Alarm system normal and abnormal data detection method and apparatus based on mean change detection

A technology of abnormal data detection and change detection, applied in other database retrieval based on metadata, other database retrieval, electronic digital data processing, etc. performance, ensuring safe production and efficient operation, and the effect of accurate results

Active Publication Date: 2017-08-25
SHANDONG UNIV OF SCI & TECH
View PDF3 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] In the alarm system for analog quantity and alarm delayer, the alarm system plays a vital role in ensuring the safe production and efficient operation of coal-fired generating units. However, the common problem at present is that there are too many alarm signals. Sometimes hundreds of alarms are generated in just a few tens of minutes, causing operators to be unable to deal with these alarms in time

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
  • Alarm system normal and abnormal data detection method and apparatus based on mean change detection
  • Alarm system normal and abnormal data detection method and apparatus based on mean change detection
  • Alarm system normal and abnormal data detection method and apparatus based on mean change detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] Such as figure 1 Shown: A method for detecting normal and abnormal data of an alarm system based on mean value change detection, comprising the following steps:

[0043] (1) obtain process signal, described process signal is a series of warning process signals with time as coordinate;

[0044] (2) Calculate the test statistic of the process signal; the test statistic takes time as coordinates and corresponds to the alarm process signal;

[0045]The test statistic is a statistic related to the cumulative number of samples of continuous monotonous variation of the signal;

[0046] (3) Obtain the moment when the test statistic in (2) is the largest, set the first threshold, and verify whether this moment is the process signal mean value change moment by the method of hypothesis testing;

[0047] (4) Dichotomy is used to divide the process signal described in (1) using the process signal mean value change time obtained in (3);

[0048] (5) Repeat steps (2), (3), and (4) ...

Embodiment 2

[0058] Embodiment 2: the present invention also proposes a kind of alarm normal and abnormal data detection device based on mean value change detection, comprising:

[0059] an acquisition unit, configured to acquire a process signal, the process signal being a series of alarm process signals with time as the coordinate;

[0060] A calculation statistic unit is used to calculate the test statistic of the process signal; the test statistic takes time as the coordinate and corresponds to the alarm process signal;

[0061] The determination unit is used to obtain the moment when the test statistic is the largest, and verify whether the moment is the change moment of the mean value of the process signal through the method of hypothesis testing;

[0062] A process signal division unit, configured to divide the process signal described in the acquisition unit according to the mean change point obtained by the determination unit by using a dichotomy;

[0063] The alarm data identifi...

Embodiment 3

[0066] The following is the data of the process signal x:

[0067]

[0068] First, for the entire data segment x(1:3100), the test statistic U t,T The calculation result is as figure 2 As shown, such that |U t,T |The largest moment t=2600, at this time P=1.03×10 -55 , choose the probability of making the first type of error α=0.01, since P image 3 As shown, for the data segment x(1:2599), such that |U t,T |The largest moment, at this time, choose the probability of making the first type of error, we can know that the largest moment t=502, the corresponding P=1.25×10 -172 , for data segment x(2600:3100), such that U t,T |The largest moment t=2954, corresponding to P=0.9457. The comparison shows that the P value of the former is less than α, and the P value of the latter is greater than α. , so the sampling point at time t=502 is the mean change point of data segment x(1:2599), but the sampling point at time t=2954 is not the mean change point of data segment x(2600:31...

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 present invention discloses an alarm system normal and abnormal data detection method and apparatus based on mean change detection. The method comprises the following steps: obtaining a process signal, and calculating a test statistic of the continuous monotonically varying cumulative sample number of the signal; obtaining the time corresponding to the maximum value of the test statistic, and verifying whether the time is the process signal mean change time by using the hypothesis test method; using the dichotomy to identify whether all process signals are mean change signals, and recording all the mean change signals; according to the chronological order, dividing the process signal into a plurality of sub-segments with the mean change signals as the beginning and the end, and using the T test method to determine whether each of the sub-segment is the normal data segment or the abnormal data segment; and for the single analog quantity and alarm delay alarm system, obtaining the normal data and the abnormal data, so as to provide the basis for the calculation and optimization of the performance index of the alarm system, improve the performance of the alarm system, and ensure the safe production and efficient operation of the coal-fired generating units.

Description

technical field [0001] The invention relates to an alarm normal and abnormal data detection method and device based on mean value change detection. Background technique [0002] In the alarm system for analog quantity and alarm delayer, the alarm system plays a vital role in ensuring the safe production and efficient operation of coal-fired generating units. However, the common problem at present is that there are too many alarm signals. Sometimes hundreds of alarms will be generated in just tens of minutes, causing operators to be unable to deal with these alarms in time. Therefore, obtaining the normal data and abnormal data of the alarm system is very important for the calculation and optimization of the performance indicators of the alarm system, and obtaining the normal data and abnormal data of the alarm system also plays an important role in calculating the alarm probability density function, so that the alarm system can be Performance metrics calculation and optimiz...

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): G06F17/30
CPCG06F16/907
Inventor 王建东
Owner SHANDONG UNIV OF SCI & 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