Circuit load regression analysis and prediction method and system

Abstract

The invention discloses a circuit load regression analysis and prediction method and system. The method refers to a method for circuit load prediction based on load data of users mounted below a circuit; through network frame topology analysis in a power system, a district transformer and a large-customer transformer mounted below the circuit are subjected to related analysis; feature vectors of a training set and a test set are established by collecting operating data and external data (meteorological data, special event data and the like) of the district regions and large customers; and load prediction is carried out on each district region and large customer by adopting a regression prediction method, and a regression equation of the circuit is established according to the load prediction of each district region and large customer, so as to form the prediction method for the circuit load based on user view. The circuit load is predicted through the operating data of the district regions and large customers in the circuit, so that a relation between the circuit and the mounted transformer in the power topology can be reflected effectively; and the relation between the circuit load and users thereof is displayed by multi-dimensional data in a more effective manner.

Description

[0001] Technical field: [0002] The invention relates to the field of information technology, in particular to a line load regression analysis and prediction method and system. [0003] Background technique: [0004] Line load forecasting is an important basis for formulating the operation mode of the power grid, calculating the transmission capacity of the line, and then doing a safety check of the dispatching plan and identifying the security status of the power grid. At the same time, as the basis for the follow-up formulation of generation dispatching plans considering safety constraints, line load forecasting is of great significance for realizing the safety, energy saving and economy of the entire power grid dispatching. [0005] At present, most of the line load forecasting methods are based on the historical operation data of the line itself and some external data (meteorological data, holiday information, etc.) to construct the feature vector of line prediction. This...

AI Technical Summary

Problems solved by technology

This type of method ignores the transformer equipment information mounted under the line, and cann

Images