Modeling of the mechanical behavior of polyethylene/polypropylene blends using artificial neural networks

Polymers possess good thermal and electrical insulation properties, low density, and high resistance to chemicals, thus they have been widely used in industrial applications. Nevertheless, they are mechanically weaker and exhibit lower strength and stiffness than metals. However, their mechanical behavior can be enhanced through different techniques such as blending. Accurate estimation of the mechanical behavior is essential in structural design. Since the process of experimental measurements of a blend's properties can be costly and time consuming, this paper explores the potential use of artificial neural networks (ANNs) in the field of polymer characterization. It addresses the use of ANNs in modeling the tensile curves and mechanical properties of two commonly utilized polymers (polyethylene PE and polypropylene PP) and their blends. Blends of different proportions have been considered. The experimentally acquired data is used to train and test the neural network's performance. The key system inputs for the ANN modeler are blend ratio and percent strain, and the system output is the stress. The ANN-predicted outputs were compared and verified against the experimental date. The study indicates that a multilayered ANN can simulate the effect of the blending ratio on the mechanical behavior and properties to a high degree of accuracy. It also demonstrates that ANN approach is an effective tool that can be adopted to reduce cost and time of the experimental work. Moreover, the results show that ANNs demonstrate promising potential in the area of polymer characterization.