Improvement of the Unequal Division Shear Zone Model Based on Sensitivity Analysis and Classification of Uncertain Parameters

Inversion of relevant parameters of cutting analysis model through experimental data is an effective method to improve the prediction accuracy of the model. In view of the present situation that the prediction accuracy of the unequal division shear zone model needs to be improved and there are a large number of uncertain parameters that need to be inversed, a method of sensitivity analysis and classification of the uncertain parameters of the model is proposed to reduce the number of parameters that need to be inversed, and the model is improved. Firstly, the Sobol method is used to analyze the global sensitivity of 10 uncertain parameters such as the shear zone thickness h, Taylor–Quinney coefficient μ and JC parameters of the unequal division shear zone model. According to the analysis results, they are divided into sensitive parameters and insensitive parameters. Then, the sensitive parameters are inversed by genetic algorithm using experimental data. Finally, the sensitive parameters are substituted into the unequal division shear zone model to improve the prediction accuracy of the model. When the workpiece material is Al 6061-T6 and 42CrMo4 steel, the uncertain parameters that need to be inversed determined by this method are reduced from 10 to 4 and 3, respectively. The verification results show that the prediction accuracy of the main cutting force and feed force of the improved model for orthogonal cutting Al 6061-T6 workpiece and 42CrMo4 steel workpiece is much higher than that of the original model.