Smart Charging Strategy Boosts Grid Efficiency and EV Savings
As electric vehicles (EVs) continue to gain traction across global markets, the integration of these vehicles into existing power grids presents both opportunities and challenges. A recent study published in the Journal of Chongqing University of Technology (Natural Science) offers a promising solution by introducing a multi-timescale optimization strategy that enhances the coordination between power generation, transmission networks, loads, and storage systems—commonly referred to as “source-grid-load-storage” integration. This innovative approach not only improves the economic efficiency of active distribution networks but also significantly increases renewable energy utilization while reducing operational costs for both utilities and EV owners.
The research, led by Jianghong Chen and Xuelian Li from the College of Electrical Engineering and New Energy at China Three Gorges University, introduces a comprehensive framework designed to address one of the most persistent issues in modern power systems: forecasting inaccuracies. Renewable energy sources such as wind and solar are inherently variable, making precise output predictions difficult. Similarly, consumer demand fluctuates throughout the day, influenced by weather patterns, seasonal changes, and human behavior. These uncertainties can lead to imbalances in supply and demand, resulting in inefficiencies such as curtailed renewable generation or excessive reliance on costly peak-load power plants.
To mitigate these risks, the team developed a three-tiered scheduling model that operates across different time horizons—day-ahead, intraday, and real-time—allowing for continuous adjustments based on updated forecasts and actual conditions. The core idea is to progressively refine the initial day-ahead plan using rolling optimization techniques, ensuring that the system remains responsive to changing conditions without sacrificing long-term stability.
The day-ahead phase serves as the foundation of the strategy. During this stage, the distribution network operator creates a 24-hour dispatch plan with hourly resolution, taking into account predicted wind and solar output, expected load demand, and electricity pricing structures. The primary objective is to minimize overall operating costs, which include expenses related to purchasing electricity from the main grid, maintaining distributed generators (DGs), operating energy storage systems (ESS), and compensating flexible loads that participate in demand response programs.
One key aspect of this phase is the incorporation of price-based demand response (DR). By offering time-of-use tariffs, consumers are incentivized to shift their electricity usage away from peak periods to off-peak hours when renewable generation is abundant and electricity prices are lower. For instance, instead of charging their EVs during the evening rush, users may choose to charge overnight when wind power output is high and demand is low. This not only helps flatten the load curve but also reduces stress on the grid during critical periods.
However, price-based DR has limitations. It relies on consumer willingness and behavioral change, which can be unpredictable and difficult to control in real-time scenarios. To overcome this, the researchers introduced incentive-based DR mechanisms that allow the grid operator to directly manage certain types of flexible loads. These are categorized based on their response characteristics, such as adjustability, reaction speed, and notice requirements.
Among the most impactful flexible resources is the fleet of electric vehicles themselves. In the proposed model, EVs are treated not just as passive consumers but as active participants in grid balancing. Through vehicle-to-grid (V2G) technology, EVs can discharge stored energy back into the network during peak demand periods, effectively turning them into mobile energy storage units. However, for this to work efficiently, it’s essential to ensure that EV owners’ mobility needs are fully met. No driver wants to wake up to a half-charged battery when they need to commute.
To address this concern, the model incorporates constraints that guarantee each participating EV reaches its target state of charge before departure. This is calculated based on the vehicle’s daily driving distance and energy consumption rate. Only those EVs that have sufficient time and battery capacity to participate in grid services without compromising their usability are enrolled in the program. According to the study, assuming an 80% participation rate among a fleet of 250 EVs, significant flexibility can be unlocked without negatively impacting user experience.
The next layer of the strategy unfolds during the intraday phase, where the initial day-ahead plan is refined using updated short-term forecasts. This stage operates on a 15-minute resolution and runs every 15 minutes, looking ahead four hours into the future. By this point, more accurate data on wind speeds, solar irradiance, and load patterns become available, allowing for better alignment between supply and demand.
In this phase, additional flexibility resources come into play, including direct-control loads such as water heaters, air conditioners, and industrial processes that can be temporarily interrupted or adjusted upon request. Unlike price-based DR, these loads are contractually obligated to respond to grid signals, providing a higher degree of controllability. The model distinguishes between different classes of incentive-based DR based on their technical capabilities and compensation rates.
For example, Class A loads—primarily industrial users—are suitable for day-ahead scheduling due to their longer response times and larger adjustment capacities. In contrast, Class B and C loads, which include residential thermostatically controlled devices, can respond much faster and are therefore integrated into intraday and real-time operations. The compensation paid to participants is determined by several factors, including the amount of load reduction, response duration, speed, and the criticality of the service, ensuring fair remuneration while maintaining economic efficiency.
A crucial innovation in the intraday model is the inclusion of electric vehicles as dispatchable assets. Instead of simply shifting charging times, EVs now actively contribute to grid stability by discharging during peak periods and absorbing excess renewable energy during low-demand hours. This dual functionality enhances both grid reliability and the economic value of EV ownership.
