综合智慧能源 ›› 2025, Vol. 47 ›› Issue (11): 87-95.doi: 10.3969/j.issn.2097-0706.2025.11.008

• 电力数据安全 • 上一篇    下一篇

面向电-气综合能源系统的协同虚假数据注入攻击建模

范雨欣a,b,c(), 李永明a,b,c(), 苏雷a,b,c,*()   

  1. a.新疆大学 a.计算机科学与技术学院 乌鲁木齐 830046
    b.新疆大学 新疆多模态智能处理与信息安全工程技术研究中心 乌鲁木齐 830046
    c.新疆大学 鹏城实验室新疆网络节点,乌鲁木齐 830046
  • 收稿日期:2025-06-25 修回日期:2025-10-10 出版日期:2025-11-25
  • 通讯作者: *苏雷(1994),男,副教授,博士,从事工业人工智能、智能感知与决策、能源互联网等方面的研究,sul@xju.edu.cn
  • 作者简介:范雨欣(1999),女,硕士生,从事虚假数据注入攻击建模与检测方法方面的研究,17667411303@163.com
    李永明(1971),男,副教授,硕士,从事计算机视觉方面的研究,lym@xju.edu.cn
  • 基金资助:
    国家自然科学基金项目(62403404);中国博士后科学基金第76批面上资助(2024M762716);新疆维吾尔自治区自然科学基金(2024D01C238)

Coordinated false data injection attacks against distributed economic dispatch of electricity-gas integrated energy system

FAN Yuxina,b,c(), LI Yongminga,b,c(), SU Leia,b,c,*()   

  1. a. Xinjiang University, School of Computer Science and Technology Urumqi 830046,China
    b. Xinjiang University, Xinjiang Multimodal Intelligent Processing and Information Security Engineering Technology Research Center Urumqi 830046,China
    c. Xinjiang University, Xinjiang Network Node of Pengcheng Laboratory, Xinjiang University,Urumqi 830046,China
  • Received:2025-06-25 Revised:2025-10-10 Published:2025-11-25
  • Supported by:
    National Natural Sciences Foundation of China (62403404);Natural Science Foundation of Xinjiang Uygur Autonomous Region (2024D01C238);China Postdoctoral Science Foundation(2024M762716)

摘要:

多能流调度利用信息技术打破了传统能源系统间的壁垒,构建了高效、互补的供能体系。然而,能源系统间大量测控数据的传输也加剧了其遭受网络攻击的风险。为深入探究综合能源系统(IES)中虚假数据注入攻击(FDIA)潜在的攻击模式,提出一种面向分布式调度的电-气协同FDIA建模方法。该方法考虑了IES中电力和燃气不同管理主体的影响,建立了分布式双层规划模型,并结合交替方向乘子法设计了双层嵌套分布式求解框架,保护了各主体数据隐私。采用凸松弛技术将燃气系统中Weymouth方程转化为易于处理的凸约束形式,便于模型的高效求解。在IEEE 39节点和比利时20节点互联的电-气IES中进行测试,结果表明协同FDIA的破坏效能可通过耦合设备在系统中传播,导致系统近70%的负荷被强制切除。分布式求解算法可在有限迭代步数内收敛,与集中式方法的相对误差控制在1.4%以内,所提方法为分析分布式架构下IES中的FDIA攻击模式提供了有效工具。

关键词: 综合能源系统, 协同虚假数据注入攻击, 分布式求解算法, 双层规划, 凸松弛, 多能流调度

Abstract:

Multi-energy flow scheduling utilizes information technology to break down the barriers between traditional energy systems, thereby forming an efficient and complementary energy supply system. However, the transmission of a large amount of measurement and control data between energy systems also increases their risk of being subjected to cyberattacks. To clarify the potential attack patterns of false data injection attacks (FDIA) in integrated energy systems (IES), a distributed scheduling-oriented modeling method for coordinated FDIA in the electricity-gas system was proposed. This method considered the influence of different management entities of electricity and gas in the IES, and established a distributed bilevel programming model. Additionally, it designed a bilevel nested distributed solution framework by integrating the alternating direction method of multipliers, protecting the privacy of data from each entity. Meanwhile, convex relaxation techniques were used to transform the Weymouth equation in the natural gas system into a tractable convex constraint form, facilitating the efficient solution of the model. Tests were conducted in an electricity-gas IES interconnected with the IEEE 39-bus and the Belgian 20-bus systems. The results showed that the destructive effect of the coordinated FDIA could be transmitted through coupling devices in the system, forcing the system to shed nearly 70% of the load. Meanwhile, the distributed solution algorithm could converge within a limited number of iterations, and the relative error compared with the centralized method was controlled within 1.4%. The proposed method provides an effective tool for analyzing the attack patterns of FDIA in IES under distributed architectures

Key words: integrated energy system, coordinated false data injection attacks, distributed solving algorithm, bilevel programming, convex relaxation, multi-energy flow scheduling

中图分类号: