| 引用本文: | 陈锋升,于仲安,王梓耀,等.快速配电网故障重构的编码改进及其应用算法[J].电力系统保护与控制,2025,53(15):113-124.[点击复制] |
| CHEN Fengsheng,YU Zhongan,WANG Ziyao,et al.Coding improvement and application algorithm for rapid distribution network fault reconfiguration[J].Power System Protection and Control,2025,53(15):113-124[点击复制] |
|
| 摘要: |
| 针对大型配电网最小环集获取耗时，以及现有优化算法难以在短时间内生成高质量故障重构方案等问题，提出一种快速配电网故障重构的编码改进及其应用算法。首先，基于Tarjan算法检测并剔除故障导致的孤岛区域，并根据最小环与反向边关联特性在最大双连通分量子图上利用宽度优先搜索(breadth first search, BFS)找到最小环集。其次，以最小环作为编码，针对环消除过程中不可行解的产生原因进行编码优化。然后，结合故障重构问题，利用抽样、图论和交叉方式对河马优化算法的3个关键搜索过程进行改进。最后，基于真实的751节点大型配电网算例，展开仿真分析。实验结果表明，所提方法获取最小环集的时间仅为现有方法的3.45%，并且在5 s内生成的故障重构策略，在网损、开关次数和电压偏差方面均优于其他算法。 |
| 关键词: 配电网 网络重构 故障 Tarjan算法 最小环 |
| DOI：10.19783/j.cnki.pspc.241445 |
| 投稿时间：2024-10-28修订日期：2025-03-25 |
| 基金项目:国家自然科学基金项目资助(52207105)；广东省基础与应用基础研究基金项目资助(2023A1515011598) |
|
| Coding improvement and application algorithm for rapid distribution network fault reconfiguration |
| CHEN Fengsheng1,YU Zhongan1,WANG Ziyao2,WU Yufeng2,XIAO Wencong2,PAN Zhenning2 |
| (1. School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China;
2. School of Electric Power, South China University of Technology, Guangzhou 510000, China) |
| Abstract: |
| To address the challenges of time-consuming minimum cycle basis acquisition in large-scale distribution networks and the inability of existing optimization algorithms to generate high-quality fault reconfiguration solutions within a short timeframe, a coding improvement and application algorithm for rapid distribution network fault reconfiguration method is proposed. First, the Tarjan algorithm is employed to detect and eliminate islanded regions caused by faults. Based on the correlation between minimum cycles and back edges, the minimum cycle basis on the maximum biconnected component subgraph is identified via breadth first search (BFS). Then, an encoding based on the identified minimum cycles is established, and encoding optimization is performed to eliminate the causes of infeasible solutions during the cycle elimination process. Subsequently, for the fault reconfiguration problem, three key search processes of the Hippopotamus optimization algorithm are enhanced using sampling, graph theory, and crossover methods. Finally, simulation analysis is conducted on a real-world large-scale distribution network with 751 buses. Experimental results demonstrate that the proposed method requires only 3.45% of the time compared to existing methods for acquiring the minimum cycle basis. Moreover, the fault reconfiguration strategy generated within 5 seconds outperforms other algorithms in terms of network loss, number of switching operations, and voltage deviation. |
| Key words: distribution network network reconfiguration fault Tarjan algorithm minimum cycle |
