CyberDep: Towards the Analysis of Cyber-Physical Power System Interdependencies Using Bayesian Networks and Temporal Data

TL;DR

This paper introduces CyberDep, an algorithm that models cyber-physical interdependencies in power systems using Bayesian networks and temporal data, aiding in vulnerability analysis.

Contribution

The paper presents a novel Bayesian network-based algorithm, CyberDep, for analyzing cyber-physical interdependencies in power systems using temporal cyber data.

Findings

Visualizes probabilistic relationships in cyber-physical systems

Quantifies interdependencies between cyber and physical components

Demonstrates effectiveness on WSCC 9-bus system data

Abstract

Modern-day power systems have become increasingly cyber-physical due to the ongoing developments to the grid that include the rise of distributed energy generation and the increase of the deployment of many cyber devices for monitoring and control, such as the Supervisory Control and Data Acquisition (SCADA) system. Such capabilities have made the power system more vulnerable to cyber-attacks that can harm the physical components of the system. As such, it is of utmost importance to study both the physical and cyber components together, focusing on characterizing and quantifying the interdependency between these components. This paper focuses on developing an algorithm, named CyberDep, for Bayesian network generation through conditional probability calculations of cyber traffic flows between system nodes. Additionally, CyberDep is implemented on the temporal data of the cyber-physical emulation of the WSCC 9-bus power system. The results of this work provide a visual representation of the probabilistic relationships within the cyber and physical components of the system, aiding in cyber-physical interdependency quantification.