A Kernel Density Estimation Method for Determining the Maximum Cell Load in Spatial Load Prediction

Abstract

A kernel density estimation method for determining the maximum value of cell load in spatial load forecasting, which is characterized in that: in the power geographic information system environment, cells are generated according to the power supply range of 10kV feeders and the cell load is obtained, and the load fluctuations of each cell Based on the analysis of characteristics and differences, all the abnormal data in the cell load are divided into four categories, and the classification index system of the cell load with abnormal data is established; the kernel density estimation curve of the cell load with abnormal data is calculated , to find the different characteristics of the cell load kernel density estimation curve containing various abnormal data, and then propose the idea of ​​obtaining a reasonable maximum value of the cell load by truncating the tail of the kernel density estimation curve; for the cell load kernel density of different types of abnormal data Estimating the difference between the curves, constructing the calculation model of the follow-up threshold required for the truncation of various cell load kernel density estimation curves containing abnormal data, and taking the load value at the truncation point as the reasonable value of the corresponding cell load maximum value.

Description

technical field [0001] The invention relates to the field of space load forecasting in urban distribution network planning, and relates to a kernel density estimation method for determining the maximum value of cell load in space load forecasting. Background technique [0002] In order to realize spatial load forecasting, it is necessary to divide the area to be predicted into multiple regular or irregular sub-districts, each sub-district is called a cell, and the electric load in a cell is called a cell load. In the process of realizing space load prediction, generally only one maximum value among many load data per unit time of each cell is used, and the cell load data collected through the SCADA system usually contains many abnormal data, which will affect the If the maximum value of the cell load is directly found from the collected cell load data for spatial load prediction, the error of the prediction result will inevitably increase. Therefore, eliminating the abnorma...

AI Technical Summary

Problems solved by technology

In the process of realizing space load prediction, generally only one maximum value among many load data per unit time of each cell is used, and the cell load data collected through the SCADA system usually contains many abnormal data, which will affect the If the maximum value of the cell load is directly searched for the spatial load prediction from the collected cell load data, the error of the prediction result will inevitably increase.

Images