Smart Charging Strategy Balances Grid, Station, and User Interests
As the global push for electrified transportation accelerates, electric vehicles (EVs) are no longer just a symbol of clean energy—they are becoming a central component of modern power systems. With millions of EVs hitting the roads, their charging behavior is increasingly influencing grid stability, utility operations, and consumer costs. Uncontrolled, EV charging can exacerbate peak demand, strain distribution networks, and increase electricity prices. However, when intelligently managed, EVs can serve as flexible resources that support grid resilience, reduce operational costs, and deliver savings to end users.
A groundbreaking study published in Electric Drive introduces a novel approach to EV charging management that successfully aligns the often-competing interests of power grids, charging station operators, and vehicle owners. The research, led by Zhou Yufan and Professor Gao Hui from the College of Automation and College of Artificial Intelligence at Nanjing University of Posts and Telecommunications, in collaboration with Long Yi from State Grid Chongqing Power Supply Company, proposes a customized charging strategy grounded in dynamic pricing and advanced optimization algorithms.
Titled Customized Charging Strategy for EV Considering Dynamic Electricity Prices, the paper presents a holistic framework designed to transform EVs from potential grid stressors into valuable assets. By integrating real-time grid conditions, user preferences, and economic incentives, the strategy enables a more balanced, efficient, and sustainable charging ecosystem.
The Challenge of Uncontrolled EV Charging
The rapid adoption of EVs is reshaping energy demand patterns. Unlike traditional appliances, EVs require high-power, long-duration charging, often concentrated during evening hours when drivers return home. This behavioral trend leads to a phenomenon known as “load clustering,” where a surge of vehicles plug in simultaneously, creating sharp peaks in electricity demand.
These peaks strain local distribution infrastructure, particularly at the neighborhood level, where transformers and feeders may not be designed to handle such concentrated loads. The consequences are tangible: increased wear on equipment, higher electricity losses, reduced power quality, and in extreme cases, overloads that could lead to outages. Moreover, utilities may be forced to invest in costly infrastructure upgrades to accommodate growing EV fleets.
From the user’s perspective, uncontrolled charging often results in higher electricity bills, especially when vehicles are charged during peak tariff periods. While time-of-use (TOU) pricing schemes have been introduced in many regions to encourage off-peak charging, they are static and do not reflect real-time grid conditions. As a result, their effectiveness in managing EV load is limited.
Charging station operators, meanwhile, face a dual challenge. On one hand, they must attract and retain customers by offering competitive pricing and reliable service. On the other, they must manage their energy procurement costs and avoid penalties from grid operators for contributing to peak demand. Striking the right balance between customer satisfaction and operational profitability is a complex task.
Existing solutions have typically addressed only one or two of these stakeholders’ needs. Some focus on grid-side optimization without sufficient regard for user convenience. Others prioritize user cost savings but neglect the broader impact on the distribution network. Few have attempted to create a truly integrated solution that benefits all parties equally.
A Tripartite Approach to Charging Optimization
The research team recognized that a sustainable EV charging ecosystem must account for the interdependence of the grid, the charging station, and the end user. Their proposed strategy is built on the principle of mutual benefit—ensuring that no single stakeholder bears an undue burden while all gain from the optimization process.
At the core of the strategy is a dynamic pricing mechanism that moves beyond traditional TOU models. Instead of fixed rates based on time of day, the new pricing model adjusts electricity costs in real time based on the actual load on the local transformer. When the transformer is lightly loaded, prices are low, encouraging users to charge. When the transformer approaches its capacity, prices rise, discouraging additional load and preventing overloads.
This dynamic pricing is not arbitrary; it is carefully calibrated to reflect four operational zones: off-peak, mid-peak, peak, and critical peak. These zones are defined by the transformer’s load ratio—the percentage of its maximum capacity currently being used. By linking pricing directly to grid conditions, the system provides a powerful economic signal that guides user behavior in a way that supports grid stability.
For example, if a neighborhood experiences a sudden spike in demand due to evening cooking loads, the transformer’s load ratio increases, triggering a price adjustment. EV owners who are flexible in their charging schedules are incentivized to delay charging until the load decreases, thereby avoiding the most expensive periods. This creates a self-regulating mechanism that naturally smooths out demand fluctuations.
But dynamic pricing alone is not enough. Users may lack the time, knowledge, or willingness to manually adjust their charging behavior. To address this, the researchers developed a customized charging optimization model that automatically generates an ideal charging plan for each vehicle based on its specific needs.
Personalized Charging Plans for Maximum Efficiency
The customization process begins when a user connects their EV to a smart charging station. The system retrieves key information: the vehicle’s battery capacity, current state of charge (SOC), the user’s desired SOC upon departure, and the expected departure time. With this data, the algorithm calculates the total energy required and the available charging window.
Rather than charging at maximum power as quickly as possible, the system determines the optimal charging profile—how much power to draw at each 15-minute interval—over the entire charging period. The goal is twofold: minimize the user’s total charging cost and reduce the fluctuation in the local grid’s load.
These two objectives are often in tension. Charging at the cheapest possible rate might involve drawing power during very low-load periods, which could create new imbalances if many users do the same. Conversely, charging in a way that perfectly smooths the grid load might require using more expensive electricity. The optimization model resolves this conflict by combining both goals into a single weighted objective function, allowing operators to fine-tune the balance between cost savings and grid stability.
To solve this complex multi-variable optimization problem, the researchers enhanced the Artificial Bee Colony (ABC) algorithm—a nature-inspired metaheuristic known for its global search capabilities. The standard ABC algorithm can be slow to converge and prone to getting stuck in local optima, especially in high-dimensional problems like EV charging scheduling.
