Optimization method of distribution line route variation model based on ReliefF and t-SNE

Abstract

The invention provides a method based on an optimization method of distribution line route variation model based on ReliefF and t-SNE. The method firstly collects the daily input power quantity of thedistribution line and the daily power consumption of each transformer connected with the distribution line, and then obtains the daily line loss power quantity, and secondly calculates the input-power quantity difference ratio according to the relationship between the input power quantity of the distribution line and the power consumption of the transformer. The line loss-power quantity difference ratio is calculated from the relationship between the line loss of distribution lines and the power consumption of transformers, which form sequences and are subjected to wavelet packet decomposition and reconstruction, and the percentage of energy to total energy in each frequency band is calculated as the primary feature. Then the ReliefF algorithm is used to calculate the weight of each primary feature. The SNE method reduces the high-dimensional feature to the low-dimensional feature. Finally, the BP neural network is used to optimize the linear relationship model. This method can reducethe workload of manual site survey, and realize the optimization of linear relationship model by selecting key features and reducing feature dimensions.

Description

technical field [0001] The present application relates to the field of electric power technology, in particular to a method for optimizing a distribution line variation relationship model based on ReliefF and t-SNE. Background technique [0002] In order to further improve the efficiency and benefits of power grid production and operation management, optimize the line loss management process, promote the integration of dispatching, operation and maintenance, and marketing business, and make full use of the system data resources of each department, to achieve the goal of "optimal technical line loss and minimum management line loss" Target. In the process of line loss management in the same period, due to the small number of lines with voltage levels of 35kV and above, simple topological structure, few changes, high-quality and timely update of historical ledger data, the level of line loss management is relatively high, and the corresponding line loss The assessment indicat...

AI Technical Summary

Problems solved by technology

Due to the large number of 10kV lines and the transformers connected to them, the traditional carpet-type on-site search requires a lot of manpower and material resources, and the timeliness is low. This is also the key reason for the slow increase in the line loss compliance rate of 10kV sub-lines during the same period.

Images