Research On Spatial Load Forecasting Method Based On Rank Set Pair Analysis Theory

Spatial load forecasting is a prediction of the magnitude and location of future power load in the power supply area,or a prediction of the spatial and temporal characteristics of the power load in the predicted area.Compared with the total load forecasting,the SLF is not limited to the prediction of the load value,but is more focused on the prediction of the load distribution in the power supply area.The SLF which contains both the future load value and the location information of load development,has important guiding significance for determining the the layout and scale of substations and selecting the transmission path and model of power lines.The analysis of power load characteristics is the basis of spatial load forecasting.This paper analyzes the power load characteristics from both time and space,and discusses the application of power load characteristics in the spatial load forecasting.At the same time,the classification of load under different criteria and the influence of climate,population,economy and other factors on the development and change of power load are introduced.In addition,the geographic information system is introduced in the spatial load forecasting,and the construction process of the power GIS based on MapInfo Professional 7.0 is elaborated.Since the change of power load is affected by many factors,the development of power load has both deterministic side of annual growth and the uncertain side of random fluctuation.In view of this uncertainty,this paper comprehensively considers many factors affecting the change of total power load,and determines the main factors affecting the change of total power load through correlation analysis.The credibility and blind number theory are used to determine the target years of the main factors affecting the total power load change.Using the advantages of set pair analysis theory in dealing with uncertain problems,and combining the concept of set pair analysis with the basic idea of cluster analysis,spatial load forecasting methods based on credibility and set pair analysis clustering theory is proposed.On the basis of the spatial load forecasting methods based on credibility and set pair analysis clustering theory,when the spatial load forecasting using the set pair analysis clustering theory is used,there is no uniform requirement for the quantization and classification criteria of the set element symbols.The model is not objective and scientific,which leads to the low accuracy of spatial load forecasting results.The spatial load forecasting method based on rank set pair analysis theory avoids the selection of quantization criteria by rank transformation of set elements,which makes the established rank set analysis and prediction model more scientific and reasonable,and effectively improves the prediction accuracy.Finally,the practicality and validity of the two prediction methods are verified by engineering example,and the prediction accuracy of the spatial load forecasting method based on rank set pair analysis theory is higher.