Prediction of tensile strength of polymers by multilayer perceptron artificial neural network

Polymer is a large molecule made up of a large number of repeating units. We have two types of synthetic and natural polymers that play an important role in both industry and life. Polymers are widely used in industry today and come in a variety of strengths. Strength is the amount of force that a material can withstand and the yield strength is the maximum strength that the material then undergoes the waxing deformation. The mechanical properties of materials are measured through important tests such as tensile tests. Factors affecting strength include impact absorption, hardness, percentage of elongation, toughness, strain and modulus. The present paper introduces a method that predicts the yield strength of polymers with the multilayer perceptron neural network. Perceptron is the simplest type of learning neuron modeling. Having known data on yield strength and elongation, modulus of elasticity, stiffness and tensile strength, the neural network forms an algorithm to find the yield strength of unknown polymers, the value of which could not be measured by tensile test. The real and the tensile test are closer. As a result, the data obtained from the experimental test can be predicted with acceptable accuracy by the model obtained in the neural network, which are consistent.

کلیدواژه ها:

Tensile strength ، Polymers ، Neural network