基于TTAO优化算法优化VMD和RF算法的刀具磨损预测研究OA

Measuring and accurately predicting the wear of milling tools is an important means of improving processing efficiency and reducing production costs.To improve the accuracy of tool wear prediction,the triangulation topology aggregation optimizer(TTAO)algorithm is used to optimize the variational modal decomposition(VMD)decomposition parameters K and α of cutting force signals at different tool wear stages.This is combined with a random forest(RF)algorithm to predict tool wear values.First,different wear stages are classified based on the wear mechanism of cutting tools,and VMD decomposition is performed on cutting force signals from different periods.Then,the TTAO algorithm is used for parameter optimization to improve signal decomposition accuracy.Secondly,the optimized VMD algorithm is combined with permutation entropy to denoise cutting force data from different wear stages,and wear features are extracted to construct a wear prediction training model for tool wear prediction.Finally,the predicted matrix samples are input into the random forest algorithm for tool wear identification.The results show that targeted noise reduction of cutting force data at different wear stages can effectively reduce the error in tool wear prediction,providing a new method for improving the accuracy of tool wear prediction,the error assessment metrics show a 33.14%reduction in mean absolute percentage error(eMAPE),a 39.46%decrease in mean absolute error(eMAE)and a 40.07%decline in root mean square error(eRMSE).