Modelling and Simulation of Milling Forces Using an Arbitrary Lagrangian–Eulerian Finite Element Method and Support Vector Regression

A new method is presented to predict milling forces synthetically. Firstly, the 3D simulation model of the milling process is established using the arbitrary Lagrangian–Eulerian finite element method. And the simulated accuracy is calibrated by milling tests. Then the simulation model is taken as a virtual milling test system to replace extensive real milling experiments. Secondly, the specific cutting coefficients in the mechanistic milling forces model are identified by the support vector regression method using the training sample generated from the established virtual milling test system. Lastly, this methodology was validated by the slot milling operation of 2024‐T3 aluminum sheets. The results show that this new approach can dramatically eliminate the experimental machining work and achieve good estimation accuracy.