MECHANICAL PROPERTIES MODELING OF SEVERELY PLASTICALLY DEFORMED BIODEGRADABLE ZK60 MAGNESIUM ALLOY FOR BONE IMPLANTS

Abstract

MAGNESIUM (MG) AND ITS ALLOYS CAN BE UTILIZED IN BONE REGENERATION PROTOCOLS OWING TO THEIR MECHANICAL ROBUSTNESS, IN SITU DEGRADATION, AND CAPABILITY OF MOTIVATING THE GENERATION OF NEW BONES. HERE, SOME DRAWBACKS OF MG-BASED ALLOYS, I.E. THE YIELD STRENGTH (Σ_YS) AND ULTIMATE TENSILE STRENGTH (Σ_UTS), WERE ADDRESSED MOSTLY THROUGH PARALLEL TUBULAR CHANNEL ANGULAR PRESSING (PTCAP). IN VIEW OF THE MULTIFACETED VARIABLES OF THE PTCAP WITH NONLINEAR INTERACTIONS, A PRECISE DETERMINATION OF THE MATERIAL PROPERTIES REQUIRES A LARGE NUMBER OF EXPERIMENTS, WHICH INCREASES THE PROCESSING TIME AND COSTS. THEREFORE, IN THE PRESENT RESEARCH, TWO MACHINE LEARNING (ML)-BASED MODELS WERE DEVELOPED TO ACHIEVE APPROPRIATE COMBINATIONS OF MECHANICAL PROPERTIES FOR BONE IMPLANT PURPOSES. SPD METHOD, NUMBER OF THE PASS, AND PROCESSING TEMPERATURE WERE THE THREE INPUT PARAMETERS CONSIDERED IN THE CREATION OF THE MODELS TO ASSESS Σ_YS, Σ_UTS, AND ELONGATION TO FAILURE (Δ) OF THE PROCESSED ALLOYS, WHERE THE PERFORMANCE OF MODELS WAS EVALUATED IN TERMS OF CORRELATION COEFFICIENT (R), THE MEAN SQUARED ERROR (MSE), AND THE MEAN ABSOLUTE ERROR (MAE). THE RESULTS OF MODELING VERIFIED THAT THE PROPOSED GENE EXPRESSION PROGRAMMING (GEP) AND GENETIC PROGRAMMING (GP) MODELS WERE CAPABLE OF PRECISELY PREDICTING THE MECHANICAL PERFORMANCE OF THE SPD-PROCESSED BIODEGRADABLE ZK60 MG ALLOY FOR ORTHOPEDIC APPLICATIONS.