Cyber security of a power grid: State-of-the-art

Sun, Chuanwen; Hahn, Adam; Liu, Chen‐Ching

doi:10.1016/j.ijepes.2017.12.020

Cited by 370 publications

(168 citation statements)

References 48 publications

Supporting

Mentioning

165

Contrasting

Unclassified

Order By: Relevance

“…• Instead of re-generating the adversarial examples at each iteration, the presented training only generates the adversarial examples once (before the first iteration). • To avoid the complex optimization process in (10) and (11), the adversarial examples are generated simply based on the attack models, e.g., (5)- (8). For instance, if the scaling attack is employed to generate adversarial examples, each input training data point is modified based on the attack parameters provided by (5).…”

Section: Accelerated Adversarial Trainingmentioning

confidence: 99%

Enabling Cyberattack-Resilient Load Forecasting through Adversarial Machine Learning

Tang

Jiao

Zhang

et al. 2019

2019 IEEE Power &Amp; Energy Society General Meeting (PESGM)

View full text Add to dashboard Cite

In the face of an increasingly broad cyberattack surface, cyberattack-resilient load forecasting for electric utilities is both more necessary and more challenging than ever. In this paper, we propose an adversarial machine learning (AML) approach, which can respond to a wide range of attack behaviors without detecting outliers. It strikes a balance between enhancing a system's robustness against cyberattacks and maintaining a reasonable degree of forecasting accuracy when there is no attack. Attack models and configurations for the adversarial training were selected and evaluated to achieve the desired level of performance in a simulation study. The results validate the effectiveness and excellent performance of the proposed method.

show abstract

Section: Accelerated Adversarial Trainingmentioning

confidence: 99%

Enabling Cyberattack-Resilient Load Forecasting through Adversarial Machine Learning

Tang

Jiao

Zhang

et al. 2019

2019 IEEE Power &Amp; Energy Society General Meeting (PESGM)

View full text Add to dashboard Cite

show abstract

“…Not all of the factors studied are essentially interconnected, but separate groups of input parameters. A transition is needed to a set of independent parameters containing the necessary information about the variation or dispersion of the initial set of factors of the process under study [15]. It is proposed to use:…”

Section: Feature Selectionmentioning

confidence: 99%

“…Thus, it can be concluded that the network attacks detection systems based on the use of artificial intelligence methods as a key element of ensuring cybersecurity of the critical infrastructure [13,14,15] of the APCS in the concept of the development of the digital economy are of relevance and need to be improved.…”

Section: Introductionmentioning

confidence: 99%

Intrusion detection system on the basis of data mining algorithms in the industrial network

Gurin¹,

Vulfin²,

Vasilyev³

et al. 2019

Proceedings of the v International Conference Information Technology and Nanotechnology 2019

View full text Add to dashboard Cite

The purpose of the work is to increase the security of the industrial network of an automated process control system based on intelligent network traffic analysis algorithms. The analysis of the problem of detecting and recording actions of violators on the implementation of a network attack on an automated process control system in the industrial network of an enterprise has been performed. A structural and functional model of the monitoring system of the industrial network of industrial control systems is proposed. An algorithm is developed for the intellectual analysis of network traffic of industrial protocols and a software package that implements the proposed algorithms as part of a monitoring system to evaluate the effectiveness of the proposed solution on field data. Data Science M A Gurin, A M Vulfin, V I Vasilyev and A V Nikonov V International Conference on "Information Technology and Nanotechnology" (ITNT-2019) 554 IPS/IDS WEB services DMZ ERP/MES level DB and data collection DMZ Department network Department network AD File Server HMI SCADA/ DCS

show abstract

“…(). Specific attention has been paid in the literature to power systems (see, e.g., the papers by Ericsson, ; Fovino, Guidi, Masera, & Stefanini, ; Onyeji, Bazilian, & Bronk, ; Sun, Hahn, & Liu, ), industrial control systems (dealt with in the papers by Cherdantseva et al., ; Ralston, Graham, & Hieb, ), and nuclear plants (as in the papers by Park, Suh, & Park, ; Shin, Son, & Heo, ). Whatever the specific infrastructure, securing it requires an investment in cybersecurity tools and policies.…”

Section: Introductionmentioning

confidence: 99%

“…Cyber risks represent a major source of concern for several utility and industrial infrastructures, due to both the risk arising in individual infrastructures and that deriving from the interconnections and interdependencies among them, as shown in the papers by Kröger (2008) and by Maglaras et al (2018). Specific attention has been paid in the literature to power systems (see, e.g., the papers by Ericsson, 2010;Fovino, Guidi, Masera, & Stefanini, 2011;Onyeji, Bazilian, & Bronk, 2014;Sun, Hahn, & Liu, 2018), industrial control systems (dealt with in the papers by Cherdantseva et al, 2016;Ralston, Graham, & Hieb, 2007), and nuclear plants (as in the papers by Park, Suh, & Park, 2016;Shin, Son, & Heo, 2017). Whatever the specific infrastructure, securing it requires an investment in cybersecurity tools and policies.…”

Section: Introductionmentioning

confidence: 99%

Robustness of Optimal Investment Decisions in Mixed Insurance/Investment Cyber Risk Management

Mazzoccoli

Naldi

2019

Risk Analysis

View full text Add to dashboard Cite

An integrated risk management strategy, combining insurance and security investments, where the latter contribute to reduce the insurance premium, is investigated to assess whether it can lead to reduced overall security expenses. The optimal investment for this mixed strategy is derived under three insurance policies, covering, respectively, all the losses (total coverage), just those below the limit of maximum liability (partial coverage), and those above a threshold but below the maximum liability (partial coverage with deductibles). Under certain conditions (e.g., low potential loss, or either very low or very high vulnerability), the mixed strategy reverts however to insurance alone, because investments do not provide an additional benefit. When the mixed strategy is the best choice, the dominant component in the overall security expenses is the insurance premium in most cases. Optimal investment decisions require an accurate estimate of the vulnerability, whereas larger estimation errors may be tolerated for the investment‐effectiveness coefficient.

show abstract

Cyber security of a power grid: State-of-the-art

Cited by 370 publications

References 48 publications

Enabling Cyberattack-Resilient Load Forecasting through Adversarial Machine Learning

Enabling Cyberattack-Resilient Load Forecasting through Adversarial Machine Learning

Intrusion detection system on the basis of data mining algorithms in the industrial network

Robustness of Optimal Investment Decisions in Mixed Insurance/Investment Cyber Risk Management

Contact Info

Product

Resources

About