Intrusion detection of cyber physical energy system based on multivariate ensemble classification

Li, Yunfeng; Xue, Wenli; Wu, Ting; Wang, Huaizhi; Zhou, Bin; Aziz, Saddam; He, Yihui

doi:10.1016/j.energy.2020.119505

Cited by 32 publications

(6 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same scheme was used in [19] to build a framework to secure fog-to-things environment by gathering the classifiers DT, K-NN and RF. Likewise, the authors of [20] chose to uphold the protection of the cyber physical energy systems with a majority voting scheme. To do so, they aggregate the results of three individual classifiers: LightGBM, extreme learning machine (ELM), and XGBoost.…”

Section: Distribution Of Papers By Types Of Ensemble Learning Methodsmentioning

confidence: 99%

Intrusion Detection System using Ensemble Learning Approaches: A Systematic Literature Review

Arqane

Boutkhoum²,

Boukhriss³

et al. 2022

Int. J. Onl. Eng.

View full text Add to dashboard Cite

Undoubtedly, the advancements in Machine Learning (ML) and especially ensemble learning models enable researchers to develop numerous fields in ways we had never imagined before. Intrusion Detection System (IDS) is one of these fields that benefits from the aforementioned techniques to identify and classify the security threats with more accuracy. In fact, IDS technology has become an essential component for any defense strategy, since it provides a healthy environment for businesses. Besides, it protects future network infrastructures from intruders and suspicious network activities. In this context, this paper presents a systematic literature review on ensemble learning-based IDS and the datasets used to evaluate the proposed methods. Furthermore, we illustrate the distribution of the reviewed papers by year of publication, datasets utilized and ensemble learning methods. Finally, we discuss the finding of this review and highlight some limitations of the aforementioned models. The overall purpose of this literature review is to provide a guideline on how to choose ensemble learning models to solve IDS issues based on the most efficient and recent trends

show abstract

Section: Distribution Of Papers By Types Of Ensemble Learning Methodsmentioning

confidence: 99%

Intrusion Detection System using Ensemble Learning Approaches: A Systematic Literature Review

Arqane

Boutkhoum²,

Boukhriss³

et al. 2022

Int. J. Onl. Eng.

View full text Add to dashboard Cite

show abstract

“…As an alternative to FDI detection, machine learning has been proposed. Injection attacks can also lead to the security breach of an entire web server, resulting in a denial-ofservice attack [39,40] .…”

Section: Research Backgroundmentioning

confidence: 99%

A Review on the Evaluation of Feature Selection Using Machine Learning for Cyber-Attack Detection in Smart Grid

Mohammed,

Al-Jumaily,

Singh

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The Smart Grid is a modern power grid that relies on advanced technologies to provide reliable and sustainable electricity. However, its integration with various communication technologies and IoT devices makes it vulnerable to cyber-attacks. Such attacks can lead to significant damage, economic losses, and public safety hazards. To ensure the security of the smart grid, increasingly strong security solutions are needed. This paper provides a comprehensive analysis of the vulnerabilities of the smart grid and the different approaches for detecting cyber-attacks. It examines the different vulnerabilities of the smart grid, including system vulnerabilities and cyber-attacks, and discusses the vulnerabilities of all its elements. The paper also investigates various approaches for detecting cyber-attacks, including rule-based, signature-based, anomaly detection, and ma-chine learning-based methods, with a focus on their effectiveness and related research. Finally, prospective cybersecurity approaches for the smart grid, such as AI approaches and blockchain, are discussed along with the challenges and future prospects of cyberattacks on the smart grid. The paper's findings can help policymakers and stakeholders make informed decisions about the security of the smart grid and develop effective strategies to protect it from cyber-attacks.

show abstract

“…Previous work discussed various ways to detect attacks and malicious activities on the grid. For instance, in [18,19,20], the authors proposed deep learning algorithms to create intrusion detection models based on readings obtained from various input sources such as the Phasor Measurement Unit (PMU), control panel logs, snort network alerts, and relay logs. While these mechanisms mainly depend on monitoring the power grid performance and state estimation to detect abnormal behaviors, they may fail to detect covert and stealthy attacks that are hidden or rendered as normal [21].…”

Section: Introductionmentioning

confidence: 99%

Edge-Based Detection and Localization of Adversarial Oscillatory Load Attacks Orchestrated By Compromised EV Charging Stations

Sarieddine¹,

Sayed²,

Torabi³

et al. 2023

Preprint

View full text Add to dashboard Cite

Recent reports indicate that Electric Vehicle Charging Stations (EVCS) are susceptible to remote exploitation through their vulnerable software/cyber components. More importantly, compromised EVCS can be leveraged to perform coordinated oscillatory load attacks against the interconnected power grid, leading to power grid instability, increased operational costs, and power line tripping. In this paper, we investigate an edge-based approach for the detection and localization of coordinated oscillatory load attacks initiated by exploited EV charging stations against the power grid. We rely on the behavioral characteristics of the power grid in the presence of interconnected EVCS while combining cyber and physical layer features to implement deep learning algorithms for the effective detection of oscillatory load attacks at the EVCS. We evaluate the proposed detection approach by building a real-time test bed to synthesize benign and malicious data, which was generated by analyzing real-life EV charging data collected during recent years. The results demonstrate the effectiveness of the implemented approach with the Convolutional Long-Short Term Memory model producing optimal classification accuracy (99.4%). Moreover, our analysis results shed light on the impact of such detection mechanisms towards building resiliency into different levels of the EV charging ecosystem while allowing power grid operators to localize attacks and take further mitigation measures. Specifically, we managed to decentralize the detection mechanism of oscillatory load attacks and create an effective alternative for operator-centric mechanisms to mitigate multi-operator and MitM oscillatory load attacks against the power grid. Finally, we leverage the created test bed to evaluate a distributed mitigation technique, which can be deployed on public/private charging stations to average out the impact of oscillatory load attacks while allowing the power system to recover smoothly within 1 second with minimal overhead.

show abstract

Intrusion detection of cyber physical energy system based on multivariate ensemble classification

Cited by 32 publications

References 36 publications

Intrusion Detection System using Ensemble Learning Approaches: A Systematic Literature Review

Intrusion Detection System using Ensemble Learning Approaches: A Systematic Literature Review

A Review on the Evaluation of Feature Selection Using Machine Learning for Cyber-Attack Detection in Smart Grid

Edge-Based Detection and Localization of Adversarial Oscillatory Load Attacks Orchestrated By Compromised EV Charging Stations

Contact Info

Product

Resources

About