A post-disaster restoration model for power-gas-transportation distribution networks considering spatial interdependency and energy hubs.

• A coordinated restoration model is proposed for power-gas-transportation distribution systems. • Repair crews and energy hubs are simultaneously optimized to enhance post-disaster resilience. • Spatial interdependency between power-gas systems and transportation systems is considered to ensure the feasibility of the obtained restoration decisions. • Modified Boolean logic equations are proposed to represent the spatial interdependency. Recently, extreme events have caused widespread blackouts over energy systems. In addition, the increasing interactions between power distribution networks and natural gas networks bring new challenges to the fast recovery of energy systems. Mobile emergency resources, such as repair crews (RCs), play an essential role in the restoration. However, power distribution networks and gas networks have geographical correlations with transportation systems in urban areas, therefore, there exists spatial interdependency between infrastructural systems. This property results in the dependency between RCs' routings and road states in transportation systems. To this end, this paper proposes a sequential restoration model for power-gas-transportation distribution networks. The post-disaster restoration model contains the dispatch of both RCs and fixed resources. The spatial interdependency is modeled by modified Boolean logic equations to obtain feasible recovery schedules in practice. Convex relaxations and linearization methodologies are utilized to transform the proposed model into a mixed-integer second-order cone programming (MISOCP) model. Simulations are carried out on 33–14–20 and 136–20–39 bus coupled systems. Numerical results indicate that the proposed restoration strategy could increase the post-disaster resilience index by 19.5 % in 33-bus power system, 8.7 % in 136-bus power system, 25.6 % in 14-node gas system, and 27.2 % in 20-node gas system, respectively. [ABSTRACT FROM AUTHOR]