Multi-agent Power Grid Restoration Under Uncertainty Considering Coupled Transportation-Power Networks

TL;DR

This paper tackles post-disaster power grid restoration under uncertainty by coupling the power distribution network with the transportation network used by repair crews. It introduces a two-stage stochastic mixed-integer programming framework where Stage 1 pre-positions and assigns heterogeneous repair crews across multiple pre-generated damage scenarios, and Stage 2 solves a multi-agent capacitated vehicle routing problem to dispatch crews to damaged nodes under each realized scenario. Key contributions include modeling multiple crew types as distinct agents, integrating a realistic road network with IEEE 8500-node feeder data mapped to the DFW area, and employing scenario decomposition to manage computational complexity. The framework enables proactive, data-driven restoration planning that can adapt to uncertain damages, transportation disruptions, and crew mobilization constraints, with potential to improve response times and service restoration in real-world utilities.

Abstract

Restoring power distribution systems after extreme events such as tornadoes presents significant logistical and computational challenges. The complexity arises from the need to coordinate multiple repair crews under uncertainty, manage interdependent infrastructure failures, and respect strict sequencing and routing constraints. Existing methods often rely on deterministic heuristics or simplified models that fail to capture the interdependencies between power and transportation networks, do not adequately model uncertainty, and lack representation of the interrelated dynamics and dependencies among different types of repair crews--leading to suboptimal restoration outcomes. To address these limitations, we develop a stochastic two-stage mixed-integer programming framework for proactive crew allocation, assignment, and routing in power grid restoration. The primary objective of our framework is to minimize service downtime and enhance power restoration by efficiently coordinating repair operations under uncertainty. Multiple repair crews are modeled as distinct agents, enabling decentralized coordination and efficient task allocation across the network. To validate our approach, we conduct a case study using the IEEE 8500-node test feeder integrated with a real transportation network from the Dallas-Fort Worth (DFW) region. Additionally, we use tornado event data from the DFW area to construct realistic failure scenarios involving damaged grid components and transportation links. Results from our case study demonstrate that the proposed method enables more coordinated and efficient restoration strategies. The model facilitates real-time disaster response by supporting timely and practical power grid restoration, with a strong emphasis on interoperability and crew schedule coordination.

Figures (8)

• Figure 1: Overview of the proposed process for restoring the distribution grid under uncertainty. The framework begins with scenario generation using tornado history, power grid, and transportation network data. In Stage 1, repair crew types and capacities are allocated in anticipation of damage. In Stage 2, crews are routed to damaged nodes and scheduled using Gantt charts, reflecting a realistic and coordinated multi-crew restoration plan. The approach integrates infrastructure interdependencies and stochastic failure modeling to support proactive and efficient decision-making.
• Figure 2: Scenario generation based on variations in repair times ($\mathcal{T}$), repair demands ($\mathcal{D}$), and transportation network availability. Scenarios are derived using historical tornado data, power grid infrastructure (including transformers, substations, switches, and distribution lines), and transportation network data. Each scenario captures different combinations of failures to reflect realistic post-disaster conditions and their impact on optimal crew routing from depots ($\mathbb{D}$) to damaged nodes ($\mathbb{N}$).
• Figure 3: Gantt chart illustrating the sequential deployment of specialized repair crews during power grid restoration. The timeline shows how inspection and assessment, tree trimming, line repairs, and final inspections are coordinated over a 24-hour period to ensure efficient and timely recovery.
• Figure 4: Projection of the IEEE 8500-node distribution test network on the DFW transportation network. A complete graph constructed from the damaged nodes and depots.
• Figure 5: Second stage decisions: Crew-2 (Line crew) routing within different scenarios