考虑地域差异的配电网空间负荷聚类及一体化预测方法

Spatial Load Clustering and Integrated Forecasting Method of Distribution Network Considering Regional Difference

针对基于智能算法的负荷密度指标法对样本依赖性强且在各地实际应用困难的不足,提出一种考虑地域差异的配电网空间负荷聚类及一体化预测方法。该方法首先通过大量调研得到分布在不同地区、分属不同类型的负荷样本及所处地区信息;然后利用基于日负荷曲线的负荷分类校验及精选方法对所有调研样本进行分类精选;再根据区域分类、负荷分类对精选样本构成的全样本空间进行两级划分,得到分层级子样本空间;最后根据待测地块的属性信息对子样本空间进行匹配,选取与其最相似的子样本空间作为训练样本,构建支持向量机模型预测各地块的负荷密度,进而得到电力负荷的空间分布。工程实例分析表明了该方法的实用性和有效性。

In view of the load density method based on intelligent algorithm heavily dependent on the sample and hardly applicable, a spatial load clustering and integrated forecasting method of distribution network considering regional difference is proposed. Firstly, power users and regional information on different cities and different types are obtained through massive investigation. Secondly, according to an analysis of the inadequacy of the current load classification mode, a method based on typical daily load curves is put forward to check the sample load classification label and select the proper samples. Thirdly, a full sample space is constructed by the selected samples, which is further classified into a hierarchical sub-sample space according to a two-level partition of regional property and checked classification label. Finally, a support vector machine （SVM） is built with the most similar sub＋sample space to the plot to be predicted, thus the spatial load distribution is obtained. An actual case shows the practicality and effectiveness of the proposed method, which meets the accuracy requirements.