Hydroelectric unit fault diagnosis method and system based on knowledge map of fusion features

Abstract

The invention discloses a fault diagnosis method and system for a hydroelectric unit based on a knowledge graph of fusion features, and belongs to the technical field of fault diagnosis. Including: extracting the structured vibration data features of the hydropower unit according to the vibration data of the hydropower unit; extracting the unstructured text data features of the hydropower unit according to various diagnosis reports of the hydropower unit; The text data features are fused with heterogeneous knowledge to obtain fusion features; according to the fusion features, a knowledge graph of hydropower unit fault diagnosis is constructed; according to the knowledge graph of hydropower unit fault diagnosis and the current state characteristics of hydropower units, the current state of hydropower units is reasoned and diagnosed. The present invention constructs a two-layer structure of hydropower unit fault diagnosis knowledge map by fusing structured data and unstructured data, which not only ensures the rational and full utilization of text and data knowledge, but also effectively guarantees that under the condition of data update The stability of the map.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of hydroelectric units, and more particularly relates to a method and system for fault diagnosis of hydroelectric units based on a knowledge map of fusion features. Background technique [0002] Whether the operation status of the hydroelectric generating set is safe and reliable is directly related to whether the hydropower station can provide reliable power safely and economically, and is also directly related to the safety of the hydropower station itself. With the development of science and technology, the fault diagnosis of hydroelectric generator sets is gradually developing from manual diagnosis to intelligent diagnosis, from offline diagnosis to online diagnosis, and from on-site diagnosis to remote diagnosis. [0003] The core of fault diagnosis is feature extraction, and classifiers are used to classify faults after feature extraction. At present, there are mainly fault tree fau...

AI Technical Summary

Problems solved by technology

[0005] In view of the defects of the prior art, the purpose of the present invention is to solve the technical problems of difficulty in fusion and utilization of various fault information, low accuracy and stability in the fault diagnosis method of the prior art

Images