Research On The Expert System For Optimizing Hard-milling Process

The high-speed milling technology was applied to the machining of steels in their hardened state for manufacturing of molds and dies shortly after its renaissance. Despite being harnessed by numerous advantages, the high-speed milling of hardened steels comes also with a major demerit of dire shortening of tool life. By developing an expert system for optimizing high speed hard-milling process, the hard-milling parameters are experimentally investigated for the purpose of enhancing tool life, improving surface finish and material removal rate. The major research works are as follows: On the basis of analyzing the framework of the expert system for optimizing high speed hard-milling process, a new expert system utilizing fuzzy reasoning was created. And the fuzzy rule-base was adjusted for maximum accuracy by employing the simulated annealing algorithm.The effects of the following parameters: work-piece material's microstructure and hardness, tool coating, helix angle, rake angle, milling-orientation, methods of cooling and lubricating, cutting speed, feed rate, radial depth of cut on tool life, roughness of work-piece's side surface and end surface, and cutting forces were investigated by series of hard-milling experiments. The experimental results were analyzed using ANOVA and significance of effects of all the tested parameters, on performance measures, was determined.A new expert system for optimizing high speed hard-milling process was developed by using wxCLIPS, a productive development and delivery expert system tool. The experimental and ANOVA results were utilized for the making of fuzzy rule-base. And also the simulated annealing algorithm was employed to adjust the fuzzy rule-base for maximum accuracy.In the following stage an innovative machine learning technique was utilized for creation of a self-developing expert system, that can: self-retrieve and self-store the experimental data; automatically develop fuzzy sets for numeric variables involved; automatically generate rules for optimization and prediction rule-bases; resolve the conflict among contradictory rules; and automatically update the interface of expert system according to newly introduced hard-milling variables.