Milling force forecasting of aeroengine blade based on BP neural network and multivariate linear regression
                                             HUANG Bing1, XU Yun1,2, LIAO Yinghua1, SHI Yan1,2, LI Zhirong1
(1.School of Mechanical Engineering，Sichuan University of Science & Engineering, Yibin 644000, China;2.Sichuan Key University Laboratory of Process Equipment and Control Engineering, Yibin 644000, China)

Abstract: Aiming at the problem of milling force change prediction and control about the aeroengine blade during the mechanical processing, by studying the main process parameters of milling machining system, the orthogonal test method was used to determine several sets of process parameter schemes, the twodimensional milling model of AdvantEdge FEM blade was established, and the simulation test of blade milling was completed. The milling force of simulation test was extracted, by using the variance analysis to determine the strength of the influence of different process parameters on the change of milling force. Then, by using the strong influential parameters to design contrast test, the BP neural network milling force prediction model and the multivariate linear regression milling force prediction model were established, and the prediction ability of the two prediction models on the milling force was compared and analyzed. The research results indicate that the influence of milling depth on the change of milling force is the strongest, while the others are weaker. The prediction accuracy and stability of BP neural network are better than that of multiple linear regression on the whole.

Key words: aeroengine blade; milling; milling process parameters; predicting milling force