Scholarlink

<p>The increasing integration of cyber-physical systems in smart grids has heightened vulnerabilities to cyberattacks, particularly false data injection attacks (FDIAs), which manipulate measurement data to bypass traditional bad data detection mechanisms. This paper proposes a robust model-driven framework for detecting FDIAs in power systems by combining an adaptive extended Kalman filter (AEKF) with weighted least squares (WLS) state estimation. First, an improved dynamic state prediction method is introduced, leveraging three-parameter exponential smoothing to minimize linearization errors in nonlinear systems and enhance the accuracy of state estimation. Second, the AEKF dynamically adjusts process and measurement noise covariances to suppress the impact of malicious data, ensuring robust state estimation under varying attack intensities. Third, a novel state estimation discrepancy detection (SEDD) method is developed by exploiting the inherent differences between AEKF (dynamic) and WLS (static) estimators, enabling reliable detection of stealthy FDIAs. Simulations of IEEE 14-bus and 30-bus systems demonstrate the effectiveness of the framework.</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

Enhancing Smart Grid Security: A Robust Framework for Detecting False Data Injection Attacks

  • 期刊论文

  • 发表日期：

    2025年9月6日

查看全文
  • Wenlong Jiang,
  • Kangrui Lan,
  • Yunfei Li,
  • Yaru Li,
  • Rui Wang,
  • Zhen Bao,
  • Jinglong Qiu,
  • Wengen Gao

摘要

The increasing integration of cyber-physical systems in smart grids has heightened vulnerabilities to cyberattacks, particularly false data injection attacks (FDIAs), which manipulate measurement data to bypass traditional bad data detection mechanisms. This paper proposes a robust model-driven framework for detecting FDIAs in power systems by combining an adaptive extended Kalman filter (AEKF) with weighted least squares (WLS) state estimation. First, an improved dynamic state prediction method is introduced, leveraging three-parameter exponential smoothing to minimize linearization errors in nonlinear systems and enhance the accuracy of state estimation. Second, the AEKF dynamically adjusts process and measurement noise covariances to suppress the impact of malicious data, ensuring robust state estimation under varying attack intensities. Third, a novel state estimation discrepancy detection (SEDD) method is developed by exploiting the inherent differences between AEKF (dynamic) and WLS (static) estimators, enabling reliable detection of stealthy FDIAs. Simulations of IEEE 14-bus and 30-bus systems demonstrate the effectiveness of the framework.