Fault diagnosis method based on knowledge graph technology

Abstract

A fault diagnosis method based on a knowledge graph technology introduces a knowledge graph into a train fault application scene, and constructs an association relationship among theme elements such as systems, parts and components of a product and an association relationship between faults and main elements by constructing the fault graph. Through the fault map, the mutual relation between parts, faults and the like can be clearly displayed, the influence relation of the faults on related parts can be accurately judged by further combining specific fault phenomena, the correlation of the faults can be accurately judged through the frequency of the faults, fault reasons and influence degrees of all factors are visually displayed, the influence of the faults on all dimensions of the product and the fault reasons are deeply excavated, and the potential influence of the faults on the train is effectively reduced.

Description

technical field [0001] The invention relates to the field of intelligent fault diagnosis, in particular to a fault diagnosis method based on knowledge map technology. Background technique [0002] The traditional fault diagnosis mainly relies on the mechanism model, judges the fault phenomenon of the train through real-time data, and uses big data technology to provide a series of trend analysis, composition analysis and other capabilities. This method can judge the specific fault phenomenon and has a good effect in fault detection However, it is impossible to diagnose the relationship between each fault phenomenon from the perspective of the whole vehicle. Specifically, there are the following shortcomings: the fault diagnosis range of the traditional fault analysis method is relatively one-sided, and it is difficult to realize all fault causes at the overall level of the product. The limitations of understanding and design experience make it difficult to comprehensively id...

AI Technical Summary

Problems solved by technology

[0002] The traditional fault diagnosis mainly relies on the mechanism model, judges the fault phenomenon of the train through real-time data, and uses big data technology to provide a series of trend analysis, composition analysis and other capabilities. This method can judge the specific fault phenomenon and has a good effect in fault detection However, it is impossible to diagnose the relationship between each fault phenomenon from the perspective of the whole vehicle. Specifically, there are the following shortcomings: the fault diagnosis range of the traditional fault analysis method is relatively one-sided, and it is difficult to realize all fault causes at the overall level of the product. The limitations of understanding and design experience make it difficult to comprehensively identify failure modes and deep-seated reasons; based on the analysis results of the failure analysis model, there are many random factors, and there are often deviations, and the failure data of the product's entire life cycle is scarce and scattered, which is difficult to achieve through the model System analysis, the monitoring ability and analysis ability of the existing model are limited, it is difficult to mine sporadic faults, and it is difficult to judge the relationship between the failures of various subsystems and components

Images