An Extreme Value Theory-Based Catastrophe Bond Design for Cyber Risk Management of Power Systems

Abstract

Cybersecurity is emerging as one of the most critical issues for the power system operation in recent years. The most recent studies have shown that cyber-insurance can be an effective solution for the cyber risk management of power grids. In these early attempts, actuarial frameworks and premium schemes were designed for the insurance companies to cope with the risks on power system cybersecurity. However, due to the potentially catastrophic consequences of malicious cyberattacks on power grids, the tail risk events may expose the insurance companies to undue financial risks even if applicable premiums have been designed and collected, which will demotivate the insurance companies from entering the market and providing insurance to the power system stakeholders. In this paper, a Catastrophe (CAT) bond scheme is proposed for insurance companies to address the tail risk of power system cybersecurity by seeking protection from the capital market. The CAT bonds are designed based on the extreme value theory (EVT) to quantify the underlying risk of tail cybersecurity events in power systems. A stochastic model is developed and used in this paper to evaluate the potential losses of power system stakeholders due to malicious cyberattacks on the grids. An example on the IEEE Reliability Test System (RTS-96) was conducted and analyzed to demonstrate the validity and performance of the proposed EVT based CAT bond scheme. The results of the example show that the proposed CAT bond design can effectively manage the insolvency risk of insurers when providing cyber insurance to various stakeholders in power systems.