A fault diagnosis method and device for distribution network switch equipment

Abstract

The embodiment of the invention provides a fault diagnosis method and device for distribution network switch equipment, and the method comprises the steps: building a fault tree model according to thephysical structure and historical diagnosis records of the distribution network switch equipment; performing association rule mining on the historical diagnosis record of the distribution network switch equipment, and adding the mined association rule into the fault tree as a lateral connection to obtain a Bayesian network topology; wherein the association rule is composed of a plurality of defects meeting the minimum support degree and the confidence degree; and training the topology according to the historical diagnosis record to obtain a Bayesian network with determined parameters, solvingthe maximum likelihood equipment defect according to the Bayesian network, and taking the defect with the maximum conditional probability as a fault diagnosis result. According to the embodiment of the invention, the uncertainty of expert competent experience is overcome, the diagnosis result is accurate to equipment parts, and meanwhile, the dependence of manual experience can be reduced to themaximum extent.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of intelligent operation and maintenance of distribution network equipment, and more specifically, to a fault diagnosis method and device for distribution network switching equipment. Background technique [0002] In the traditional fault diagnosis method using Bayesian network, the establishment of Bayesian network topology is based on fault events, and the influence of the physical structure of the device itself on fault diagnosis is ignored. Aiming at this deficiency, some literatures have studied the transformation of fault tree based on equipment physical structure - Bayesian network. However, for some lateral connections that the fault tree fails to consider, it relies more on the experience of engineering personnel, and the views are too subjective and not comprehensive enough. [0003] When it is necessary to carry out random inspection of the distribution network switchgear, t...

AI Technical Summary

Problems solved by technology

However, for some lateral connections that the fault tree fails to consider, it relies more on the experience of engineering personnel, and the views are too subjective and not comprehensive enough

[0003] When it is necessary to carry out random inspection of the distribution network switchgear, the original monitoring method relies heavily on manual experience. When it is known that a certain function of the equipment cannot be realized, it is often necessary to manually inspect the equipment item by item according to the inspection instructions. The experience accumulated over a long working time can better deal with various faults under complex equipment conditions

The main disadvantage of this detection method is that the diagnosis efficiency is low, and the diagnosis results are often empirical items, and it is difficult to reveal knowledge such as the cause of deep-seated fault defects.

Images