The final stage—real-time correction—operates on a five-minute cycle, providing fine-tuning adjustments to the intraday schedule. At this point, ultra-short-term forecasts (typically 15 minutes ahead) are used to detect any deviations between planned and actual generation or load. If a mismatch is detected, corrective actions are taken immediately to maintain balance.
For instance, if wind output drops unexpectedly, the system may activate fast-responding resources such as battery storage, microturbines, or Class C direct-control loads to compensate for the shortfall. Conversely, if solar production exceeds expectations, the network can increase EV charging rates or store surplus energy in batteries, minimizing curtailment.
To quantify the impact of these deviations, the model introduces a penalty mechanism based on power imbalance. When actual output exceeds scheduled levels (indicating potential waste), a reduced penalty rate is applied. When output falls short (posing a risk of under-supply), a higher penalty is imposed, encouraging proactive management. This dynamic pricing scheme ensures that the system prioritizes reliability while remaining economically viable.
The effectiveness of the proposed strategy was validated through simulations using real-world data from a high-penetration renewable energy system. The test case included a 7 MW peak load served by 4 MW of wind capacity, 2 MW of photovoltaic (PV) generation, a 0.8 MW microturbine, and a 1.5 MWh/0.3 MW battery storage system. With 250 EVs in the fleet and 15% of total load classified as flexible, the results demonstrated substantial improvements across multiple performance metrics.
Compared to a baseline scenario without multi-timescale coordination, the optimized approach reduced the distribution network’s electricity procurement cost by 29% from the day-ahead to the real-time phase. Load dispatch costs increased slightly due to higher compensation payments for incentive-based DR, but this was more than offset by savings in energy purchases and reduced curtailment of renewable generation.
Electric vehicle owners also benefited significantly. Their charging costs dropped by 46% compared to uncoordinated charging behavior, thanks to strategic scheduling that maximized the use of low-cost, renewable-rich periods. Moreover, the integration of EVs helped boost renewable energy utilization from 92.42% in the day-ahead phase to 98.48% in real-time operations—an improvement of over six percentage points.
These findings underscore the importance of treating EVs not merely as transportation devices but as integral components of a smarter, more resilient energy ecosystem. By enabling bidirectional energy flow and leveraging advanced forecasting and optimization tools, utilities can unlock vast amounts of hidden flexibility within the transportation sector.
Beyond technical achievements, the study highlights the need for supportive policy frameworks and market designs that encourage consumer participation. Transparent pricing signals, reliable communication infrastructure, and standardized V2G interfaces are all essential for scaling up such solutions. Furthermore, privacy and cybersecurity concerns must be addressed to build public trust in automated energy management systems.
From a broader perspective, this research contributes to the ongoing transformation of power systems worldwide. As countries accelerate their transition to clean energy, the ability to integrate variable renewables at scale will determine the pace and success of decarbonization efforts. Traditional centralized models, built around large fossil-fuel plants with predictable output, are increasingly inadequate for managing decentralized, dynamic, and interactive networks.
The multi-timescale optimization approach presented here offers a blueprint for the next generation of smart grids—one where distributed energy resources, intelligent loads, and digital control systems work in harmony to deliver reliable, affordable, and sustainable electricity. It exemplifies how engineering innovation, combined with economic incentives and consumer engagement, can drive systemic change.
Moreover, the methodology is adaptable to various grid configurations and regulatory environments. Whether applied in urban microgrids, rural electrification projects, or industrial parks, the principles of hierarchical scheduling, rolling optimization, and real-time correction remain universally relevant. As battery costs decline and EV adoption rises, the potential for grid-supportive charging will only grow.
Looking ahead, future work could explore the integration of additional distributed resources such as hydrogen production, heat pumps, and building energy management systems. Machine learning techniques could further enhance forecasting accuracy, while blockchain-based platforms might enable peer-to-peer energy trading among prosumers.
Ultimately, the vision is a fully integrated energy landscape where mobility, heating, and power generation are no longer siloed sectors but interconnected elements of a unified, intelligent network. In such a world, every EV parked at a charging station becomes a node in a vast, responsive, and self-optimizing grid—one that balances supply and demand in real time, minimizes waste, and empowers consumers.
This study, conducted by Jianghong Chen, Xuelian Li, Teng Yuan, Ximu Wang, and Weiliang Li from the College of Electrical Engineering and New Energy at China Three Gorges University, represents a significant step toward that future. Published in the Journal of Chongqing University of Technology (Natural Science), the paper provides both theoretical rigor and practical applicability, demonstrating how coordinated control strategies can unlock the full potential of modern power systems. Its findings offer valuable insights for policymakers, utility operators, technology developers, and consumers alike, paving the way for a cleaner, more efficient, and more equitable energy future.
Jianghong Chen, Xuelian Li, Teng Yuan, Ximu Wang, Weiliang Li, College of Electrical Engineering and New Energy, China Three Gorges University; Journal of Chongqing University of Technology (Natural Science), doi: 10.3969/j.issn.1674-8425(z).2024.03.038