A medium-voltage distribution network power failure event judgment and completion method based on a support vector machine

Abstract

The invention discloses a medium-voltage distribution network power failure event judgment and completion method based on a support vector machine, and belongs to the technical field of distribution network operation. The method comprises four links, namely distribution network model construction, vector machine construction, vector machine solution analysis and fault type judgment and combination. According to the medium-voltage distribution network power failure event judgment method based on the support vector machine, the final power failure node and type of the power failure event are determined, the power failure event of missed mining and false mining of the medium-voltage distribution network is complemented, the problems of missed reporting, false reporting and false reporting inmedium-voltage and low-voltage distribution network power failure event statistics are solved, and the statistical accuracy of the power supply reliability of the distribution network is improved.

Description

technical field [0001] The invention belongs to the technical field of distribution network operation, and in particular relates to a method for judging and complementing a power outage event in a medium-voltage distribution network based on a support vector machine. Background technique [0002] The current distribution network is large in scale and wide in coverage, the equipment operating environment is relatively harsh, the quality of the collection devices is uneven, and some terminals have missed reports of power outage events, and the completeness rate of power outage user data pushed to the power quality system is less than 70%. Due to the pressure of assessment, some power companies may even make false reports. However, through the tracking and analysis of missed reports of medium-voltage power outages, it is found that the functional problems of the device itself are the most prominent, accounting for about 80%. As a result, uploading and power-on events cannot be...

AI Technical Summary

Problems solved by technology

[0002] The current distribution network is large in scale and wide in coverage, the operating environment of equipment is relatively harsh, the quality of collection devices is uneven, and some terminals have missed reports of power outage events, and the completeness rate of power outage user data pushed to the power quality system is less than 70%

In the early stage of construction, this module is mainly dedicated to integrating the data of various business systems and analyzing and completing power outage events. However, the power outage event completion rules are only based on simple logic rules and engineering experience. % user power outage means that the entire line is out of power, etc. The empirical completion method is highly dependent on the power grid topology file information, and the accuracy and intelligence of the completion still need to be further improved

Images