A design of fuzzy inference systems to predict tensile properties of as-cast alloy

Abstract In this study, a design of Mamdani type fuzzy inference systems is presented to predict tensile properties as-cast alloy. To improve manufacturing light weight cast components, understanding mechanical components under load important. The ability deterministic models the performance component limited due uncertainty and imprecision in casting data. are introduced as promising solution. Compared other artificial intelligence approaches, Mandani allow for better result interpretation. were designed from data experts’ knowledge optimized using genetic algorithm. was used set up values engine initial database parameters. rule base automatically generated which collected testing experiments. A algorithm with real-valued coding optimize quality constructed evaluated by comparing predicted actual properties, including yield strength, Y.modulus, ultimate as-case alloy two series experimental obtained results showed that has satisfactory accuracy similar or than several machine learning methods. evaluation also demonstrated good reliability stability approach.