Research on cutting force prediction in milling process based on StackNet

In order to analyze the cutting force prediction in the milling process of titanium alloy materials, a method based on multi-layer StackNet network structure is proposed to predict the cutting force. Firstly, the milling experiments of titanium alloy with different speed, feed rate and axial cutting depth are carried out on VMC600 machining center, and the collected experimental data are predicted and analyzed by StackNet. Secondly, the single model and the 2-layer StackNet structure of the two different methods are used to carry out prediction analysis. In the case of clarifying the prediction advantages of two-layer StackNet, we conducted the research of StackNet with different layers furtherly. The results of the prediction experiments show that the prediction effect of multi-layer StackNet structure is improved by 78.19%. When the number of network layers is 6, the prediction accuracy is the best. The multi-layer StackNet structure is used to predict the change of cutting force in the process of aluminum alloy milling, which provides an important reference for the rational formulation of machining technology.