. The proposed approach is developed in Python and its achievement is evaluated using multiple variables, including RMSE (0.098), MAE (0.114) and R2 (0.998). The learned algorithm is then applied to unproved compositions to forecast their Wear and frictional performance, allowing for fast and intelligent decision-making. A contrast with conventional experimental approaches is used to evaluate the effectiveness of the suggested methodology, which shows promise for substantial reductions in time and resources.