Critical nodes identification and cascading failure mitigation strategies in multi-layer coupled networks of interdependent critical infrastructures.

To mitigate cascading failures in critical infrastructure networks, existing studies have conducted extensive research on the identification and protection of critical nodes. However, they have overlooked the differential characteristics between structural centrality and stability, which may lead to limitations in the selection of critical nodes in multi-layer coupled networks and result in insufficient cascading failure mitigation effects. To address this gap, this study constructs a multi-layer coupled network integrating hospital management coordination, water supply, and power supply, focusing on flow-based physical interactions between layers and the supporting role of management coordination. Based on Coupled PageRank and an improved Station Lyapunov-stability based (SLB) method, critical nodes are identified, and a path-based critical node protection strategy is proposed. The results show that in load-driven coupled networks, unstable nodes are high-risk nodes prone to inducing cascading failures, and the impact of nodes with high structural centrality on the spread of cascading failures cannot be ignored. In addition, based on the proposed protection method, selecting and protecting a small number of critical nodes can effectively mitigate the spread of network cascading failures in the early stage. These findings are helpful for the management and protection of multi-layer coupled infrastructure networks under cascading failures.