基于配用电信息分区分类的短期空间负荷预测

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。

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

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.