Parameter Optimization of a CNC Turning Process using an ANN-GA Method

摘要

Surface Roughness is one of the important parameters judging the quality of machining. Surface roughness causes friction and wear and tear between the mating parts which reduces the life of a machine component. Hence there is a need to minimize the surface roughness. This can be achieved by proper combination of cutting parameters. That is done by modeling using Artificial Neural Network (ANN) and optimization using Genetic Algorithm (GA). In the present work, the surface roughness in CNC turning operation is minimized by using an ANN-GA method. The inputs to the ANN include variables like cutting speed, feed and depth of cut. Speed is in terms of rpm, feed is in terms of mm/rev and depth of cut is in terms of mm. Then the cutting process is modeled using Artificial Neural Network in MAT LAB by taking the inputs as cutting speed, feed, depth of cut and output is taken as the surface roughness. Then the Artificial Neural Network model is trained for minimization of error using MAT LAB Software. Then the trained ANN model is exported into the Genetic Algorithm tool box of MAT LAB. Then in GA tool box surface roughness is minimized using GA. The optimum cutting parameters are noted.