Prediction Of Surface Roughness In NC Machining

Surface roughness is one of the most important requirements in NC machining. To seek optimal machining parameters so as to control machined surface roughness, more and more researchers hope to model more precise surface roughness prediction models, on the basis of fewer experiments, don't be controlled by traditionally experience expressions or refer to guides/handbooks and achieved optimum machining.Simulating machined surface topography is very critical since it affects directly the surface quality, especially the surface roughness.In addition to this, simulation results will be used as basic data when process parameters are optimized.Based on lots of literatures which related with surface roughness prediction, the dissertation deals with the ball-end milling, this paper mainly study the key techniques including influence factors and influence rules of machined surface roughness, prediction of surface roughness models, as well as prediction of machined surface roughness on multi-axis NC machining.For influence factors and rules of machined surface roughness, this paper studied on the influence factors of machined surface roughness such as path interval, cutting speed, machining inclination angle and feed rate.For study of the surface roughness prediction method, based on analyzing the shortcomings of the past methods of building a surface roughness prediction model, regression analysis and response surface methodology were adopted to build a prediction model of surface roughness. Through the verification of the built prediction model of surface roughness, result shows that the prediction accuracy and generalization of the model are very high and the model can conveniently predict the effects of milling parameters on surface roughness of machined surface, wich contributes to accurately understand the variation law of quality of machined surface following milling parameters and provides the foundation for properly selecting cutting parameters ang controlling surface quality.