Comprehensive Evaluation and Analysis Method of Material Aging Based on Machine Learning

Abstract

In this paper, the mechanical properties of polypropylene (PP) and polycaprolactam (PA6) were analyzed after they were immersed in three common automobile stain solutions, namely, No.95 gasoline, glass water and car wash liquid, respectively. The common machine learning methods such as grey correlation degree and technique for order preference by similarity to ideality were used to analyze and process the mechanical property data of the material after immersion, and the reliability of the application of machine learning in material property exploration was verified by material mechanism analysis and tensile fracture morphology analysis. The results showed that glass water and car washing liquid had little effect on the mechanical properties of polypropylene and had great effect on the mechanical properties of polycaprolactam, 95 gasoline had little effect on the mechanical properties of polycaprolactam and had great effect on the mechanical properties of polypropylene. The comprehensive anti-aging value of polycaprolactam soaked in 95 gasoline was the highest in all six groups of tests, and the anti-aging performance was the best. The comprehensive anti-aging value of polypropylene soaked in No.95 gasoline is the lowest among all the six groups of tests, and the anti-aging performance is the worst.

Keywords: mechanical properties, machine learning, comprehensive evaluation, Anti-aging performance