Unlock instant, AI-driven research and patent intelligence for your innovation.

Self-adaptive anomaly detection system applied to industrial Internet

An industrial Internet, anomaly detection technology, applied in the field of adaptive anomaly detection system, can solve problems such as economic loss and high work pressure, and achieve the effect of strong versatility, advanced solutions, and strong universality

Inactive Publication Date: 2020-08-14
GUANGDONG UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In addition, operation and maintenance engineers are engaged in emergency rescue and disaster relief 7 x 24 hours a year, 365 days a year, and the work pressure is extremely high
Moreover, if the operation and maintenance personnel cannot troubleshoot the fault in time, the economic loss caused by equipment failure will exceed 100w / production line. For example, the economic loss caused by the downtime of a cloud service provider for less than 30 minutes is more than 100 million

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
  • Self-adaptive anomaly detection system applied to industrial Internet

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0020] Such as figure 1 As shown, this embodiment discloses an adaptive anomaly detection system applied to the industrial Internet. The detection system mainly includes a machine learning algorithm module connected and operated in sequence, an intelligent operation and maintenance algorithm layer module, an intelligent operation and maintenance technology layer module, And intelligent operation and maintenance application layer module.

[0021] Specifically, the intelligent operation and maintenance algorithm layer module includes an index clustering unit, an index distribution prediction unit, a similar curve identification unit, and an abnormal interval comparison unit that are sequentially connected and operated. The intelligent operation and maintenance algorithm layer module receives the processing results of the machine learning algorithm module and imports them into the indicator clustering unit.

[0022] Specifically, the intelligent operation and maintenance technol...

Embodiment 2

[0029] This embodiment discloses an adaptive anomaly detection system applied in the field of industrial Internet. Flexible technology and program adjustment, with strong versatility.

[0030] In view of this, the present invention proposes an adaptive anomaly detection system applied in the field of industrial Internet. Including machine learning algorithm, intelligent operation and maintenance algorithm layer, intelligent operation and maintenance technology layer and intelligent operation and maintenance application layer.

[0031] Preferably, the machine learning algorithm generally refers to relatively common algorithms in the field of industrial manufacturing, such as ARIMA, time series data decomposition, and the like.

[0032] Preferably, the intelligent operation and maintenance algorithm layer includes index clustering, index distribution prediction, similarity curve identification, and abnormal interval comparison.

[0033] Preferably, the intelligent operation an...

Embodiment 3

[0036] This embodiment discloses an adaptive anomaly detection system applied in the field of industrial Internet. Such as figure 1 , including machine learning algorithm, intelligent operation and maintenance algorithm layer, intelligent operation and maintenance technology layer, and intelligent operation and maintenance application layer. The operation and maintenance algorithm layer includes index clustering, index distribution prediction, similar curve identification, and abnormal interval comparison, and the intelligent operation and maintenance technology layer includes unsupervised anomaly detection, supervised anomaly detection, automatic algorithm selection and parameter adjustment, index auxiliary labeling, multi- Index classification, great change adaptation, and extreme threshold configuration. The intelligent operation and maintenance application layer includes anomaly detection, anomaly location, root cause analysis, and anomaly prediction.

[0037] In this te...

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 self-adaptive anomaly detection system applied to the industrial Internet. The system comprises a machine learning algorithm, an intelligent operation and maintenance algorithm layer, an intelligent operation and maintenance technology layer and an intelligent operation and maintenance application layer. The machine learning algorithm generally refers to a universal algorithm in the field of industrial manufacturing, such as ARIMA, time series data decomposition and the like; the intelligent operation and maintenance algorithm layer comprises index clustering, index distribution prediction, similar curve identification and abnormal interval comparison, and the intelligent operation and maintenance technology layer comprises unsupervised anomaly detection, supervised anomaly detection, algorithm automatic selection and parameter adjustment, index auxiliary labeling, multi-index classification, giant change adaptation and limit threshold configuration. The intelligent operation and maintenance application layer comprises anomaly detection, anomaly positioning, root cause analysis and anomaly prediction. The system is advanced in solution and high in universality, can adapt to various systems in the boundary, can perform flexible technical and scheme adjustment based on different characteristics and conditions of the systems, and has very high universality.

Description

technical field [0001] The invention relates to the technical fields of industrial Internet, machine learning, and artificial intelligence, and in particular to an adaptive anomaly detection system applied to the industrial Internet. Background technique [0002] The role of operation and maintenance in the future production and life of human beings will become more and more important. It is estimated that there will be 50 billion to 100 billion networked devices in the world by 2020, and at the same time, the amount of data will increase. Covering all aspects of production and life such as Internet, finance, Internet of Things, intelligent manufacturing, telecommunications, power network, government, etc. [0003] In addition, the operation and maintenance engineers are engaged in emergency rescue and disaster relief 7 x 24 hours a year, 365 days a year, and their work pressure is extremely high. Moreover, if the operation and maintenance personnel cannot troubleshoot the ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): H04L12/26H04L12/24H04L29/06
CPCH04L41/0677H04L43/0823H04L43/50H04L63/1425
Inventor 石林欧晓龙邓富荣杨益枘黄思扬
Owner GUANGDONG UNIV OF TECH