欢迎来到OAJRC平台！

期刊栏目

投稿

审稿

期刊目次

加入编委

期刊订阅

添加您的邮件地址以接收即将发行期刊数据：

当前位置：电气工程与自动化 > 2025年 4卷 (2)期

Open Access Article

Journal of Electrical Engineering and Automation. 2025; 4: (2) ; 133-144 ; DOI: 10.12208/j.jeea.20250067.

Active distribution network outage information fusion governance method based on multi-source information fusion
基于多源信息融合的有源配电网故障信息治理方法

作者： 葛辉^1,2,3 *, 胡国洋¹, 邹德龙², 刘程子¹, 过祈睿¹, 徐俊俊¹

1 南京邮电大学自动化学院、人工智能学院江苏南京
2 国电南瑞科技股份有限公司江苏南京
3 南京邮电大学碳中和先进技术研究院江苏南京

*通讯作者：葛辉,单位：南京邮电大学自动化学院、人工智能学院江苏南京国电南瑞科技股份有限公司江苏南京南京邮电大学碳中和先进技术研究院江苏南京；

国家自然科学基金项目（52077106，62293505，62293500）南瑞集团国家重点实验室（电网运行风险防御技术与装备）开放基金“规则与数据驱动下自成长型防御策略研究”项目

发布时间: 2025-02-27 总浏览量: 134

PDF 全文下载引用本文

摘要

为实现“双碳”战略目标，分布式能源迅猛发展并广泛接入到配电网，导致有源配电网的网络结构和运行环境更加复杂，影响系统安全稳定运行的故障迅速飙升，加之故障信息多源异构的特性，使故障溯源面临巨大挑战。针对上述问题，本文提出了一种基于改进D-S证据融合理论（Dempster-Shafer evidence theory）的有源配电网多源故障信息融合治理方法。首先，基于多模态的分析数据，确立特征层融合策略以突破传统数据级融合的计算复杂度瓶颈；其次，针对开关量信息与电气量信息的时频域特性差异，构建小波变换驱动的电气量特征提取模型，充分发挥其时频局部化表征优势；随后基于D-S证据理论融合算法改进融合规则，量化处理多源证据间的矛盾关系，以充分考虑各个证据源间的矛盾情况；最终对特征数据进行融合，为故障诊断提供多维信息支撑。基于IEEE-39网络与实际网络作为算例进行仿真验证，结果表明相较于传统组合规则，本文方法所得到的融合结果综合准确率有所提升，可高效、准确地表征信息采集系统所收集到的电气量信息，为有源配电网运维系统故障信息诊断、定位以及决策响应等提供有力支撑。

关键词： 改进证据理论；多源信息融合；暂态特征提取；配电网停电故障

Abstract

To achieve the "dual-carbon" strategic goals, the rapid development and extensive integration of distributed energy resources (DERs) into power distribution networks have resulted in increasingly complex topological structures and operational environments for active distribution networks. This complexity has led to a surge in system faults that threaten grid security and stability. Coupled with the multi-source heterogeneous nature of fault information, these challenges significantly impede effective fault traceability. Addressing these issues, this paper proposes a multi-source fault information fusion governance method for active distribution networks based on an improved Dempster-Shafer (D-S) evidence theory. The methodology establishes a technical framework encompassing "feature dimension reduction-information representation-rule optimization-decision fusion". First, leveraging multimodal analytical data, a feature-level fusion strategy is established to circumvent the computational complexity bottlenecks inherent in conventional data-level fusion approaches. Second, recognizing the distinct time-frequency characteristics between switching quantity information and electrical quantity information, a wavelet transform-driven feature extraction model is constructed for electrical quantities, harnessing its inherent advantages in time-frequency localized representation. Subsequently, the D-S evidence fusion rules are innovatively redesigned through dynamic conflict factor correction mechanisms to quantitatively resolve contradictions among multi-source evidence. The final fusion decision vector provides multidimensional information support for fault diagnosis. Simulation validations conducted on both the IEEE-39 standard network and practical distribution networks demonstrate that compared with traditional combination rules, the proposed method achieves a 12.3% improvement in comprehensive accuracy while reducing feature extraction time consumption by 42%. The fusion results precisely characterize the dynamic electrical quantity information collected by monitoring systems, offering robust technical support for fault diagnosis, localization, and decision-making responses in active distribution network operation and maintenance systems. This research advances the field of multi-source heterogeneous information fusion through innovative rule design and computational architecture optimization, providing a novel technical paradigm for intelligent grid fault management.

Key words： Improved evidence theory; Multi-source information fusion; Transient feature extraction; Distribution network outage fault

参考文献 References

[1] Yuan Y, Dehghanpour K, Wang Z, et al. Multisource data fusion outage location in distribution systems via probabilistic graphical models[J]. IEEE Transactions on Smart Grid, 2021, 13(2): 1357-1371.

[2] Wang H, Lin D, Qiu J, et al. Research on multiobjective group decision-making in condition-based maintenance for transmission and transformation equipment based on DS evidence theory[J]. IEEE Transactions on Smart Grid, 2015, 6(2): 1035-1045.

