半月刊

ISSN 1000-1026

CN 32-1180/TP

+高级检索 English
基于互信息贝叶斯网络的配电网拓扑鲁棒辨识算法
作者:
作者单位:

1.四川大学电气工程学院,四川省成都市 610065;2.国网四川省电力公司计量中心,四川省成都市 610045

摘要:

基于微型同步相量测量装置(μPMU),提出一种拓扑识别新算法,通过贝叶斯网络(BN)拟合拓扑、光伏、负荷及μPMU测量电压的非线性关系,引入最大互信息网格划分度量连续型节点区间,解决了BN处理连续型数据时需人为指定区间数目、难以适应较多连续型变量的问题。基于拉丁超立方抽样生成光、荷数据保证了场景在样本空间内均匀分布,简化了BN的训练过程并提高了拓扑识别的泛化能力。通过仿真算例验证方法的有效性,结果表明,所提方法具有与实时估计匹配法相当的识别精度和更高的时效性,且识别时间不随配电网可行拓扑数量而线性增长,适用性佳。在μPMU部分失效或负荷、光伏等数据缺失时仍能保证较高识别率,具有较强的鲁棒性。

关键词:

基金项目:

国家自然科学基金资助项目(51977133)。

通信作者:

作者简介:


Robust Identification Algorithm for Distribution Network Topology Based on Mutual Information-Bayesian Network
Author:
Affiliation:

1.College of Electrical Engineering, Sichuan University, Chengdu 610065, China;2.State Grid Sichuan Electric Power Corporation Metering Center, Chengdu 610045, China

Abstract:

Based on the μPMU (micro phasor measurement unit), this paper proposes a new method for identifying distribution network topology. The nonlinear relationship among distribution network topology, photovoltaic (PV), load and μPMU measured voltage is fitted by Bayesian network (BN). The interval of continuous nodes is measured by the grid division of maximal information coefficient (MIC) among variables. MIC solves the problem that traditional BN needs to specify the interval number artificially when processing continuous data, and is difficult to be applied to the case with lots of continuous variables. The photovoltaic-load data generated by the Latin hypercube sampling (LHS) can realize the uniform distribution of the scene in the sample space. The training process of BN and improves the generalization ability of identification is effectively simplified. The effectiveness of the method is verified by a simulation example. The simulation results show that the proposed method has the same identification accuracy and the higher timeliness as the real-time estimation matching method. The identification time does not increase linearly with the number of feasible topologies. Even in the case of partial failure of the μPMU or lack of key data, such as load and PV data, it can ensure a high identification effect and a strong robustness.

Keywords:

Foundation:
This work is supported by National Natural Science Foundation of China (No. 51977133).
引用本文
[1]任鹏哲,刘友波,刘挺坚,等.基于互信息贝叶斯网络的配电网拓扑鲁棒辨识算法[J/OL].电力系统自动化,http://doi. org/10.7500/AEPS20200818001.
REN Pengzhe, LIU Youbo, LIU Tingjian, et al. Robust Identification Algorithm for Distribution Network Topology Based on Mutual Information-Bayesian Network[J/OL]. Automation of Electric Power Systems, http://doi. org/10.7500/AEPS20200818001.
复制
分享
历史
  • 收稿日期:2020-08-18
  • 最后修改日期:2021-02-08
  • 录用日期:2020-12-03
  • 在线发布日期:
  • 出版日期:
相关附件