Optimization of End milling process for minimizing surface roughness with combined artificial Neural Network and Genetic Algorithm

Through enormous development of machining methods, applying optimization method in machining process to improve quality seems to be important. One of the most important parameter of a work piece is its surface roughness. As surface roughness decrease, the quality of work piece increase. In this study, optimization of input parameter of end mill machining to reach minimum surface roughness is investigated. Among these parameters five of them selected and Taguchi method is used for the design of experiments. The process is modeled with neural network method and using try and error test 5-8-6-1 architecture. Genetic algorithm is used for process optimizing and neural network model is selected as the target function. For three different tool path strategies, optimization has been conducted and results are discussed. Using genetic algorithm decrease surface roughness to 0.85 μm. Finally selected level of Taguchi method is analyzed and levels with maximum signal to noise ratio are introduced as optimized level that have minimum surface roughness.