考虑数据潮流模型的配电网关键节点辨识  

Identification of key nodes in distribution networks considering data-driven power flow model

作　　者：琚佳彬 赵健 杨克新 JU Jiabin;ZHAO Jian;YANG Kexin(School of Electrical Engineering,Shanghai University of Electric Power,Shanghai200090,China)

机构地区：[1]上海电力大学电气工程学院,上海200090

出　　处：《水利水电技术（中英文）》2023年第9期26-36,共11页Water Resources and Hydropower Engineering

基　　金：国家自然科学基金项目(51907114);国网浙江省电力有限公司科技项目(5211DS200084)。

摘　　要：【目的】配电网关键节点发生故障时会造成较大的停电损失,对关键节点进行辨识并加强对它的监控具有重要意义。随着新型电力系统的建设,配电网拓扑变动越来越频繁使得系统档案无法及时更新,造成已知拓扑信息部分失效,这导致现有基于已知拓扑的关键节点辨识方法效果不佳。针对此问题,提出了一种考虑数据潮流模型的配电网关键节点辨识方法。【方法】首先建立了基于数据驱动的潮流模型,通过回溯上述模型生成配电网邻接矩阵,解决了拓扑信息部分失效问题;然后,基于复杂网络理论和电压灵敏度提出了改进节点度数和改进节点介数指标,并引入风险理论中的电压越限风险指标,从而构建了节点关键性评估指标集;最后,基于上述指标集建立数据驱动的配电网关键节点综合辨识模型,实现了配电网的关键节点辨识。【结果】利用改进的IEEE 33节点系统对方法进行验证,结果显示:基于数据潮流模型的邻接矩阵生成算法在迭代500次后趋于稳定,迭代800次后其生成的邻接矩阵准确率可达95%以上;与对比方法在拓扑已知条件下的辨识结果进行比较,关键性排名前四的节点辨识准确率为100%,关键性排名前十的节点辨识准确率为80%。【结论】结果表明:考虑数据潮流模型的配电网关键节点辨识方法在拓扑未知的情况下能够达到对比方法在拓扑已知情况下的关键节点辨识效果。[Objective]The failure of key nodes in distribution network will cause great loss of outage.It is of great significance to identify and strengthen the monitoring of key nodes.With the construction of new power system,the topology change of distribu-tion network becomes more and more frequent,which makes the system archives cannot be updated in time,[Results]ing in the failure of the known topology information,which leads to the poor effect of the existing key node identification method based on the known topology.To solve this problem,an identification method of key nodes of distribution network considering data flow model is proposed.[Methods]Firstly,a data-driven power flow model is established,and the adjacency matrix of distribution network is generated by tracing the above model to solve the problem of partial failure of topological information.Secondly,based on the complex network theory and voltage sensitivity,the index of improved node degree and improved node interval is proposed,and the risk index of voltage exceeding the limit is introduced in the risk theory,so as to construct the node key evaluation index set.Finally,based on the above index set,a data-driven comprehensive identification model for key nodes of the distribution net-work is established to realize the identification of key nodes of the distribution network.[Results]The improved IEEE 33 node system is used to verify the method.The result show that the adjacency matrix generation algorithm based on data flow model be-comes stable after 500 iterations,and the accuracy of the adjacency matrix generated by the algorithm can reach more than 95%after 800 iterations.Compared with the identification result of the comparison method under the condition of known topology,the identification accuracy of the top four key nodes is 100%,and the identification accuracy of the top ten key nodes is 80%.[Conclusion]The key node identification method considering data flow model can achieve the identification effect of the key node identification method under

关 键 词：配电网 关键节点 数据潮流模型 复杂网络理论 风险理论 风险评估 影响因素 拓扑关系 

分 类 号：TM73[电气工程—电力系统及自动化]