Electric Vehicles and Drones Team Up to Restore Power After Disasters
In the wake of increasingly frequent extreme weather events, the resilience of urban power grids has become a critical concern for cities worldwide. When hurricanes, wildfires, or other natural disasters strike, power outages can cascade through communities, leaving homes and businesses in darkness and disrupting essential services. The challenge is not just in repairing physical damage to power lines, but also in overcoming the invisible yet equally crippling breakdown of communication networks that are essential for managing the grid. Now, a groundbreaking new approach developed by researchers at Xi’an Jiaotong University proposes a novel solution: using electric vehicles (EVs) and drones in a coordinated effort to rapidly restore electricity, even when traditional communication systems are down.
This innovative strategy, detailed in a recent study published in the journal Automation of Electric Power Systems, tackles the complex interdependence between the physical power grid and the digital communication network that controls it. The research team, led by Liu Dafu, Zhong Jian, Yang Qiming, Chen Chen, Li Gengfeng, and Bie Zhaohong, introduces a “cyber-physical collaborative” recovery method that leverages two emerging technologies—vehicle-to-grid (V2G) capabilities and unmanned aerial vehicles (UAVs)—to create a faster, more resilient response to disaster recovery.
The core of the problem lies in the modern power grid’s reliance on a sophisticated information layer. Distribution networks are no longer simple systems of wires and transformers; they are complex cyber-physical systems. To restore power after a disaster, operators need to send commands to remote switches to reconfigure the network, forming isolated “microgrids” around available power sources. However, these commands depend on a functioning communication network. When a disaster knocks out power, it often also disables the cellular towers and data links that the grid relies on, creating a “communication blind spot.” Without communication, operators are blind and powerless, unable to see the state of the grid or control its switches. This creates a vicious cycle: no power means no communication, and no communication means no ability to restore power.
Previous recovery strategies have often treated these two layers—physical power and digital communication—separately. Some methods focus on dispatching mobile generators to provide power, while others look at repairing communication lines. The innovation of this new research is its holistic, integrated approach. It recognizes that the solution to a coupled problem must also be coupled.
The proposed method unfolds in a carefully orchestrated sequence of events, designed to break the deadlock between power and communication. The process begins in the critical moments immediately following a disaster. As power lines fail and communication networks go dark, the first priority is to re-establish a lifeline of communication. This is where the drones come in. The researchers propose deploying drones equipped with portable, high-capacity wireless communication base stations—essentially, flying cell towers.
These drones are dispatched to areas with the most severe communication blackouts. Once in position, they hover at a high altitude and activate their base stations, creating a temporary, localized emergency network. This network is not designed for streaming video or social media; it is a minimal, robust system built for one purpose: restoring control. It provides just enough bandwidth to send and receive the critical control signals needed to operate the grid.
With this emergency communication network in place, the next phase of the recovery can begin. This is where the electric vehicles play their crucial role. The researchers’ model envisions a fleet of EVs, which, during the disaster, have been directed to safe “refuge stations.” These are not just parking lots; they are strategic nodes in the recovery plan. Once the drone establishes a communication link, the central command center can now send signals to these EVs.
The command is simple: drive to a designated V2G station. A V2G station is more than just a charging point; it is a bidirectional interface. When an EV connects, it can not only receive power but also send power back into the grid. By guiding a group of EVs to a V2G station, the system effectively creates a mobile, distributed power source—a “virtual power plant” on wheels.
The brilliance of the strategy lies in the coordination. The drone restores the communication, which allows the command center to guide the EVs. The EVs, once in place, provide the power. This power can then be used to re-energize a local microgrid. The first priority for this power is often the very ground-based communication base stations that were knocked out by the initial power failure. By powering these stations, the system creates a self-sustaining recovery loop. The drone’s temporary network gets the first base station back online, which then provides a more permanent and robust communication link. This stronger link can then be used to coordinate the next wave of drone and EV deployments to the next affected area, allowing the recovery effort to spread outward like a wave.
This method represents a significant leap forward in disaster recovery planning. It transforms EVs from passive consumers of electricity into active participants in grid resilience. Instead of being a burden on the system during an outage, they become a vital resource. The research team emphasizes that this is not a theoretical exercise. They have built a sophisticated mixed-integer linear programming model that simulates the entire process, incorporating the complex constraints of drone flight paths, EV travel times, traffic congestion, and the electrical physics of the distribution network.
