Abstract
Integration of Information and Communications Technology (ICT) into the distribution system makes today’s power grid more remotely monitored and controlled than it has been. The fast increasing connectivity, however, also implies that the distribution grid today, or smart grid, is more vulnerable. Thus, research into intrusion/anomaly detection systems at the distribution level is in critical need. Current research on Intrusion Detection Systems for the power grid has been focused primarily on cyber security at the Supervisory Control And Data Acquisition, and single node levels with little attention on coordinated cyberattacks at multiple nodes. A holistic approach toward system-wide cyber security for distribution systems is yet to be developed. This paper presents a novel approach toward intrusion prevention, using a multi-agent system, at the distribution system level. Simulations of the method have been performed on the IEEE 13-Node Test Feeder, and the results compared to those obtained from existing methods. The results have validated the performance of the proposed method for protection against cyber intrusions at the distribution system level.
Highlights
Reference [15] proposes a network-based intrusion detection system (NIDS) which includes deep packet inspection to ensure that benign-looking packets do not create an unstable system state
The intent of this paper is to develop a light-weight intrusion detection system
The likelihood of collision occurring is dependent on the probability that a link is used at the same time by neighbors. It is reduced when the probability of a neighbor communicating decreases. This is the unique feature of the link drop max consensus protocol, when compared to other max consensus protocols, e.g., [32]–[34]; it avoids the sharing of redundant information
Summary
T RADITIONAL distribution grids have seen significant improvements that have made the grid more automated and remotely controllable. A similar concept to provide authentication and integrity is proposed in references [10] and [14] These methods are static; they are not designed to detect other forms of attacks such as Denial of Service, or password hacks. Reference [15] proposes a network-based intrusion detection system (NIDS) which includes deep packet inspection to ensure that benign-looking packets do not create an unstable system state. The authors formulate a bi-level model for coordinating the two types of attacks, which minimizes the attack cost according to a predefined budget, while seeking to maximize the reward This technique does not provide preventive and mitigation steps when an attack happens. In [23], the authors propose a method to detect and correlate attacks, using data collected from IDSs installed at different substations. Their data is sent to a central agent that performs anomaly detection
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
