| 引用本文: | 陈 春,阳汉琨,肖轩怡,等.面向配电网低压用户的停电区域研判方法[J].电力系统保护与控制,2025,53(16):28-38.[点击复制] |
| CHEN Chun,YANG Hankun,XIAO Xuanyi,et al.A method for identifying the outage area of low-voltage distribution network users[J].Power System Protection and Control,2025,53(16):28-38[点击复制] |
|
| 摘要: |
| 在配电网低压停电区域研判过程中，由于中压配电终端在上报故障信息时存在漏报与误报现象，导致难以准确判断低压停电区域。为解决这一问题，提出了一种基于动态模糊贝叶斯网络的配电网停电区域研判方法。整合用户侧和配电终端的相关数据，利用中低压配电网的典型拓扑结构，构建动态贝叶斯网络(dynamic Bayesian network, DBN)，以用户停电事件为核心推理各区域停电事件发生的概率。在此基础上，评估当前推理结果是否需要修正。若需要，则将该推理结果作为模糊推理系统的输入，利用隶属度函数和推理规则，经过去模糊化处理进一步修正结果，最终推断出最可能的停电区域。通过分析某城市配电网的实际故障数据发现，当模拟信息缺失率为10%时，模型研判准确度达到83.59%，验证了该模型在信息不完全的条件下依然能保持较高的判断精度。 |
| 关键词: 低压配电网 动态贝叶斯网络 模糊理论 用户停电事件 停电区域 |
| DOI：10.19783/j.cnki.pspc.241411 |
| 投稿时间：2024-10-22修订日期：2024-12-31 |
| 基金项目:湖南省自然科学基金优秀青年项目资助(2023JJ 20039)；南方电网公司科技项目资助(031800KC23120003) |
|
| A method for identifying the outage area of low-voltage distribution network users |
| CHEN Chun1,YANG Hankun1,XIAO Xuanyi1,CAO Yijia1,AN Yi2 |
| (1. State Key Laboratory of Disaster Prevention & Reduction for Power Grid, Changsha University of Science &
Technology, Changsha 410114, China; 2. Electric Power Science Research Institute of State Grid
Jiangxi Electric Power Co., Ltd., Nanchang 330096, China) |
| Abstract: |
| In the process of identifying low-voltage outage areas in distribution networks, missing or false reports from medium-voltage distribution terminals often make it difficult to accurately determine the actual outage zones. To address this problem, this paper proposes a dynamic fuzzy Bayesian network-based outage area identification method for distribution networks. By integrating relevant data from both the user side and distribution terminals, and utilizing the typical topology of medium and low-voltage distribution networks, a dynamic Bayesian network (DBN) is constructed to infer the probability of outages occurring in various regions. Based on this initial inference, a verification step is conducted to determine whether results need to be corrected. If correction is required, the inference result is fed into a fuzzy reasoning system, where it is further refined using membership functions and inference rules, followed by defuzzification, to ultimately deduce the most probable outage area. An analysis of the actual fault data from a city’s distribution network show that, even with a 10% rate of simulated information loss, the accuracy of the model is 83.59 %, demonstrating its strong judgment accuracy under incomplete information conditions. |
| Key words: low-voltage distribution network dynamic Bayesian network fuzzy theory user outage event outage area |
