| 引用本文: | 刘海涛,许伦,郝思鹏,张小莲,张潮.基于配电网分区的分布式混合储能优化方法[J].电力自动化设备,2020,40(5): |
| LIU Haitao,XU Lun,HAO Sipeng,ZHANG Xiaolian,ZHANG Chao.Optimization method of distributed hybrid energy storage based on distribution network partition[J].Electric Power Automation Equipment,2020,40(5): |
|
| 摘要: |
| 为了提高配电网对呈现不同出力特征的分布式电源的消纳能力，实现含多种分布式电源及大负荷并涉及不同出力特征的配电网优化运行，提出一种基于配电网分区的分布式混合储能优化方法。考虑多种分布式电源及大负荷的概率特征，通过K-means算法得到各分布式电源及大负荷的离散化模型，并计算离散化后配电网的电气距离矩阵，得到配电网分区的相似矩阵；采用吸引子传播(AP)聚类算法将配电网按照节点间的相似度大小分为多个分区，进而确定安装混合储能系统的聚类中心节点；考虑储能系统的充放电效率和荷电状态进行混合储能系统的容量配置，采用希尔伯特-黄变换将各分区的不平衡功率分解为高频分量和低频分量，分别作为功率型储能和能量型储能容量配置的参考功率，以确定最优配置方案。算例分析结果表明，所提优化方法可以有效提高储能的利用效率以及配置方案的经济性。 |
| 关键词: 分布式混合储能 配电网 离散化模型 分区 AP聚类算法 希尔伯特-黄变换 容量配置 |
| DOI：10.16081/j.epae.202004009 |
| 分类号:TM727 |
| 基金项目:国家自然科学基金资助项目(51607083);2018江苏省高校重大项目(18KJA470002) |
|
| Optimization method of distributed hybrid energy storage based on distribution network partition |
|
LIU Haitao1,2, XU Lun1, HAO Sipeng1,2, ZHANG Xiaolian1,2, ZHANG Chao1
|
|
1.School of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, China;2.Jiangsu Collaborative Innovation Center for Smart Distribution Network, Nanjing 211167, China
|
| Abstract: |
| In order to improve the absorption capacity of distribution network to DGs(Distributed Generations) with different output characteristics and realize the optimal operation of distribution network with multiple DGs and large loads involving different output characteristics, an optimization method of distributed hybrid energy storage based on distribution network partition is proposed. Considering the probabilistic characteristics of multiple DGs and large loads, the discretization models of DGs and large loads are obtained by using K-means algorithm, and the electrical distance matrix of distribution network after discretization is calculated to obtain the similar matrix of distribution network partition. The distribution network is divided into multiple partitions according to the similarity between nodes by using the AP(Attractor Propagation) clustering algorithm, and then the clustering center nodes for installing the hybrid energy storage system are determined. Considering the charging/discharging efficiency and state-of-charge of energy storage system, the capacity configuration of the hybrid energy storage system is carried out. The unbalanced power of each partition is decomposed into high frequency component and low frequency component by Hilbert-Huang transform, which are respectively used as the reference power of capacity configuration for power-type and energy-type energy storage, so as to determine the optimal configuration scheme. The example results show that the proposed optimization method can effectively improve the utilization efficiency of energy storage and the economy of configuration scheme. |
| Key words: distributed hybrid energy storage distribution network discretization model partition AP clustering algorithm Hilbert-Huang transform capacity configuration |
