Cutter residual life prediction method based on deep learning and time sequence regression model

Abstract

The invention discloses a cutter residual life prediction method based on deep learning and a time sequence regression model, solves the problem that the existing cutter residual life prediction precision is not high, and belongs to the field of numerical control cutter predictive maintenance. The method comprises the steps that tool vibration, cutting force and acoustic emission signals of each channel sampling point in the machining process are collected, and four-dimensional data including the mean value, the variance, the skewness and the kurtosis are calculated; inputting the four-dimensional data into a deep convolutional residual neural network tool wear monitoring model, and outputting a tool wear value; and smoothing the tool wear value, and outputting a wear sequence. Inputting the residual life into a difference integration moving average autoregressive tool wear advanced prediction model, predicting a tool wear value of N advanced steps, and predicting the residual life at the current moment as the maximum residual life value when the value of the Nth step does not reach a threshold value; when the Mth step value reaches or exceeds the threshold value, the remaining life at the current moment is predicted as M-1 cutting strokes, and M is smaller than or equal to N.

Description

technical field [0001] The invention relates to a tool residual life prediction method based on deep learning and time series regression model, and belongs to the field of numerical control tool predictive maintenance. Background technique [0002] In machining, the performance and quality of tools directly affect the production efficiency and processing quality of millions of machine tools, which indirectly affects the production technology level and economic benefits of the entire manufacturing field. When the tool is used for more than its life, the tool will be wasted, increase the manufacturing cost, and reduce the productivity. When the tool is used for more than its life, the machining accuracy of the parts will be reduced, the productivity will be reduced, and the machine tool will be damaged, resulting in major equipment losses. . Therefore, from the perspective of reducing the production cost of enterprises or ensuring the machining accuracy of parts and equipment...

AI Technical Summary

Problems solved by technology

[0008] Aiming at the problem that the existing tool remaining life prediction relies on specific prior knowledge and professional knowledge, which easily leads to low prediction accuracy, the present invention provides a tool remaining life prediction method based on deep learning and time series regression model

Images