LS-SVM-based method for detecting anomaly slot of sensor detection data

A technology for detecting data and detection methods, which is applied in electrical digital data processing, special data processing applications, instruments, etc., and can solve problems such as difficulty in judging short-term trends or pattern changes or anomalies in time series

Active Publication Date: 2014-11-19
HARBIN INST OF TECH
View PDF2 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing single test point anomaly det

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
  • LS-SVM-based method for detecting anomaly slot of sensor detection data
  • LS-SVM-based method for detecting anomaly slot of sensor detection data
  • LS-SVM-based method for detecting anomaly slot of sensor detection data

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0018] Specific implementation mode one: see figure 1 Describe this embodiment, a kind of sensor detection data abnormal segment detection method based on LS-SVM described in this embodiment, it comprises the following steps:

[0019] Step 1. Set the required detection confidence c, the time series length n and the minimum number of abnormal points m, and the settings of n and m meet the limits of the confidence c, and both n and m are positive integers;

[0020] Step two, from t 0 From time to time, point anomaly detection is performed on the data within the time series length n, and the LS-SVM point anomaly detection with a confidence probability of p is obtained to obtain the number of prediction residuals and data abnormal points within the time series length n;

[0021] Step 3. Determine whether the number of abnormal points in the time series of length n in step 2 is at least m abnormal points, that is, |E n (t 0 )|≥m, m is a positive integer, |E n (t 0 )|represents...

specific Embodiment approach 2

[0030] Specific embodiment 2: This embodiment is a further description of the method for detecting abnormal fragments of sensor detection data based on LS-SVM described in specific embodiment 1. In this embodiment, in step 1, the time series length n, minimum abnormal The number m of points and the degree of confidence c, and the relationship between the three satisfies the following formula:

[0031] P(|E n (t 0 )|≥m)>c,

[0032] Among them, in a time series segment of length n, the probability P(|E n (t 0 )|≥m) is expressed as P(|E n (t 0 )|≥m)=P(m)+P(m+1)+…+P(n), P(m) is the probability of m outliers appearing in the time series length of n, P(|E n (t 0 )|) is the occurrence of |E ​​in a time series segment of length n n (t 0 )|the probability of an outlier, P ( | E n ( t 0 ) | ...

specific Embodiment approach 3

[0035] Specific embodiment three, this embodiment is a kind of LS-SVM-based sensor detection data abnormal segment detection method described in specific embodiment one to further explain, in this embodiment, step two from t 0 From time to time, point anomaly detection is performed on the data within the time series length n with the LS-SVM point anomaly detection with a confidence probability of p. The method to obtain the number of abnormal points of the data within the time series length n is:

[0036] Step 21: Set the training data set, perform phase space reconstruction on the data in the training data set, and obtain the input vector and output vector;

[0037] Step 22: Using the Z-zeros method to normalize the obtained input vector and output vector, and normalize the input vector and output vector to a range of [-1, 1];

[0038] Step two and three: select the kernel function of the LS-SVM algorithm, and set the parameters of the LS-SVM prediction model, and train the L...

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 LS-SVM-based method for detecting an anomaly slot of sensor detection data, and relates to the field of anomaly detection of spacecraft monitoring data. The LS-SVM-based method aims at solving the problem that a short-time trend appearing in a time series or changes or anomalies appearing in a mode can not be easily judged in an existing single-detecting-point anomaly detection mode, and includes the steps of (1) setting the needed detection fiducial probability p, the detection slot length n and the minimum number m of abnormal points in the slot; (2) carrying out LS-SVM point anomaly detection with the point anomaly detection fiducial probability p on data in the time series length n from the moment t<0>, and obtaining predication residual errors and the number of the data anomaly points in the time series length n; (3) determining the positions of the anomaly points according to the residual errors and the number of the data anomaly points. The LS-SVM-based method can be applied in the aerospace flight vehicle monitoring field.

Description

technical field [0001] The invention relates to a method for detecting abnormal fragments of sensor detection data based on LS-SVM. The invention belongs to the field of anomaly detection of spacecraft monitoring data. Background technique [0002] With the rapid development of national defense modernization and the urgent need of national security, my country's demand for various types of spacecraft continues to grow, and higher requirements are put forward for the completeness and reliability of spacecraft functions. In order to ensure the high reliability and long life of such aerospace equipment, a lot of testing work is always inseparable in the process of design, development, production, use and maintenance. Taking satellites as an example, as a large-scale multifunctional and complex system, a large amount of test data will be recorded during the entire life cycle of a satellite's birth, launch, and on-orbit maintenance. These data are often in time series, In parti...

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): G06F17/30
CPCG05B23/0254
Inventor 刘大同彭宇宋歌庞景月彭喜元
Owner HARBIN INST 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