Manufacturing and Modeling of Polypropylene-based Hybrid Composites by Using Multiple-Nonlinear Regression Analysis

Abstract

In this research, artichoke stem particles (AS) and wollastonite (W) were used as an organic and inorganic fillers in order to improve the mechanical properties of polypropylene (PP). In this regard, PP-matrix composites containing AS and W were produced as non-hybrid and hybrid using a high speed thermo-kinetic mixer. Mechanical properties of polymer composites were investigated by the tensile test. Experimental results reveal that the highest modulus of elasticity was obtained in PP-W and the highest tensile strength was obtained in raw PP while the lowest ultimate strain value was obtained in PP-W-AS. Then, multiple-nonlinear regression analysis was employed to determine the effect of weight ratios of W and AS in PP on modulus of elasticity, tensile strength and ultimate strain. Experimental results were expressed with polynomial, rational and trigonometric models. The results show that the proposed models have well fitted with the experimental results. The coefficient of determination (R2) values were found between 0.95 and 1 in all models. Also, boundedness check control of the proposed models which gives information about whether models are realistic or not was carried out by calculating the maximum and minimum values produced by the relevant model.