Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
FAULT DIAGNOSIS METHOD AND SYSTEM BASED ON STANDARD SELF-LEARNING DATA ENHANCEMENT
Document Type and Number:
WIPO Patent Application WO/2024/045555
Kind Code:
A1
Abstract:
A fault diagnosis method and system based on standard self-learning data enhancement, the method comprising: constructing a fault diagnosis model on the basis of a one-dimensional convolutional neural network (S1); training the fault diagnosis model by means of a cross-adversarial training mode of standard self-learning and data enhancement, so as to obtain a complete data set and an intelligent fault diagnosis model under a strong non-stationary working condition (S2); and inputting an acquired vibration signal to be diagnosed into the trained intelligent fault diagnosis model, so as to obtain a bearing fault diagnosis result (S3). The one-dimensional convolutional neural network is taken as a basic framework, an incomplete training data set is utilized, and by means of the cross-adversarial training mode of standard self-learning and data enhancement, disturbance data is generated, and the fault diagnosis model under the strong non-stationary working condition is obtained, thereby improving the accuracy of fault diagnosis.

Inventors:
AN ZENGHUI (CN)
ZHANG YUXI (CN)
YANG RUI (CN)
WANG HOULIANG (CN)
YAN YINGLONG (CN)
Application Number:
PCT/CN2023/081715
Publication Date:
March 07, 2024
Filing Date:
March 15, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV SHANDONG JIANZHU (CN)
International Classes:
G01M13/045
Domestic Patent References:
WO2022077901A12022-04-21
Foreign References:
CN115753103A2023-03-07
CN113095413A2021-07-09
CN110608884A2019-12-24
CN111539152A2020-08-14
CN112067294A2020-12-11
CN112284735A2021-01-29
CN113203566A2021-08-03
CN114295377A2022-04-08
US20200285900A12020-09-10
US20210278478A12021-09-09
US20220269925A12022-08-25
Attorney, Agent or Firm:
JINAN SHENGDA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: