Predict And Experiment Research Of Surface Roughness For Milling

The surface roughness is one of the important indicators of measure the surfaceof the complete process. If we can accurately predict the surface roughness beforethe actually processing, we can improve the quality of work piece surface and savecost. Titanium alloy has the characteristics of high activation, low heat conductivityand low elastic modulus. In the cutting performance, these features are large cuttingforce, high cutting temperature and made the cutting tool severe, these factors will bean adverse effect on the integrity of the machined surface of parts and influence thesurface roughness.The surface roughness is the important indicators to measure the surface of thetitanium alloy in milling. Through the establishment of the surface roughnessprediction model, it can provide the basis for optimization of milling parameters. Inthis paper, using the SVM method can set up the model in cutting speed, feed rateand cutting depth. Meantime, the prediction model based on RSM theory isestablished as well. Then, some machine experiments are executed tocontradistinctively verify prediction accuracy of the both models.In this paper, using the orthogonal experiment on titanium alloy materials,select the process parameters of cutting speed, feed rate and cutting depth to designtwo sets of tests. Based on the analysis of the experimental results by direct-viewinganalysis method and variance analysis method, we can do further research. Using theSVM method can set up the model in cutting speed, feed rate and cutting depth. Letthe test results into the forecasting model to compare the predictive value with theexperiment value, and then calculate the relative prediction error. The results showthat the SVM method has lower relative error. Finally, using the RSM method canset up the model in cutting speed, feed rate and cutting depth. Compared to two kindsof model test results, seek a high-precision model about specific process parameters.The result shows that, SVM method is more suitable for milling titanium surfaceroughness production.