Title:
HYDROELECTRIC SIGNAL ABNORMALITY DETERMINATION METHOD BASED ON FREQUENT ITEM SET REASONING
Document Type and Number:
WIPO Patent Application WO/2024/093533
Kind Code:
A1
Abstract:
The present invention relates to the technical field of hydroelectric monitoring. Disclosed is a hydroelectric signal abnormality determination method based on frequent item set reasoning. The present invention comprises: acquiring original data, extracting signal data and action data from the original data and preprocessing same, and constructing signal-action pairs in which the signal data and the action data are combined; performing relationship mining on the signal-action pairs on the basis of an apriori algorithm and a prefixSpan algorithm, so as to acquire an association relationship set, a coexistence relationship set and a causal relationship set; and testing, according to the association relationship set, the coexistence relationship set and the causal relationship set, a signal stream formed by newly arriving continuous signals, determining whether there is an abnormality in the signal stream, and labelling a signal interval in which the abnormality is present. According to the present invention, signal relationships are defined from the perspective of a frequent mode, a relationship mining method is proposed on the basis of an apriori algorithm and a prefixSpan algorithm, signal pairs corresponding to an association relationship, a coexistence relationship and a causal relationship are obtained, and automatic abnormality determination is performed on incoming log signals on the basis of signal relationship pairs, thereby improving the abnormality detection efficiency.
Inventors:
DONG SHILEI (CN)
HUANG FEIHU (CN)
SONG WEIPING (CN)
HU ZHOUMING (CN)
LI YU (CN)
YANG FAN (CN)
DING XI (CN)
GAO PAN (CN)
ZHAO HONGLEI (CN)
WU HONGYU (CN)
TIAN PAN (CN)
HUANG FEIHU (CN)
SONG WEIPING (CN)
HU ZHOUMING (CN)
LI YU (CN)
YANG FAN (CN)
DING XI (CN)
GAO PAN (CN)
ZHAO HONGLEI (CN)
WU HONGYU (CN)
TIAN PAN (CN)
Application Number:
PCT/CN2023/118155
Publication Date:
May 10, 2024
Filing Date:
September 12, 2023
Export Citation:
Assignee:
AOSTAR INF TECH CO LTD (CN)
International Classes:
G06N5/04; G06F18/20
Foreign References:
| CN115470831A | 2022-12-13 | |||
| CN112183656A | 2021-01-05 | |||
| CN112888008A | 2021-06-01 | |||
| CN110245168A | 2019-09-17 | |||
| US20170293670A1 | 2017-10-12 |
Attorney, Agent or Firm:
SCIHEAD IP LAW FIRM (CN)
Download PDF: