Experimental investigation of high pressure hybrid jet assisted turning and process optimization by genetic algorithm and neural network

High pressure hybrid jet assisted machining is an efficient method to improve the machining conditions. The significant advantage of this method is concentration of high pressure fluid on machining zone and reduction in tool-chip interface. In this research, a collection of pumps and water jet pump was used in order to supply a high pressure jet. A specific tool holder was designed and manufactured. Experiments were designed to investigate the process parameters such as jet pressure, cutting speed, feed rate and depth of cut. During the experiments, the cutting forces and surface roughness were measured. Results showed the existence of the optimum conditions of the minimum cutting force and surface roughness in hybrid jet assisted machining. Thus, the optimization of process parameters is necessary in order to use hybrid machining more efficiently. A neural network with suitable topology was utilized which was trained by genetic algorithm to obtain the predictive model. Ultimately, genetic algorithm was applied to optimize process parameters. Results showed the ability of employed algorithm to predict the optimum parameters with considerable accuracy.