Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD FOR PREDICTING RESIDUAL LIFE OF NUMERICAL CONTROL MACHINE TOOL BASED ON HYBRID NEURAL MODEL
Document Type and Number:
WIPO Patent Application WO/2022/268043
Kind Code:
A1
Abstract:
The present invention relates to the field of life prediction of numerical control machine tools, and in particular, to a method for predicting the residual life of a numerical control machine tool based on a hybrid neural model, comprising: constructing a hybrid neural network model, constructing, for a tool data sampling frequency, a sample of PLC working condition signal data and vibration and current signals in a same time period, and generating one piece of sample data; using a combined sub-model of a convolutional neural network and a long short-term memory network to learn the sample data to obtain a first feature vector representing the tool life; using an NFM neural network to hash a working condition signal of a sampling point into a unique index value, then generating a learnable vector table having a specific dimension, and learning to obtain a second feature vector representing the tool life; and inputting a current working duration of a tool and the obtained feature vectors to a multilayer perceptron for fusion, so as to predict the tool life. The present invention can effectively and accurately predict the tool life according to sparse features in the tool.

Inventors:
HUANG QINGQING (CN)
HAN YAN (CN)
KANG ZHEN (CN)
ZHANG YAN (CN)
WANG PING (CN)
Application Number:
PCT/CN2022/099990
Publication Date:
December 29, 2022
Filing Date:
June 21, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
INSTITUTE OF INDUSTRIAL INTERNET CHONGQING UNIV OF POSTS AND TELECOMMUNIC (CN)
International Classes:
B23Q17/09
Foreign References:
CN113305645A2021-08-27
CN110472800A2019-11-19
CN112631128A2021-04-09
CN110153802A2019-08-23
US8781982B12014-07-15
Attorney, Agent or Firm:
CHONGQING PHAETON LAW FIRM (CN)
Download PDF: