New Strategy Cuts Energy Costs for 5G and EVs
As the global push for electrification and digital connectivity accelerates, two of the most transformative technologies—electric vehicles (EVs) and 5G networks—are converging on a collision course with energy infrastructure. The surge in EV adoption and the rollout of 5G infrastructure have created unprecedented demand on power grids, posing challenges for energy efficiency, cost management, and network reliability. A groundbreaking study led by Kai Chen from North China Electric Power University, in collaboration with Yu Fu of Guizhou Power Grid, introduces an innovative energy management framework that could reshape how 5G base stations and EVs coexist within the modern power ecosystem. Published in the Transactions of China Electrotechnical Society, this research proposes a real-time, collaborative strategy that leverages vehicle computing power to reduce energy costs and improve network performance.
The integration of 5G technology into transportation systems is more than just a bandwidth upgrade—it represents a fundamental shift in how vehicles interact with their environment. 5G enables ultra-low latency communication, essential for vehicle-to-everything (V2X) applications, autonomous driving, and smart traffic management. However, this technological leap comes at a steep energy cost. Each 5G macro base station (MBS) can consume up to 12,000 kWh annually, with cooling systems alone accounting for nearly 30% of that consumption. Meanwhile, the global EV fleet is projected to reach tens of millions within the next few years, each requiring regular charging and generating additional strain on distribution networks. When these two forces intersect, the result is a complex web of interdependent energy and data flows that traditional grid management systems are ill-equipped to handle.
Chen and Fu’s work addresses this challenge by reframing EVs not just as energy consumers, but as active participants in the energy network. Their model, grounded in advanced optimization theory, treats parked EVs with idle computing capacity as distributed edge computing nodes that can offload processing tasks from nearby 5G base stations. This approach, termed “computing hotspot transfer,” allows base stations to reduce their computational load, which in turn lowers heat generation and the energy required for cooling. By shifting computation to vehicles, the system achieves a dual benefit: base stations save on electricity, and EV owners are compensated for providing computing services, creating a mutually beneficial ecosystem.
At the heart of the proposed strategy is a sophisticated online management algorithm that operates without requiring prior knowledge of future network conditions—a critical advantage in real-world deployment. Unlike traditional optimization methods that rely on day-ahead forecasts, this approach uses an improved Lyapunov optimization framework to make real-time decisions based on current system states. The algorithm continuously monitors variables such as real-time electricity prices, network traffic, ambient temperature, and EV availability, dynamically adjusting energy flows to minimize long-term costs. This adaptability is particularly valuable in environments with high uncertainty, such as those influenced by renewable energy generation or fluctuating user demand.
One of the most innovative aspects of the research is its recognition of the deep coupling between information and energy flows in the 5G-EV ecosystem. Previous studies often treated energy management and network optimization as separate domains. Chen and Fu’s model, however, acknowledges that changes in one directly affect the other. For example, when a base station reduces its computational load by offloading tasks to an EV, it not only saves energy but also frees up communication bandwidth. This additional capacity can be used to provide higher-quality service to EVs, ensuring that charging commands from grid operators are transmitted with minimal delay. Conversely, if communication delays occur, they can disrupt the timing of EV charging, leading to inefficiencies and increased costs. The new strategy explicitly accounts for this interdependence, creating a feedback loop that optimizes both energy and data performance.
To ensure fairness and incentivize participation, the researchers designed a Stackelberg game-based mechanism that governs the interaction between EV owners and network operators. In this model, EV owners act as leaders, setting the price for their computing resources, while base stations act as followers, deciding how much computing load to offload based on those prices. This market-like structure ensures that EV owners are fairly compensated for their contribution, while operators can strategically allocate resources to minimize expenses. The game-theoretic approach guarantees a Nash equilibrium, meaning that neither party can unilaterally change their strategy to achieve a better outcome, thus promoting stability and cooperation.
The model also incorporates safety and security considerations, which are often overlooked in academic research but are paramount in real-world deployment. In dense urban environments, EVs may connect to multiple base stations, creating potential vulnerabilities to cyberattacks. To mitigate this risk, the algorithm includes a privacy entropy metric that measures the randomness of EV-to-base-station connections. By encouraging a more uniform distribution of connections across the network, the system reduces the likelihood of targeted attacks and enhances overall communication security. This feature is particularly important for critical grid operations, where data integrity and reliability are non-negotiable.
Simulation results demonstrate the effectiveness of the proposed strategy across a range of scenarios. In a test environment with three 5G base stations and 70 EVs, the method reduced the time-averaged electricity cost for base stations by over 10% compared to conventional approaches. Even more impressively, the cost savings were achieved without increasing the charging burden on EVs, thanks to the compensation mechanism. The system also showed robust performance under varying conditions, including sudden spikes in network traffic and different renewable energy configurations. In scenarios with high solar penetration, the algorithm effectively coordinated photovoltaic generation, battery storage, and computing offloading to maximize cost efficiency.
The implications of this research extend beyond the immediate context of 5G and EVs. As smart cities evolve, the integration of transportation, communication, and energy systems will become increasingly common. The principles demonstrated in this study—real-time optimization, distributed resource utilization, and cross-domain coordination—can be applied to other sectors, such as smart buildings, industrial automation, and microgrids. By treating computing and communication infrastructure as flexible energy assets, cities can build more resilient and sustainable systems that adapt to changing conditions in real time.
Moreover, the strategy aligns with broader environmental goals. By reducing the energy consumption of 5G base stations, it lowers carbon emissions associated with network operations. It also supports the integration of renewable energy by providing additional flexibility in demand-side management. As governments and corporations commit to net-zero targets, solutions like this one will play a crucial role in decarbonizing the digital economy.
From a policy perspective, the research highlights the need for regulatory frameworks that encourage collaboration between different stakeholders in the energy and telecommunications sectors. Currently, EV charging, grid operations, and mobile network management are often governed by separate entities with different priorities. A unified approach, supported by advanced algorithms like the one proposed by Chen and Fu, could unlock significant efficiency gains and cost savings. Policymakers should consider creating incentives for data sharing, interoperability standards, and joint planning initiatives that enable such synergies.
For industry, the findings suggest new business models centered around vehicle-to-grid (V2G) and vehicle-to-network (V2N) services. Automakers could offer computing-as-a-service features in their EVs, allowing owners to earn revenue while their vehicles are parked. Telecommunications companies could integrate energy management into their network operations, offering premium connectivity services to EV users in exchange for computing access. Energy providers could use the system to balance load and reduce peak demand, improving grid stability and deferring costly infrastructure upgrades.
The scalability of the solution is another key advantage. The simulations show that as the number of EVs and base stations increases, the benefits of collaboration grow even more pronounced. In larger networks, the availability of idle computing resources increases, providing greater opportunities for load shifting and cost reduction. This scalability makes the strategy well-suited for deployment in metropolitan areas, where both EV density and 5G coverage are expected to be highest.
Despite its many strengths, the approach is not without challenges. One potential limitation is the reliance on EV owners’ willingness to participate in computing offloading. While the compensation mechanism addresses economic incentives, user experience and privacy concerns must also be considered. Future work could explore ways to make participation seamless and transparent, perhaps through automated enrollment or opt-in programs tied to loyalty rewards.
Another area for further research is the integration of different types of edge computing devices. While the current model focuses on EVs, other mobile platforms such as buses, trucks, and even smartphones could also contribute computing power. Expanding the scope of the system to include these devices could further enhance its flexibility and resilience.
In conclusion, the research by Kai Chen and Yu Fu represents a significant step forward in the convergence of transportation, communication, and energy systems. By reimagining EVs as active participants in the digital infrastructure, their strategy unlocks new possibilities for energy efficiency, cost reduction, and network optimization. As the world moves toward a more connected and electrified future, innovations like this one will be essential for building sustainable, intelligent, and resilient systems that serve the needs of both people and the planet.
Kai Chen, Yu Fu. Transactions of China Electrotechnical Society. DOI: 10.19595/j.cnki.1000-6753.tces.231896