Hardening critical components to enhance grid resilience by fusing the data of power networks and natural disasters

Power grids are highly vulnerable to natural disasters that physically damage grid components, leading to cascading failures and restoration challenges. Hardening critical components is an effective method to enhance grid resilience, where resilience assessment and critical component identification are important and should be realized by considering the data of the power network and natural disasters. This paper proposes a gradient-targeted hardening strategy to enhance power grid resilience by improving the survival capability of critical components. Firstly, we introduce a power grid resilience metric that is expressed in a universal form and can assess the grid resilience under various types of natural disasters. Next, we propose a resilience-criticality metric that quantifies the criticality of each component with direct relevance to the grid resilience. Both metrics are estimated using the Monte Carlo method, which significantly reduces computational costs while ensuring high accuracy. Further, we design a gradient hardening resource allocation strategy for enhancing power grid resilience with limited resources. Based on the component resilience-criticality values, the critical components are grouped into different clusters. Each cluster corresponds to a hardening level and the components in the same cluster share the same amount of resources. Simulation results verify the effectiveness of the targeted strategy in enhancing power grid resilience, allowing for cost-effective resource allocation and improved return on investment.