Optimization of dry sliding wear and hardness characteristics of AL-5083 metal matrix composites reinforced with SiC, Mg, AND Sr particles

• Published in: Engineering Research Express Vol. 7; no. 2; pp. 25557 - 25573
• Main Authors: Thanikonda, Praveen Kumar, Rambabu, Govada, Ch, Ratnam, Kaladari, Adarsha Kumar
• Format: Journal Article
• Language: English
• Published: IOP Publishing 30.06.2025
• Subjects: Al-5083 alloy; brinell hardness; dry sliding wear; metal matrix composites (MMCs); signal to noise (S/N) ratio; supervised machine learning model-baseline random forest regression; taguchi method
• ISSN: 2631-8695; 2631-8695
• DOI: 10.1088/2631-8695/addd5e

This work provides an investigation of the dry slide wear properties and Brinell hardness test results for a stir-cast aluminum alloy, namely Al-5083, that has been reinforced with Silicon Carbide (SiC) Magnesium (Mg) and Strontium (Sr) particles. During the synthesis process, various weight percentages of SiC (2–12 wt%), Mg (2–12 wt%), and Sr (0.02–0.12 wt%) particles were used. An investigation was carried out to analyse the properties of wear in both the aluminum alloy and its composites. A pin-on-disc testing equipment was used to subject them to a wear test. The effect of wear factors on metal matrix composites (MMCs) dry sliding wear was studied, including factors such as, sliding speed, applied load, slide-distance, and 0%–12% of reinforcement. Results were systematically gathered by performing a sequence of experiments utilizing Taguchi’s methodology as per ASTM standards. To better model nonlinear behavior and enhance prediction, a supervised Machine Learning model-baseline Random Forest Regression-was introduced, achieving an R 2 value of 0.954. This integration of data-driven modelling enhances the novelty of the study compared with traditional methods such as ANOVA, Taguchi’s L7 design, and linear regression approaches. Tests were conducted to compare the experimental outcomes predicted by the aforementioned connection.