Spatial load prediction method based on a rank set pair analysis theory

Abstract

The invention discloses a spatial load prediction method based on a rank set pair analysis theory. The method is characterized by comprising the following steps: in a power geographic information system, generating I-type cells according to the power supply range of each 10kV feeder line in a to-be-predicted area, and generating a plurality of historical data sets and a target data set from the historical load data of the I-type cells according to different set capacities; obtaining a corresponding rank order set through rank order transformation, and respectively forming a set pair by each historical data rank order set and the target data rank order set; respectively carrying out set pair analysis on the set pairs under different capacities, determining a load prediction value of each I-type cell by searching a historical data set similar to a target data set, and selecting the prediction value corresponding to the set capacity with the minimum relative error as the load prediction value of each I-type cell; and generating II-type cells by using equal-size grids, and solving load prediction values of the II-type cells according to the I-type cell load prediction values and the land use information so as to obtain a gridded space load prediction result.

Description

technical field [0001] The invention relates to the field of space load forecasting in distribution network planning, and is a space load forecasting method based on rank set pair analysis theory. Background technique [0002] As the basic work of power system planning, spatial load forecasting is used to determine the capacity and optimal location of power supply equipment, which greatly improves the economy of power system construction. Spatial load forecasting is an important research topic in the field of urban power grid planning and construction. Because of its basic nature, it occupies an important position in urban power grid planning and plays a key role in planning quality. [0003] The space load forecasting methods used in the prior art usually predict directly on the basis of the annual maximum value of historical load data, and do not fully exploit the characteristics of historical load data and space power load, resulting in low prediction accuracy. However, ...

AI Technical Summary

Problems solved by technology

[0003] The space load forecasting methods used in the prior art usually predict directly on the basis of the annual maximum value of historical load data, and do not fully exploit the characteristics of historical load data and space power load, resulting in low accuracy of forecast results

However, when the existing set pair analysis theory is used to predict the space load, not only the historical data needs to be preprocessed, but also the elements in the set need to be classified during the forecasting process. Requirements, making the model results not objective and scientific enough, leading to low accuracy of prediction results

Images