Judgement method for cutter failure

Abstract

The invention discloses a judgement method for a cutter failure. The method comprises the steps that a current sensor is used for collecting a current signal of a main shaft of a processing machine table, and the current signal is repeatedly sampled with 1 s as time length, marked and labelled; through feature learning, a feature value of the current signal is extracted; according to the feature value of the current signal, current signal data is cleaned, the cleaned signal is subjected to normalization, and input data is obtained; through machine learning and depth learning methods, the relation between the current signal and cutter breaking and cracks is analyzed, the input data is used as a training sample for training, and then a cutter failure judgement model is established; the inputdata is used as a testing sample to be input into the cutter failure judgement model testing, and a cutter failure assessment result is obtained. According to the judgement method for the cutter failure, cutter wear can be effectively monitored and predicted online in real time, the cutter failure states of various related factors including cutter breaking and edge cracking are predicted in advance, hidden safety dangers in the production process are greatly reduced, and the finished product rate of products is increased.

Description

technical field [0001] The invention belongs to the technical field of processing tool detection, and in particular relates to a method for judging tool failure. Background technique [0002] The tool is subjected to two main loads during the machining process: 1) the centrifugal force generated by high-speed rotation; 2) the cutting force generated by contact with the workpiece. As the machining time increases during the machining process, the degree of wear of the tool itself also increases. When the amount of wear reaches a certain level, serious wear phenomena such as broken knife and edge chipping may even occur, which will lead to tool failure. In the absence of effective tool wear monitoring and prediction methods, tool wear will not only increase the defective rate of production products, but also cause hidden dangers to safe production. At present, manufacturing companies can only find various factors related to tool breakage and edge chipping through daily experi...

AI Technical Summary

Problems solved by technology

When the amount of wear reaches a certain level, there may even be serious wear phenomena such as broken knife and edge chipping, which will lead to the failure of the knife

In the absence of effective tool wear monitoring and prediction methods, tool wear will not only increase the defective rate of production products, but also cause hidden dangers to safe production

At present, manufacturing companies can only find various factors related to tool breakage and edge chipping through daily experience, which leads to the problem of inaccurate judgment

Images