基于K-means聚类算法和智能电表数据的居民峰值负荷概率估计模型

为解决低压变电站容量长期规划中,居民负荷的随机波动性导致峰值负荷难以准确估计问题,本文提出了一种基于K-means聚类算法和智能电表数据的居民峰值负荷概率估计模型,此方法以解决以往确定性峰值负荷估计难以衡量估计误差问题。首先,根据家庭人口数将居民分为两大类,即中小规模居民和大规模居民;其次,基于Kmeans聚类算法多次迭代抽样,分别求取两类居民的标准多样性最大化需求曲线,即峰值负荷概率分布和居民人口数之间的函数关系;最后,根据新居民人口数确定两类居民的峰值负荷概率分布,分别乘以对应的贡献因子系数后再求取其联合概率分布,可得到新入住居民的总峰值负荷概率分布。仿真算例表明,本文提出的居民峰值负荷概率估计模型能较准确地新居民的总峰值负荷。

K-means clustering algorithm and smart meter data based residential peak load probability estimation model

In order to solve the problem that the peak load is difficult to be accurately estimated due to the random fluctuation of residential load in the long-term capacity planning of low-voltage substation,In this paper,a probability estimation model of residential peak load based on K-means clustering algorithm and smart meter data is proposed to solve the problem that it is difficult to measure the estimation error in the previous deterministic peak load estimation.First,the residents are divided into two categories according to the number of households,namely small and medium-sized residents and large-scale residents.Secondly,K-means clustering algorithm based iterative sampling is performed multiple times to obtain the normalized maximum diversified demand(NMDD)curve of the two types of residents.This curve describes the the functional relationship between the peak load probability distribution and the number of resident populations.Finally,the peak load probability distribution of the two types of residents is determined according to the number of new residents.The above two probability distributions are respectively multiplied by contribution factor(CF)coefficient.Then the joint probability distribution is calculated to get the total peak load probability distribution of new residents.The simulation example shows that the proposed peak load probability estimation model can accurately predict the total peak load of new residents.