[3] Cui Q, Zhu J, Shu J, et al. Comprehensive evaluation of electric power prediction models based on DS evidence theory combined with multiple accuracy indicators[J]. Journal of Modern Power Systems and Clean Energy, 2021, 10(3): 597-605.

[4] Li S, Li W, Chen L, et al. Real-time robust state estimation for large-scale low-observability power-transportation system based on meta physics-informed graph TimesNet[J]. IEEE Transactions on Smart Grid, 2024.

[5] 刘崇崇.基于多数据源信息融合的配电网故障诊断方法研究[D].华北电力大学,2016.

[6] Wu K, Xiao F, Zhang Y. A Fractal-Based Complex Belief Entropy for Uncertainty Measure in Complex Evidence Theory[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2024, 55(2): 910-924.

[7] Wu X, Wang D, Cao W, et al. A genetic-algorithm support vector machine and DS evidence theory based fault diagnostic model for transmission line[J]. IEEE Transactions on Power Systems, 2019, 34(6): 4186-4194.

[8] 曹海.基于D-S证据理论的变压器诊断系统研究[J].四川电力技术,2009,32(05):46-49.

[9] Jiao Z, Gong H, Wang Y. A D-S evidence theory-based relay protection system hidden failures detection method in smart grid[J]. IEEE Transactions on Smart Grid, 2016, 9(3): 2118-2126.

[10] Zhou M, Zhu S S, Chen Y W, et al. A generalized belief entropy with nonspecificity and structural conflict[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021, 52(9): 5532-5545.

[11] Patel T K, Panda P C, Swain S C, et al. A fault detection technique in transmission line by using discrete wavelet transform[C]. 2017 Second International Conference on Electrical, Computer and Communication Technologies (ICECCT), IEEE, 2017.

[12] 谢凯.基于D-S证据理论的小电流接地系统单相接地故障选线与定位技术研究[D].中国石油大学,2010.

[13] Gao X, Pan L, Deng Y. Quantum pythagorean fuzzy evidence theory: A negation of quantum mass function view[J]. IEEE Transactions on Fuzzy Systems, 2021, 30(5): 1313-1327.

[14] 席磊,陈采玉,陈洪军,等.基于图注意力与多尺度并行融合卷积的虚假数据注入攻击定位检测[J].高电压技术,1-11.

[15] Zhou M, Zhu S S, Chen Y W, et al. A generalized belief entropy with nonspecificity and structural conflict[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021, 52(9): 5532-5545.

[16] Xu B ,Yin X ,Yin X , et al.Fault Diagnosis of Power Systems Based on Temporal Constrained Fuzzy Petri Nets.[J].IEEE Access,2019, 7:101895-101904.

[17] 乔帅君.基于多源信息融合的智能电网故障诊断方法研究[D].中国矿业大学,2022.

[18] 王守鹏,赵冬梅.电网故障诊断的研究综述与前景展望[J].电力系统自动化,2017,41(19):164-175.

[19] 徐欢,杨秋勇,邹文景,等.深度学习在电力系统异常损耗数据识别中的应用研究[J].南京邮电大学学报(自然科学版),2024,44(02):101-110.

[20] Liu L, Yuan Y, Wang Z, et al. Integrated Framework of Multisource Data Fusion for Outage Location in Looped Distribution Systems[J]. IEEE Transactions on Smart Grid, 2025.

[21] 高鹏翔.基于多源数据融合的配电网运行故障特征信息提取技术研究[D].华北电力大学(北京),2019.

[22] 陈伟彪,陈亦平,姚伟,等.采用WAMS时序信息的故障诊断方法及应用[J].电网技术,2017,41(06):1956-1963.

[23] 何俊涛,车仁飞,孟庆萌等.基于广域录波数据和FCM聚类的电网故障诊断方法[J].电力自动化设备,2019, 39(06): 179-184.

[24] 王娜,刘静渝,李皓然,等.基于D-S证据理论的多智能体系统冲突数据融合机制研究[J].重庆大学学报,2025, 48(02):22-34.

[25] 袁杰,王福利,王姝,等.基于D-S融合的混合专家知识系统故障诊断方法[J].自动化学报,2017,43(09):1580-1587.

[26] He C, Wang Y, Du W, et al. Phase Noise Suppression in Phase-Sensitive OTDR Using Fault-Tolerant DS Evidence Theory[J]. Journal of Lightwave Technology, 2024, 42(24): 8900-8912.

[27] Ni Q, Li X, Chen Z, et al. A mechanism and data hybrid-driven method for main circuit ground fault diagnosis in electrical traction system[J]. IEEE Transactions on Industrial Electronics, 2023, 70(12): 12806-12815.

[28] 李月.改进的D-S证据理论的信息融合技术在发电机组旋转设备故障诊断中的应用[D].华东理工大学,2012.

引用本文

葛辉, 胡国洋, 邹德龙, 刘程子, 过祈睿, 徐俊俊, 基于多源信息融合的有源配电网故障信息治理方法[J]. 电气工程与自动化, 2025; 4: (2) : 133-144.

入驻平台

回到首页

搜索文章: