Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An unsupervised industrial system anomaly detection method based on deep transfer learning

An industrial system, transfer learning technology, applied in the fields of data mining, deep learning and neural networks

Active Publication Date: 2019-05-03
GUANGDONG UNIV OF TECH
View PDF11 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, using traditional deep learning techniques to solve anomaly detection problems is relatively limited.

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
  • An unsupervised industrial system anomaly detection method based on deep transfer learning
  • An unsupervised industrial system anomaly detection method based on deep transfer learning
  • An unsupervised industrial system anomaly detection method based on deep transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] The unsupervised industrial system anomaly detection method based on deep transfer learning of the present invention comprises the following steps:

[0018] 1) Collect sensor data;

[0019] 2) Data preprocessing;

[0020] 3) Divide the dataset,

[0021] 4) Model building;

[0022] 5) Model training;

[0023] 6) Model testing;

[0024] 7) Determine whether there is a fault.

[0025] When a new machine is added to the industrial system, repeat steps 1) to 6) to get a new model, and then redeploy the model.

[0026] The above step 2) data preprocessing divides sensor data into three categories, the first category has a certain range of data; the second category is data that increases with time; the third category is data with only fixed state quantities; The data in a certain range of change is processed by normalization, which compresses the range of the training set data to [0,1]; the data that increases with time adopts differential processing, and the differentia...

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 an unsupervised industrial system anomaly detection method based on deep transfer learning. According to the invention, labeled machine sensor sequence data from a migration source and unlabeled sensor sequence data from a migration target are utilized to train an industrial system abnormality detection model with good generalization ability, and the industrial system abnormality detection model is trained and tested to finally generate a trained industrial system abnormality discrimination model. By means of the model, the received machine sensor sequence data can be analyzed, and whether the machine is abnormal or not can be judged.

Description

technical field [0001] The invention relates to the fields of data mining, deep learning, neural network and the like, and in particular to an unsupervised industrial system anomaly detection method based on deep transfer learning. Background technique [0002] In recent years, with the rapid development of deep learning technology, deep learning has blossomed in all walks of life. One of the applications in industrial systems is anomaly detection. For example, in a power plant, by collecting data from sensors such as temperature and air pressure on the boiler, an abnormality discrimination model is used to determine whether there is an abnormality; Whether there is any abnormality; in the telecommunications marketing system, by collecting server hardware load and network traffic data, it is judged whether there is any abnormality through the abnormality discrimination model. The examples mentioned above all use industrial sensors to receive data. Among them, we call the la...

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): G06F16/245G06N3/04
CPCG05B23/024G05B23/0221G06N3/084G06N3/045G05B19/4065G06N3/088
Inventor 蔡瑞初李梓健温雯郝志峰王丽娟陈炳丰许柏炎李俊峰
Owner GUANGDONG UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com