To overcome these limitations, the team introduced an adaptive normal decay coefficient that dynamically adjusts the search behavior based on the iteration count. In the early stages of optimization, the algorithm explores a wide range of solutions to avoid premature convergence. As the process progresses, it gradually focuses on refining the most promising solutions, accelerating convergence toward the global optimum.
This adaptive ABC algorithm proved significantly more effective than its traditional counterpart in simulation tests. It reached optimal charging plans faster and with greater consistency, demonstrating its suitability for real-world deployment in dynamic environments.
Real-World Simulation and Performance Evaluation
To validate their approach, the researchers conducted extensive simulations based on real-world data from a residential community in East China. The test site featured a 1,250 kVA transformer serving a typical urban neighborhood, with historical load profiles reflecting daily household energy use.
They compared two charging strategies: Strategy 1, representing the current norm, combined static time-of-use pricing with uncontrolled (unordered) charging. Strategy 2 implemented the proposed dynamic pricing and customized charging optimization.
Simulations were run for scenarios with 50, 100, and 150 EVs—representing low, medium, and high adoption levels. The results were analyzed across three key metrics: grid load quality, user charging costs, and charging station profitability.
From the grid’s perspective, the benefits of Strategy 2 were dramatic. Under Strategy 1, EV charging exacerbated existing load peaks, increasing the peak-to-valley difference from 376 kW (with 50 EVs) to 752 kW (with 150 EVs). Load fluctuation rates rose from 37.12% to 53.27%, and the transformer’s maximum load ratio reached 74.65%—dangerously close to the 80% safety threshold.
In contrast, Strategy 2 effectively flattened the load curve. Even with 150 EVs, the peak-to-valley difference was reduced to 422 kW, load fluctuation dropped to 32.22%, and the maximum load ratio fell to 57.07%. This improvement means the same transformer can support more EVs without requiring upgrades—extending its operational life and deferring capital expenditures.
The enhanced load management also unlocked additional charging capacity. The study found that under Strategy 2, the transformer could safely accommodate up to 250 EVs—nearly 100 more than under Strategy 1. This represents a significant increase in service capacity without any hardware investment.
For users, the financial benefits were equally compelling. Under Strategy 1, the average charging cost hovered around 1.02 yuan per kWh across all scenarios. Strategy 2 reduced this cost to 0.8168 yuan/kWh with 50 EVs, 0.869 yuan/kWh with 100 EVs, and 0.9368 yuan/kWh with 150 EVs. Even in the most congested scenario, users saved nearly 7% on their charging bills.
Crucially, these savings were achieved without compromising user convenience. Drivers retained full control over their departure times and target battery levels. The system automatically adjusted the charging schedule to meet these preferences at the lowest possible cost, eliminating the need for manual intervention.
Charging station operators also emerged as clear beneficiaries. While they collected a slightly lower energy charge due to off-peak charging, they offset this through a stable service fee of 0.45 yuan/kWh and a small subsidy from the grid operator for their load-balancing services.
With 150 EVs, the station’s daily revenue reached 2,275 yuan, up from 1,508 yuan with 100 EVs—demonstrating strong scalability. More importantly, the station contributed to grid stability, enhancing its value as a responsible energy partner and potentially qualifying for additional incentives in future regulatory frameworks.
A Scalable Model for the Future of Mobility
The success of this strategy lies in its practicality and scalability. It does not require new hardware or changes to existing grid infrastructure. Instead, it leverages smart meters, communication networks, and cloud-based optimization—technologies already being deployed in modern smart grids.
The dynamic pricing mechanism is transparent and easy to understand: higher load means higher prices, lower load means lower prices. This simplicity encourages user trust and participation. The automated optimization removes the burden from users, making efficient charging the default rather than the exception.
Moreover, the framework is inherently adaptable. The weight given to cost savings versus load smoothing can be adjusted based on local conditions. In areas with aging infrastructure, grid stability might be prioritized. In newer developments with robust equipment, cost savings could take precedence.
The model also lays the groundwork for future advancements. As vehicle-to-grid (V2G) technology matures, the same optimization engine could be extended to manage bidirectional energy flows, turning EVs into mobile energy storage units that provide grid services such as frequency regulation and emergency backup.
Regulators and utility planners should take note of this research. It demonstrates that with the right incentives and intelligent control, EVs can be a solution to grid challenges rather than a source of them. Policies that encourage dynamic pricing, support smart charging infrastructure, and reward load-balancing services will be essential to unlocking this potential.
For automakers and charging network providers, the study offers a blueprint for next-generation charging services. Integrating such optimization into vehicle infotainment systems or mobile apps could become a key differentiator, offering users not just convenience but tangible financial and environmental benefits.
Conclusion: A Win-Win-Win for the Energy Transition
The transition to electric mobility is not just about replacing internal combustion engines with batteries. It is about reimagining the relationship between transportation and energy. This research by Zhou Yufan, Gao Hui, and Long Yi represents a significant step in that direction.
By harmonizing the interests of the grid, the charging station, and the user, their customized charging strategy proves that smart management can turn a potential problem into an opportunity. It shows that with the right algorithms and economic signals, EVs can help stabilize the grid, reduce costs for consumers, and create sustainable business models for service providers.
As EV adoption continues to grow, such integrated solutions will become increasingly vital. The days of treating EV charging as a simple plug-and-play activity are ending. The future belongs to intelligent, adaptive, and cooperative systems that make the most of every kilowatt-hour.
The work published in Electric Drive offers a compelling vision of that future—one where electric vehicles are not just driven, but intelligently integrated into the fabric of the energy system.
Zhou Yufan, Gao Hui, Long Yi, Nanjing University of Posts and Telecommunications, State Grid Chongqing Power Supply Company, Electric Drive, DOI: 10.19457/j.1001-2095.dqcd25189