Tool condition monitoring method based on ELM-SDAE algorithm

Abstract

The invention belongs to the technical field of tool condition monitoring, and provides a tool condition monitoring method based on an ELM-SDAE algorithm. Firstly, vibration signals in the process ofmachining are collected by a three-way acceleration sensor; then, missing values are interpolated by adopting a linear interpolation method, a naive random undersampling and white noise adding methodis adopted to balance data, and effective values of a main vibration direction are obtained by vector superposition of the vibration signals; then the ELM-SDAE network is constructed, and a network model is trained by adopting a greedy layer-by-layer method; and finally, real-time vibration in the actual machining process is input into the ELM-SDAE network after data preprocessing, the network outputs the current status of a tool, and real-time monitoring of the status of the tool is achieved. The method reduces participation of human and expert experience, avoids the situation that the training process may fall into local optimum, slow learning rate and so on, which can lead to failure of model training and lack of generalization ability, and greatly reduces the training time of the network.

Description

technical field [0001] The invention belongs to the technical field of tool state monitoring, and specifically relates to a tool state monitoring method based on an extreme learning machine-stacked denoising autoencoder (Extreme Learning Machine-Stacked Denoising Autoencoders, ELM-SDAE) algorithm. Background technique [0002] In the field of machining, the state of the tool directly affects the machining accuracy and surface quality of the machined parts. In conventional cutting machining, the wear condition is generally estimated by experienced machining personnel based on machining noise, chip color, cutting vibration and cutting time of the tool. However, due to the limitations of workers’ experience, it is impossible to accurately judge whether the tool is worn or not. It may happen that the worn tool continues to process, which reduces the processing quality; it may also happen that the unworn tool is replaced in advance, causing waste and improving the quality of the ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a tool state monitoring method based on the ELM-SDAE algorithm, which solves the problems that the existing monitoring method relies on manual experience to extract signal features, takes a long time in the training process and easily falls into local optimum, and realizes the real-time monitoring of the tool state. monitor

Images