Cutter life prediction method based on cooperation of monitoring data and degradation model

A tool life prediction method based on cooperation of monitoring data and a degradation model comprises the following steps: firstly, acquiring historical monitoring data of a tool multi-source sensor, and preprocessing the historical monitoring data by using a maximum-minimum normalization method; secondly, fusing multi-source sensing monitoring data of the cutter by adopting a graph convolutional neural network method, and constructing a composite health index of the cutter to represent degradation characteristics of the cutter; secondly, modeling a time-varying evolution trend of the composite health index based on a Wiener process, particularly, considering the problem that degradation tracks of different individuals of the same type of cutters are inconsistent, describing a drift coefficient of the cutter as a random variable to characterize the uncertainty of the cutter, and estimating model parameters by using intermediate evolution process data; and finally, dynamically optimizing and adjusting a fusi