Piecewise linear representation-based single-variable alarm system performance evaluation method and system
An alarm system, piecewise linear technology, applied in the direction of alarm, resources, office automation, etc., can solve the problems of large error in performance indicators, affecting the effectiveness and accuracy of the alarm system, and not reflecting the performance of the alarm system well
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0041] Such as figure 1 As shown, Embodiment 1 of the present disclosure provides a method for evaluating the performance of a univariate alarm system based on a piecewise linear representation, comprising the following steps:
[0042] Given a predetermined number K of data segments, cluster adjacent data samples with the same trend to obtain historical data for univariate x(t) A piecewise linear representation of , where t∈[1,T];
[0043] Through hypothesis testing, judge the kth data segment is the normal working condition data segment or the abnormal working condition data segment, where k∈[1,K];
[0044] After dividing all data segments into normal, abnormal and undefined data segments, the performance index of the alarm system is calculated according to the division results.
[0045] Cluster adjacent data samples with the same trend, specifically:
[0046] Using piecewise linear representations (PLRs) to convert historical data into Divided into several short time...
Embodiment 2
[0107] Embodiment 2 of the present disclosure provides a univariate alarm system performance evaluation system based on piecewise linear representation, including:
[0108] The data acquisition module is configured to: acquire the alarm history data of the alarm system;
[0109] The data segmentation module is configured to: cluster adjacent data samples with the same trend according to the number of preset data segments to obtain a segmented linear representation of historical data;
[0110] The data segment classification module is configured to: judge the attributes of each data segment through hypothesis testing, including at least two attributes of the normal working condition data segment and the abnormal working condition data segment;
[0111] The performance evaluation module is configured to: calculate each performance index of the alarm system according to the obtained attributes of each data segment.
[0112] The working method of the evaluation system is the same...
Embodiment 3
[0114] Embodiment 3 of the present disclosure provides a medium on which a program is stored. When the program is executed by a processor, the steps in the method for evaluating the performance of a single-variable alarm system based on piecewise linear representation as described in Embodiment 1 of the present disclosure are implemented. , the steps are specifically:
[0115] Obtain the alarm history data of the alarm system;
[0116] According to the number of preset data segments, cluster adjacent data samples with the same trend to obtain a segmented linear representation of historical data;
[0117] Through hypothesis testing, determine the attributes of each data segment, including at least two attributes of the normal working condition data segment and the abnormal working condition data segment;
[0118] Each performance index of the alarm system is calculated according to the obtained attributes of each data segment.
[0119] The more detailed steps are the same as ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


