基于配用电信息分区分类的短期空间负荷预测

Short-term Spatial Load Forecasting Based on Partition and Classification of Power Distribution Information

精确负荷预测是实现电力系统安全经济运行和电网科学管理的基础。首先提出一种基于分区分类的空间负荷预测方法,该方法依托GIS系统中的地理空间信息和电网信息,进行区域划分;然后将区域内负荷分为工业、商业、居民、其他等类型,基于配用电历史信息,采用模糊聚类方法对区域内的负荷类型进行归类,获得影响负荷的主要因素;再以不同类型的负荷预测为基础,考虑跨空间和非跨空间负荷转移等情况,获得精细化的空间负荷预测方法;最后通过我国南方某实际电网的算例分析,验证了该方法较传统的空间负荷预测方法在预测精度上有较大提升。

Accurate load forecasting is the basis for the realization of safe and economic operation of power system and the scientific management of power grid. In the paper , a spatial load forecasting ( SLF ) method based on partition and classification is proposed , which relies on the geospatial information and grid information in the geographic information system ( GIS ) to carry out the regional partition. The power load in one area is classified into industrial , commercial , residential , and other types , and a fuzzy clustering method is used to classify the load in this area based on the historical information of power distribution , thus the main factors affecting load are obtained. Considering the trans-spatial and non-trans-spatial load transfer , a SLF method with higher accuracy is obtained based on the prediction of different types of load. Finally , through the analysis of a numerical example in one actual power grid of southern China , it is verified that compared with the traditional SLF method , the forecasting accuracy is improved obviously.