Prediction of Wear Characteristics of Polymer Composites by ANN Modified by GA

Wear characteristics of any material are highly improved by reinforcing it with particulates. Wear resistance enhanced materials are very much essential in industries today. To reduce the cost of the composites, naturally available materials are preferred as reinforcements. Industrial waste in the powder form is reinforced with Nylon in various concentrations and wear tests were conducted at different parameters. The specific wear rate was found by experiments. Artificial Neural Network which is equivalent to biological network is usually used to predict the characteristics of the materials. An artificial neural network was developed to predict the wear rate of these composites. To get precise results, various techniques are being followed by researchers in developing the architecture of the neural network. The architecture of the developed network was optimized by applying genetic algorithm to obtain high accuracy in the predicted values. The neural network developed was able to predict the wear rates with more than 98% accuracy.