Power distribution network power failure event precision diagnosis method, device and system based on data driving

Abstract

The invention discloses a power distribution network power failure event precision diagnosis method, device and system based on data driving. The method comprises the steps: generating a primary powerfailure fragment set of a power distribution transformer based on obtained measurement data of the power distribution transformer; calculating integral electric quantity checking indexes of the primary power failure fragments of the distribution transformer, eliminating part of unreal power failure fragments in the primary power failure fragment set, and generating a middle-stage power failure fragment set of the distribution transformer; inputting the meter bottom change amplitude of the distribution transformer corresponding to the intermediate power failure fragments of each distribution transformer into a frozen electric quantity check index calculation formula to obtain a frozen electric quantity check index set; inputting the frozen electric quantity check index set and the intermediate power failure fragment duration of each distribution transformer into a classification model, and eliminating all non-real power failure fragments in the intermediate power failure fragment set to generate an advanced power failure fragment set of the distribution transformer. Operation data change characteristics of the distribution transformer are learned in a data driving mode, whether real power failure happens to the distribution transformer is judged, and accurate collection of power failure events is achieved.

Description

Technical field[0001]The invention belongs to the field of diagnosis of power outage events in a medium-voltage distribution network, and specifically relates to a data-driven method, device and system for accurately diagnosing outage events in a distribution network.Background technique[0002]As the end link of power transmission, the distribution network is the most intuitive object for users to experience the grid service. However, the distribution network has a complex structure and numerous equipment, and faces the dilemma of low operation and maintenance level and poor reliability. In traditional distribution network power outage management work, it is often dependent on the accuracy of device collection. This management method is relatively extensive, low efficiency, and underreporting is more common. Incomplete distribution network failure and power outage information will affect the decision-making of distribution network reliability analysis, patrol inspections, and grid up...

AI Technical Summary

Problems solved by technology

[0002] As the terminal link of power transmission, the distribution network is the most intuitive object for users to experience the service of the power grid. However, the structure of the distribution network is complex and there are many equipments, and it faces the dilemma of low operation and maintenance level and poor reliability.

In the traditional distribution network power outage management work, it often relies on the accuracy of device collection. This kind of management method is relatively extensive, the work efficiency is low, and the phenomenon of missing reports is more common.

The incompleteness of distribution network failure and power outage information will affect the decision-making of distribution network reliability analysis, patrol inspection, grid upgrade, etc.

Images