Abstract:Infrastructure systems are critical in modern communities but are highly susceptible to various natural and man-made disasters. Efficient post-disaster recovery requires repair-scheduling approaches under the limitation of capped resources that need to be shared across the system. Existing approaches, including component ranking methods, greedy evolutionary algorithms, and data-driven machine learning models, face various limitations when tested within such a context. To tackle these issues, we propose a novel approach to optimize post-disaster recovery of infrastructure systems by leveraging Deep Reinforcement Learning (DRL) methods and incorporating a specialized resilience metric to lead the optimization. The system topology is represented adopting a graph-based structure, where the system's recovery process is formulated as a sequential decision-making problem. Deep Q-learning algorithms are employed to learn optimal recovery strategies by mapping system states to specific actions, as for instance which component ought to be repaired next, with the goal of maximizing long-term recovery from a resilience-oriented perspective. To demonstrate the efficacy of our proposed approach, we implement this scheme on the example of post-earthquake recovery optimization for an electrical substation system. We assess different deep Q-learning algorithms to this end, namely vanilla Deep Q-Networks (DQN), Double DQN(DDQN), Duel DQN, and duel DDQN, demonstrating superiority of the DDQN for the considered problem. A further comparative analysis against baseline methods during testing reveals the superior performance of the proposed method in terms of both optimization effect and computational cost, rendering this an attractive approach in the context of resilience enhancement and rapid response and recovery.

Stochastic Planning for Resilient Infrastructure of Distribution System under Extreme Weather Events

Markov Decision Process-based Resilience Enhancement for Distribution Systems: an Approximate Dynamic Programming Approach

A Bi-Level Coordinated Dispatch Strategy for Enhancing Resilience of Electricity-Gas System Considering Virtual Power Plants

A deep reinforcement learning-based resilience enhancement framework for distribution networks under extreme weather events

Deep Reinforcement Learning Framework for Resilience Enhancement of Distribution Systems under Extreme Weather Events

A Bayesian Deep Reinforcement Learning-Based Resilient Control for Multi-Energy Micro-Gird

Resilience-driven Planning of Electric Power Systems Against Extreme Weather Events

Deep Reinforcement Learning for Distribution System Restoration Using Distributed Energy Resources and Tie-Switches

Planning for Distribution Resilience under Variable Generation: Prevention, Surviving and Recovery

Risk-informed Resilience Planning of Transmission Systems Against Ice Storms

A Deep Reinforcement Learning Framework for Automatic Operation Control of Power System Considering Extreme Weather Events

Robustness Assessment and Enhancement of Deep Reinforcement Learning-Enabled Load Restoration for Distribution Systems.

Resilience Enhancement of Distribution Grids Against Extreme Weather Events

Resilience Enhancement Strategies for Active Distribution Grid Fault Recovery in Collaboration with Dynamic Network Reconfiguration

Risk-Based Active Distribution System Planning for Resilience Against Extreme Weather Events

Enhancing Cyber-Resilience in Integrated Energy System Scheduling with Demand Response Using Deep Reinforcement Learning

Power System Emergency Control to Improve Short-Term Voltage Stability Using Deep Reinforcement Learning Algorithm

Equity-driven distribution power system planning for resilience enhancement

Dynamic Resilience Region-Based Proactive Scheduling for Enhancing the Power System Resilience

Resilience-based post disaster recovery optimization for infrastructure system via Deep Reinforcement Learning

Unified Two-Stage Reconfiguration Method for Resilience Enhancement of Distribution Systems