Cyberphysical Security and Dependability Analysis of Digital Control Systems in Nuclear Power Plants

TLDR

The use of nuclear energy is essential for modern power demands, and newly built nuclear power plants are shifting from analog to digital control systems, making cyber‑physical security and dependability critical. The study proposes layered protection strategies, including a cyber framework that blocks attacks and meets regulations, a physical framework to mitigate physical threats, and a discussion of reliability, maintainability, and availability metrics. The authors employ generalized stochastic Petri nets to quantify intrusion probabilities and develop cyber and physical protection frameworks, while evaluating dependability via reliability, maintainability, and availability metrics. A case study shows the cyber framework achieves high dependability, evidenced by favorable steady‑state probability analysis.

Abstract

The use of nuclear energy to generate electric power is crucial to meet the high energy demand of a modern economy. In newly constructed nuclear power plants (NPPs), the trend among control systems is to replace the obsolete analog hard-wired systems with the contemporary digital and cyber-based systems. Therefore, cyberphysical security as well as dependability are critical issues in safety critical NPPs. In this paper, we present different levels/layers of protection to manage cyber/physical security. We also discuss the interrelationship between cyber and physical attacks. We adopt generalized stochastic Petri nets to quantitatively evaluate the intrusion probability. We then propose a new cyberframework and show that the proposed framework not only prevents cyberattacks but also conforms to cybersecurity regulations. We also propose a physical framework to prevent potential physical attacks. Finally, we discuss dependability through three metrics, i.e., reliability, maintainability, and availability. A case study is presented to demonstrate that the proposed cyberframework is highly dependable through analyzing steady-state probabilities.