Tool wear amount prediction method based on self-attention mechanism and depth learning

A technology of tool wear and deep learning, applied in the direction of neural learning methods, manufacturing tools, measuring/indicating equipment, etc., can solve problems such as the ability to process sequence data to be improved, and the prediction accuracy of tool wear is not high, so as to improve real-time prediction effect, Improve prediction effect, effect of improved ability

Active Publication Date: 2019-10-22
ZHEJIANG UNIV
View PDF5 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method has problems such as the ability to process sequence data ne

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Tool wear amount prediction method based on self-attention mechanism and depth learning
  • Tool wear amount prediction method based on self-attention mechanism and depth learning
  • Tool wear amount prediction method based on self-attention mechanism and depth learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0041]The PHM (Fault Diagnosis and Health Management) Association is an internationally influential organization spanning academia and industry, dedicated to the research and application of the theory and method of "Fault Prediction and Health Management". The PHM Association was established in New York, USA in 2009, and holds the "PHM Association Data Competition" every year. This competition is one of the high-level similar competitions in the world. This example uses the competition data of the 2010 PHM Association Data Competition to verify the CNC machine tool wear prediction method proposed by the present invention.

[0042] The processing parameters of the CNC machine tools are: the spindle speed is 10400rpm, the feed rate in the x-axis direction is 1555mm / min, the radial cutting depth is 0.125mm, and the axial cutting depth is 0.2mm. The sensor arrangement is: as attached figure 2 As shown, the tool 2 is installed on the rotating spindle 1, the workpiece 3 is fixed o...

specific Embodiment approach

[0045] S1. Install dynamometers, acceleration sensors and acoustic sensors on the CNC machine table, fixtures and workpieces;

[0046] Install a dynamometer between the workpiece and the fixture, install three acceleration sensors in three directions of the workpiece, and install an acoustic sensor on the machine table;

[0047] S2. Carry out a milling experiment on the tool, and collect relevant processing test data through the sensor in S1;

[0048] The cutting force of the tool in the three coordinate directions of x, y and z of the CNC machine tool can be obtained through a dynamometer, and the vibration signals of the workpiece in the three coordinate directions of x, y and z of the CNC machine tool can be obtained through three acceleration sensors. Acoustic sensors can obtain sound signals during CNC machine tool processing. That is, through a dynamometer, three acceleration sensors and an acoustic sensor, a total of 7 different processing characteristic signals are ob...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a tool wear amount prediction method based on a self-attention mechanism and depth learning. The method comprises the steps of mounting a dynamometer, an acceleration sensor and an acoustic sensor in numerical control machine tool machining, collecting cutting force, vibration signals and acoustic signals in a milling process, and measuring tool wear amount; preprocessing sensor measurement data, and forming training data with a tool wear amount label; establishing a neural network prediction model based on the self-attention mechanism and the depth learning, wherein the neural network prediction model comprises a self-attention layer, a bidirectional long-short-time memory network and a full-connection network; inputting the training data into the prediction modelto train the prediction model; and inputting test data into the trained prediction model to predict the tool wear amount in real time. According to the method, the characteristic information related to tool wear in the sensor measurement data is fully mined, the dependence relation between the sensor measurement data at different moments is extracted, and the tool wear amount can be predicted in real time.

Description

technical field [0001] The invention relates to a tool wear prediction method based on a self-attention mechanism and deep learning, and belongs to the field of tool wear prediction of CNC machine tools. Background technique [0002] With the continuous advancement of modern industrial informatization and intelligence, CNC machine tools are more and more widely used in industrial production, and the precision of cutting tools plays a decisive role in the comprehensive performance of CNC machine tools and the quality of processed parts. The wear of CNC machine tool tools will lead to a decrease in the dimensional accuracy of the machined surface, an increase in roughness, and a reduction in the quality of the machined parts. It may also lead to the scrapping of the machined parts and increase the production cost. And if the wear amount of CNC machine tools can be monitored or predicted, the tools can be replaced or maintained in time to ensure the quality of the workpiece. T...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): B23Q17/09G06N3/04G06N3/08
CPCB23Q17/0957G06N3/049G06N3/08G06N3/045
Inventor 刘振宇张朔刘惠郏维强谭建荣
Owner ZHEJIANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap