Distributed power distribution network electricity price adjusting method based on convolutional network and collaborative filtering

Abstract

The invention provides a distributed power distribution network electricity price adjustment method based on a convolutional network and collaborative filtering. The method comprises the steps of 1, performing feature extraction according to power distribution network data; 2, constructing a scoring matrix; 3, singular value decomposition; 4, calculating the similarity degree; 5, confirming a to-be-recommended user; and 6, selecting recommended users according to the sequence. According to the method, output prediction of the power distribution network containing the distributed power supplies is realized by adopting a convolutional neural network method, data feature extraction is realized, output abundant time periods of the power distribution network containing the distributed power supplies are predicted, and electricity price adjustment is performed by utilizing a prediction result. The electricity price preferential information is pushed to the user by adopting a collaborative filtering method based on singular value decomposition and a model, and the singular value decomposition method is wide in application range and suitable for large-scale data; according to the model-based collaborative filtering method, collaborative filtering performance can be improved, and screening and sorting are more accurate. And the final recommended user is determined according to the similarity ranking between the user and the target user, so that the unnecessary information spreading cost is reduced.

Description

technical field [0001] The invention belongs to the technical field of power systems and automation thereof, and relates to a software calculation method for electricity price adjustment, in particular to a distributed distribution network electricity price adjustment method based on convolutional networks and collaborative filtering. Background technique [0002] With the shortage of international energy supply and environmental changes caused by air pollution, renewable energy has received increasing attention, among which clean energy represented by wind power and photovoltaics has developed rapidly. However, distributed energy is affected by weather conditions and geographical environment, and the power output is random, so it is difficult to control, which brings volatility to the distribution network, and the output of distributed energy cannot be highly matched with the peak hours of power consumption on the user side. Therefore, it is a good solution to use energy st...

AI Technical Summary

Problems solved by technology

The limitation of the existing distribution network electricity price adjustment mechanism is that the electricity price mechanism based on load demand cannot give positive feedback to the power grid output, resulting in a weak connection between the two

Images