Multi-dimensional time sequence abnormal value detection method and system based on MOTCN-AE

A MOTCN-AE, time series technology, applied in instruments, character and pattern recognition, computer parts, etc., can solve problems such as low accuracy, inability to meet the requirements of abnormal data detection accuracy, and inability to apply semi-supervised learning, etc. To achieve the effect of excellent precision

Pending Publication Date: 2020-04-10
QILU UNIV OF TECH
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, traditional semi-supervised learning often requires a certain amount of labeled data, and semi-supervised learning methods cannot be applied without labeled data.
However, traditional unsupervised learning methods are less accurate and cannot meet the requirements for abnormal data detection accuracy.

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
  • Multi-dimensional time sequence abnormal value detection method and system based on MOTCN-AE
  • Multi-dimensional time sequence abnormal value detection method and system based on MOTCN-AE
  • Multi-dimensional time sequence abnormal value detection method and system based on MOTCN-AE

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052]In one or more embodiments, a multidimensional time series outlier detection method based on MOTCN-AE is disclosed, referring to figure 2 , including the following procedures:

[0053] (1) Receive a set of signals in the form of time series;

[0054] (2) Enriching the time series with features;

[0055] (3) Use an autoencoder to reconstruct the time series after feature enrichment;

[0056] (4) Comparing the feature-enriched time series with the reconstructed time series to obtain abnormal data in the time series form signal.

[0057] The method of this embodiment first applies temporal convolutional networks to the field of anomaly detection, and proposes a new time series anomaly detection framework MOTCN-AE (Multidimensional time series outlier detection based onTCN-AE) in combination with autoencoders and rich time series data. The framework consists of three parts: First, a TCN-based autoencoder TCN-AE is proposed using an unsupervised method combined with a tem...

Embodiment 2

[0133] In one or more embodiments, a multidimensional time series outlier detection system based on MOTCN-AE is disclosed, including:

[0134] means for receiving a set of signals in time-series form;

[0135] means for feature enriching said time series;

[0136] Apparatus for reconstructing feature-enriched time series using an autoencoder;

[0137] A device for comparing the feature-enriched time series with the reconstructed time series to obtain abnormal data in the time series signal.

[0138] Since the device provided in this embodiment can be used to implement the time series anomaly detection method in Embodiment 1, the technical effect it can obtain can refer to the above method embodiment, and will not be repeated here.

[0139] Those of ordinary skill in the art know that all or part of the steps in the above method can be completed by program instruction related hardware, and the program can be stored in a computer-readable storage medium, such as ROM, RAM and o...

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 multi-dimensional time sequence abnormal value detection method and system based on MOTCN-AE. The method comprises the following steps: receiving a group of signals in a timesequence form; carrying out feature enrichment on the time sequence; reconstructing the time sequence with rich features by using an automatic encoder; and comparing the time sequence with rich features with the reconstructed time sequence to obtain abnormal data in the time sequence form signal. The automatic encoder TCN-AE provided by the invention is more suitable for time sequence modeling; the time series feature enrichment method provided by the invention can well improve the prediction precision of an algorithm framework.

Description

technical field [0001] The present invention relates to the technical field of multidimensional time series abnormal value detection, in particular to a multidimensional time series abnormal value detection method and system based on MOTCN-AE (time series abnormal value detection framework). Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] With the development of IoT technology, the world today produces more and more complex systems. Monitoring the behavior of these systems generates large amounts of multidimensional time-series data, for example: sensor parameters (temperature and pressure, etc.) I / O) and so on. The key task of managing these systems is how to detect abnormal data in a specific time series, so that operators can take actions to solve potential problems of the system. For example, in car driving data, the speed, accelera...

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): G06K9/62
CPCG06F18/2155G06F18/22G06F18/2433
Inventor 姜雪松孟超尉秀梅胡大鹏朱庆存井立超李浩
Owner QILU UNIV OF TECH
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