To test their model, the researchers used a modified IEEE 33-node distribution network coupled with a 30-node transportation network. Their simulations compared three different recovery scenarios. The first scenario relied only on the small number of EVs that happened to be charging at a V2G station when the disaster struck. This baseline approach restored a limited amount of power, but the supply quickly dwindled as those few EVs depleted their batteries.
The second scenario introduced the drones to restore communication but still relied only on the EVs that were already at V2G stations. The results were dramatically better. By restoring communication, the command center could reconfigure the grid’s topology, connecting more loads to the limited power available. This scenario increased the total power restored by 40.95% compared to the first.
The third and final scenario, which combined both drone-deployed emergency communication and the active re-dispatch of EVs from refuge stations to V2G sites, proved to be the most effective. This fully integrated approach increased the total power restored by a remarkable 92.13% compared to the baseline. The load recovery curve showed multiple sharp increases, corresponding to the moments when a new drone established a link and a new group of EVs arrived at a V2G station, injecting a fresh wave of power into the system.
The researchers also explored the sensitivity of their model to key variables. They found that increasing the number of EVs in the system led to better recovery outcomes, but only up to a point. Once the number of EVs reached 400 in their test case, further increases provided no additional benefit. This is because the system’s capacity is ultimately limited by the number of V2G charging points and the physical constraints of the power lines, which can only carry so much current. This finding is crucial for utilities and city planners, as it provides a clear target for the scale of EV resources needed for effective disaster response.
Another critical insight came from studying the placement of V2G stations. The model showed that stations located on the downstream, “load-side” of the network faults were far more valuable than those upstream. This makes intuitive sense: the areas that are hardest to reach after a fault are the ones that have been disconnected from the main power source. By positioning a V2G station in these isolated areas, the mobile power from the EVs can be used to directly restore power to the communities that need it most. Moving a V2G station upstream, closer to the main substation, actually reduced the total amount of load that could be restored, as it diminished the ability to form microgrids in the isolated zones.
The implications of this research are profound. It suggests a future where our transportation and energy infrastructures are not just connected, but deeply integrated for mutual benefit. In this future, the millions of EVs on the road are not just a solution to climate change; they are also a distributed, mobile battery bank that can be called upon in times of crisis. Similarly, the growing fleet of commercial and industrial drones is not just a tool for delivery or inspection; it is a rapid-response communication force that can be deployed to maintain the nervous system of the city.
The success of this method hinges on several key factors. First, it requires a high level of coordination between different agencies—power utilities, emergency management, and transportation authorities. Second, it depends on the widespread adoption of V2G technology, which allows for bidirectional power flow. While this technology is still in its early stages, it is rapidly gaining traction, with automakers like Ford and Nissan already offering V2G-compatible vehicles in some markets.
Third, and perhaps most importantly, it requires the active participation of EV owners. The model includes an “incentive mechanism” to encourage drivers to respond to the dispatch call. This could take the form of financial compensation, priority access to charging, or other benefits. The researchers acknowledge that human behavior is a complex variable, and their model uses coefficients to represent the “willingness” of drivers to participate. For this system to work in the real world, building public trust and clear communication about the process will be essential.
The potential applications of this technology extend far beyond just restoring power after a storm. It could be used to bolster the grid during periods of peak demand, to provide backup power for critical facilities like hospitals, or to support the integration of renewable energy sources. The ability to create dynamic, temporary microgrids offers a new level of flexibility and control over the distribution network.
The research also highlights the importance of a systems-thinking approach to infrastructure. Modern cities are a web of interconnected systems—power, water, transportation, and communication. A failure in one can cascade into others. The most resilient solutions are those that recognize these interdependencies and design for them from the start. This study is a prime example of such thinking, using the failure of one system (communication) to enable the recovery of another (power) by leveraging a third (transportation).
While the model is complex, the underlying principle is elegant in its simplicity: use what you have, where you need it. When a disaster strikes, the resources are already there—thousands of EVs with charged batteries, and drones capable of flying into the heart of the chaos. The genius of the Xi’an Jiaotong University team is in creating a blueprint for how to orchestrate these resources into a powerful, coordinated response. It is a vision of a smarter, more resilient city, where technology is not just a convenience, but a lifeline.
The work of Liu Dafu, Zhong Jian, Yang Qiming, Chen Chen, Li Gengfeng, and Bie Zhaohong from the State Key Laboratory of Electrical Insulation and Power Equipment at Xi’an Jiaotong University, published in Automation of Electric Power Systems (DOI: 10.7500/AEPS20230731011), provides a compelling roadmap for the future of urban disaster recovery. It is a future where our cars and drones are not just tools of our daily lives, but guardians of our community’s well-being.