False Data Injection Attacks on Reinforcement Learning-Based Charging Coordination in Smart Grids and a Countermeasure DOI Creative Commons

A. A. El-Shazly,

Islam Elgarhy, Ahmed T. El-Toukhy

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(23), P. 10874 - 10874

Published: Nov. 24, 2024

Reinforcement learning (RL) is proven effective in optimizing home battery charging coordination within smart grids. However, its vulnerability to adversarial behavior poses a significant challenge the security and fairness of process. In this study, we, first, craft five stealthy false data injection (FDI) attacks that under-report state-of-charge (SoC) values deceive RL agent into prioritizing their requests, then, we investigate impact these on system. Our evaluations demonstrate attackers can increase chances compared honest consumers. As result, consumers experience reduced levels for batteries, leading degradation system’s performance terms fairness, consumer satisfaction, overall reward. These negative effects become more severe as amount power allocated decreases number system increases. Since total available limited, some with genuinely low SoC are not selected, creating disparity between malicious To counter serious threat, develop deep learning-based FDI attack detector evaluated it using real-world dataset. experiments show our identify high accuracy alarm rates, effectively protecting RL-based from mitigating impacts attacks.

Language: Английский

Optimizing PID Control for Automatic Voltage Regulators Using ADIWACO PSO DOI Creative Commons
Yaw Opoku Mensah Sekyere,

Priscilla Oyeladun Ajiboye,

Francis Boafo Effah

et al.

Scientific African, Journal Year: 2025, Volume and Issue: unknown, P. e02562 - e02562

Published: Jan. 1, 2025

Language: Английский

Citations

2
A state-of-the-art review on concurrent voltage and frequency regulation problems in renewable integrated power networks DOI
Vineet Kumar, Veena Sharma, Ananya Kumar

et al.

Energy Sources Part A Recovery Utilization and Environmental Effects, Journal Year: 2025, Volume and Issue: 47(1), P. 16 - 49

Published: Feb. 7, 2025

The growing demand for electricity has intensified the shift toward renewable energy sources like wind and solar, which are environmentally friendly due to their zero carbon emissions. However, intermittent nature poses significant challenges grid security, reliability, stability. An effort is made in this paper present a comprehensive review of critical issue concurrent voltage frequency regulation renewable-integrated power systems. It provides an overview these focuses on combined control strategies across different system configurations involving storage devices. also details modeling configuration essential components within loops. Additionally, case studies presented evaluate performance limitations recent controller structures. A literature survey latest thorough bibliography topic. Furthermore, challenges, opportunities, future research directions collective discussed detail.

Language: Английский

Citations

2
Intrusion detection in smart grids using artificial intelligence-based ensemble modelling DOI Creative Commons
Amjad Alsirhani, Noshina Tariq, Mamoona Humayun

et al.

Cluster Computing, Journal Year: 2025, Volume and Issue: 28(4)

Published: Feb. 25, 2025

Language: Английский

Citations

0
Hierarchical Privacy Protection Model in Advanced Metering Infrastructure Based on Cloud and Fog Assistance DOI Open Access

Linghong Kuang,

Wenlong Shi, Jing Zhang

et al.

Computers, materials & continua/Computers, materials & continua (Print), Journal Year: 2024, Volume and Issue: 80(2), P. 3193 - 3219

Published: Jan. 1, 2024

The Advanced Metering Infrastructure (AMI), as a crucial subsystem in the smart grid, is responsible for measuring user electricity consumption and plays vital role communication between providers consumers.However, with advancement of information technology, new security privacy challenges have emerged AMI.To address these enhance data Hierarchical Privacy Protection Model based on Cloud Fog Assistance (HPPM-AMICFA) proposed this paper.The model integrates cloud fog computing hierarchical threshold encryption, offering flexible efficient protection solution that significantly enhances grid.The methodology involves setting levels by processing missing utilizing fuzzy comprehensive analysis to evaluate importance, thereby assigning appropriate levels.Furthermore, encryption algorithm developed provide differentiated strategies nodes IDs, ensuring secure aggregation data.Experimental results demonstrate HPPM-AMICFA effectively resists various attack while minimizing time costs, safeguarding grid.

Language: Английский

Citations

1
Emas: an efficient MLWE-based authentication scheme for advanced metering infrastructure in smart grid environment DOI
Noureddine Chikouche,

Fares Mezrag,

Rafik Hamza

et al.

Journal of Ambient Intelligence and Humanized Computing, Journal Year: 2024, Volume and Issue: 15(11), P. 3759 - 3775

Published: Sept. 2, 2024

Language: Английский

Citations

0
A Comprehensive Study on Network Security in the Current Scenario DOI
A. Saranya,

B. Indrani

Published: Oct. 3, 2024

Language: Английский

Citations

0
False Data Injection Attacks on Reinforcement Learning-Based Charging Coordination in Smart Grids and a Countermeasure DOI Creative Commons

A. A. El-Shazly,

Islam Elgarhy, Ahmed T. El-Toukhy

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(23), P. 10874 - 10874

Published: Nov. 24, 2024

Reinforcement learning (RL) is proven effective in optimizing home battery charging coordination within smart grids. However, its vulnerability to adversarial behavior poses a significant challenge the security and fairness of process. In this study, we, first, craft five stealthy false data injection (FDI) attacks that under-report state-of-charge (SoC) values deceive RL agent into prioritizing their requests, then, we investigate impact these on system. Our evaluations demonstrate attackers can increase chances compared honest consumers. As result, consumers experience reduced levels for batteries, leading degradation system’s performance terms fairness, consumer satisfaction, overall reward. These negative effects become more severe as amount power allocated decreases number system increases. Since total available limited, some with genuinely low SoC are not selected, creating disparity between malicious To counter serious threat, develop deep learning-based FDI attack detector evaluated it using real-world dataset. experiments show our identify high accuracy alarm rates, effectively protecting RL-based from mitigating impacts attacks.

Language: Английский

Citations

0