Fuzzy controlling withdrawal technology by numerical simulation to optimize directional solidification process of superalloy casting

Abstract

Numerical simulation methods become more accurate and intelligent to aid the directional solidification (DS) process. In the research, a fuzzy controlling model was built to optimize DS process. In this controlling model, some input variables such as mushy-zone width, temperature gradients were extracted from calculation of DS process and a multivariable fuzzy rule was structured on the nonlinear function of the input DS variables and the key technological parameter, withdrawal speed (v), meanwhile some other influence factors also considered in the fuzzy rule, such as the areas of sections, the positions of the areas ' sudden changes, etc.. The fuzzy controlling model coupled with CA-FD method could be used to optimize v real-timely during the calculation and a smooth v curve was got instead of the initial constant value. The optimized v curve was proved to be more flexible and adaptive for a steady and stray-grain free DS process by simulation and experiment.