Cybersecurity Analysis of Data-Driven Power System Stability Assessment
Tóm tắt
Machine learning-based intelligent systems enhanced with Internet of Things (IoT) technologies have been widely developed and exploited to enable the real-time stability assessment of a large-scale electricity grid. However, it has been extensively recognized that the IoT-enabled communication network of power systems is vulnerable to cyberattacks. In particular, system operating states, critical attributes that act as input to the data-driven stability assessment, can be manipulated by malicious actors to mislead the system operator into making disastrous decisions and thus cause major blackouts and cascading events. In this article, we explore the vulnerability of the data-driven power system stability assessment, with a special emphasis on decision tree-based stability assessment (DTSA) approaches, and investigate the feasibility of constructing a physics-constrained adversarial attack (PCAA) to undermine the DTSA. The PCAA is formulated as a nonlinear programming problem considering the misclassification constraint, power limits, and bad data detection, computing potential adversarial perturbations that reverse the “stable/unstable” prediction of the real-time input while remaining invisible/stealthy. Extensive experiments based on the IEEE 68-bus system are conducted to evaluate the impact of PCAAs on predictions of DTSA and their transferability.
Từ khóa
#Adversarial examples #cyber security #decision trees (DTs) #physics-constrained machine learning (ML) #transient stability assessment (TSA)Tài liệu tham khảo
10.1109/PMAPS.2018.8440373
illera, 2014, Lights off! the darkness of the smart meters, Proc Block Hat Eur
10.1109/TPWRS.2013.2283064
2021, Data center security
10.1109/PMAPS.2016.7764166
lee, 2013, Electric Sector Failure Scenarios and Impact Analyses
morisson, 2007, Review of On-line Dynamic Security Assessment Tools and Techniques
2020, Major power outage in India could be triggered by cyber attack
10.1109/EuroSP.2016.36
10.1109/CVPR.2016.282
10.1109/TPWRS.2013.2246822
10.1109/ICEI.2018.00041
10.1109/TPWRS.2009.2016528
10.1109/TSG.2018.2873001
10.1109/TPWRS.2017.2669839
10.1038/s41560-018-0128-x
10.1145/1952982.1952995
10.1201/9780203913673
10.1109/JIOT.2021.3049818
2021, NP-hardness
10.1109/MSP.2011.67
10.1109/SmartGridComm.2017.8340729
10.1109/TSG.2020.3009401
10.1109/JIOT.2022.3161790
10.1109/SmartGridComm.2018.8587547
10.1109/TIFS.2019.2928624
10.1109/PESGM41954.2020.9281719
tjeng, 2018, Evaluating robustness of neural networks with mixed integer programming, Proc Int Conf Learn Represent, 1
tramer, 2019, Adversarial training and robustness for multiple perturbations, arXiv 1904 13000
2016, Analysis of the cyber attack on the ukrainian power grid Defense use case
10.1109/TSG.2015.2508449
pal, 2006, Robust Control in Power Systems
10.1007/s42835-019-00084-2
10.1109/TIFS.2019.2902822
papernot, 2017, Transferability in machine learning: From phenomena to black-box attacks using adversarial samples, Proc Asia Conf Comput Commun, 1
10.1109/TIFS.2017.2686367
10.1109/CVPRW.2016.58
goodfellow, 2014, Explaining and harnessing adversarial examples, arXiv 1412 6572
madry, 2017, Towards deep learning models resistant to adversarial attacks, arXiv 1706 06083
10.1109/JIOT.2021.3127895
10.1109/TPWRS.2013.2266617
10.1109/TPWRS.2009.2037006
10.1016/S1474-6670(17)65191-7
10.1109/TPWRS.2015.2496302
ren, 2022, A universal defense strategy for data-driven power system stability assessment models under adversarial examples, IEEE Internet of Things Journal
10.1109/TPWRS.2010.2082575
10.1109/TPWRS.2018.2794468