Scholarlink

<p>Cyber-physical power systems (CPPSs) are vulnerable to coordinated cyber-physical attacks (CCPAs) following the attack uncertainties. It is urgent to identify valid defense measures against CCPAs. To tackle this problem, a tri-level distributionally robust optimization model (T-DRO) is proposed to minimize the risk. Firstly, a tri-level framework is constructed to portray the dynamics of defenders, attackers and operators. Secondly, a CCPA model is designed where the condition for attacks to be injected stealthily is derived considering the phasor measurement units (PMUs). Meanwhile, a DRO method based on Kullback–Leibler (KL) divergence is proposed to capture the attack resource uncertainties. Finally, an improved C&amp;CG algorithm is developed to achieve the defense allocation solution. Case studies performed on the IEEE 14-bus system and 57-bus system demonstrate that the proposed method has better performance compared with the traditional methods. We can conclude that it is capable of providing robust strategies against CCPAs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A tri-level distributionally robust optimization model against coordinated cyber-physical attacks considering attack resource uncertainties

  • 期刊论文

  • 发表日期：

    2025年3月14日

查看全文
  • Linying Huang,
  • Jizhong Zhu,
  • Di Zhang,
  • Yixi Chen,
  • Yanfeng Li

摘要

Cyber-physical power systems (CPPSs) are vulnerable to coordinated cyber-physical attacks (CCPAs) following the attack uncertainties. It is urgent to identify valid defense measures against CCPAs. To tackle this problem, a tri-level distributionally robust optimization model (T-DRO) is proposed to minimize the risk. Firstly, a tri-level framework is constructed to portray the dynamics of defenders, attackers and operators. Secondly, a CCPA model is designed where the condition for attacks to be injected stealthily is derived considering the phasor measurement units (PMUs). Meanwhile, a DRO method based on Kullback–Leibler (KL) divergence is proposed to capture the attack resource uncertainties. Finally, an improved C&CG algorithm is developed to achieve the defense allocation solution. Case studies performed on the IEEE 14-bus system and 57-bus system demonstrate that the proposed method has better performance compared with the traditional methods. We can conclude that it is capable of providing robust strategies against CCPAs.