NONLINEAR RELATIONSHIP BETWEEN PROCESSING PARAMETERS AND MECHANICAL PROPERTIES IN Ti6Al4V ALLOY BY USING FUZZY NEURAL NETWORK

摘要

The purpose of this paper is to develop a nonlinear model for prediction of mechanical properties of the Ti6Al4V alloy based on its processing parameters. A fuzzy neural network (FNN) model has been employed to establish the relationship. Four input variables (deformation temperature, deformation degree, solution and aging temperature) are arranged in the model, while four mechanical properties (ultimate tensile strength, yield strength, elongation and area reduction) are used as output variables. After the training process, the result showed that the model used to predict the properties of Ti6Al4V alloy has good learning precision and good generalization and the maximum relative error is less than 10%. Based on the model, the effects of the solution and aging temperature on the corresponding mechanical properties were deeply studied. The optimum matching of the forged Ti6Al4V would be treated at the solution temperature of 900～950°C, and aging at temperature of 500～550°C.