Abstract
Protective coatings are essential for improving magnesium (Mg) alloy performance and longevity in a variety of industrial environments. The wear and tear, or attrition loss, that these coatings endure over time, continues to be a major problem. Conventional techniques to calculate attrition loss are imprecise and unable to consider the intricate relationships between coating components and surroundings. Our research suggests a novel machine learning (ML) method to forecast attrition loss in magnesium alloys with spray-on coatings as a solution to this problem. For the forecasting model in this paper, adaptive seagull optimization integrated random forest (ASO-RF) strategy is suggested. Wear loss data that was actually obtained for AZ91D Mg-alloyed specimens that were sprayed with various settings was used in this investigation. With the help of the gathered dataset, the ASO-RF algorithm is trained, allowing for precise attrition loss forecasts under different circumstances. In terms of attribute loss prediction, the suggested ASO-RF method's performance in Matlab implementation is examined and contrasted. This method presents a viable way to maximize coating formulas and thicknesses to reduce attrition, improving the overall performance and durability of magnesium alloys.

This publication has 19 references indexed in Scilit: