A distribution network district risk early warning method and system based on support vector data description

Abstract

The invention discloses a distribution network district risk early warning method and system based on support vector data description, and belongs to the technical field of big data mining and application. The method comprises the steps of obtaining an operation risk factor of a distribution network district as a training sample set; carrying out dimension reduction processing on the training samples in the training sample set by adopting a principal component analysis method to obtain low-dimension training samples; and on the basis of the characteristic difference of low-dimension training samples, constructing a weighted support vector data description multi-classification model, and carrying out classification identification on the operation state of the power distribution area, thereby realizing risk early warning of the power distribution area. According to the invention, the power distribution area operation risk early warning model is constructed by applying the big data technology, the risk evaluation efficiency is high, and the early warning effect is good.

Description

technical field [0001] The invention relates to the technical field of big data mining and application, in particular to a risk warning method and system for a distribution network station area based on support vector data description. Background technique [0002] With the construction and development of my country's smart grid, power supply enterprises have accumulated a large amount of distribution network operation data, but the traditional reliability evaluation methods are limited by data sources, data integration, and data processing capabilities, and it is difficult to analyze the power distribution system from multiple sources. More valuable information can be mined from structural data. [0003] Conventional distribution network reliability evaluation is mainly based on the statistical value of distribution system reliability under different long-term operating conditions, mainly based on load shedding indicators, and the evaluation methods are mainly analytical and...

AI Technical Summary

Problems solved by technology

[0002] With the construction and development of my country's smart grid, power supply enterprises have accumulated a large amount of distribution network operation data, but the traditional reliability evaluation methods are limited by data sources, data integration, and data processing capabilities, and it is difficult to analyze the power distribution system from multiple sources. Mining more valuable information from structural data

[0003] Conventional distribution network reliability evaluation is mainly based on the statistical value of distribution system reliability under different long-term operating conditions, mainly based on load shedding indicators, and the evaluation methods are mainly analytical and simulation methods, and the results are mainly used for distribution network planning Although the probability and consequences of failures can be considered in the design, the statistical cycle is long and the time scale is large, usually months or years, so it is difficult to be directly applied to the short-term reliability assessment and real-time operation scheduling control of distribution network stations.

[0004] In addition, the replacement or decommissioning of transformer equipment in the distribution network area is mainly determined based on a small number of factors such as the distribution transformer model, operating life, and historical maximum load, or the cost of the entire life cycle, which lacks objective evaluation.

Images