Milling cutter wear state real-time monitoring method based on deep convolutional neural network

Abstract

The invention relates to a milling cutter wear state real-time monitoring method based on a deep convolutional neural network, and belongs to the technical field of automatic monitoring and recognition. The milling cutter wear state real-time monitoring method comprises the following steps: collecting multi-source heterogeneous state data generated when a numerical control machine tool mills a workpiece and corresponding milling tool full life cycle abrasion data, and preprocessing and labelling the data; constructing a deep learning network, and achieving accurate regression prediction of a tool abrasion loss result; constructing a deep convolutional neural network, and achieving effective recognition of the tool abrasion loss critical state; and comparing the real-time tool abrasion loss with a corresponding critical state threshold value, and timely taking measures such as tool replacement or parameter change and the like to realize real-time monitoring of the tool abrasion state. The method has the advantages of being complete in network model input signal source, fine in cutter wear critical state category division and high in model prediction accuracy, can accurately monitor the milling cutter wear state in real time, avoids cutter milling in an abnormal state, and guarantees the controllable quality of a machined product.

Description

technical field [0001] The invention relates to a real-time monitoring method for milling tool wear of a CNC machine tool based on a deep convolutional neural network, relates to the technical field of monitoring the wear amount of a CNC machine tool milling tool and identifying the critical state of tool wear, and belongs to the technical field of automatic monitoring and identification. Background technique [0002] In the milling process of CNC machine tools, the milling tool will inevitably wear out, and the wear of the tool will lead to low machining accuracy of the workpiece and unqualified product quality. In order to achieve the machining accuracy required by the product, it is necessary to monitor the wear amount of the milling tool in real time and accurately identify the critical state threshold of the tool wear, so as to detect the abnormal state of the tool in time and take preventive measures to effectively improve the product qualification rate. [0003] The e...

AI Technical Summary

Problems solved by technology

[0004] In view of the above technical needs and problems, the present invention aims to provide a real-time monitoring method for milling tool wear of CNC machine tools based on deep convolutional neural network, which can avoid the prediction of tool wear value caused by single-state signal source input and unreasonable model parameter setting The accuracy rate is not high, and the deviation of the predicted critical state is large, etc.

Images