Smart Grid Tech Boosts EV Charging Across Power Zones
As electric vehicles (EVs) surge in popularity, cities worldwide face mounting pressure on their power infrastructure. The rapid growth in EV adoption has introduced unprecedented challenges to urban distribution networks, particularly at the local transformer level. Uncoordinated charging behaviors are causing sharp spikes in electricity demand, leading to transformer overloads, increased power losses, and unstable grid conditions. In response, a team of researchers from State Grid Tianjin Electric Power Company and Tianjin University has developed an innovative solution that leverages flexible interconnection technology to enhance the capacity and efficiency of low-voltage power distribution zones.
The study, published in Electric Power Information and Communication Technology, introduces a novel cooperative optimization method for EV charging infrastructure across multiple transformer areas. By integrating intelligent soft open point (SOP) devices into the grid, the approach enables seamless power sharing between adjacent zones, effectively balancing loads and reducing strain on individual transformers. This advancement marks a significant step forward in building smarter, more resilient urban power systems capable of supporting the next generation of electric mobility.
Urban power distribution networks are typically segmented into isolated zones, each served by its own transformer. These zones operate independently, meaning surplus capacity in one area cannot be shared with a neighboring zone experiencing high demand. This rigidity becomes a critical issue as EV ownership rises. When multiple vehicle owners plug in their cars during peak hours—often in the evening after work—it creates concentrated demand that can push transformers beyond their limits. Over time, this overloading reduces equipment lifespan and increases the risk of outages.
The research team, led by Li Juan from the Economic and Technological Research Institute at State Grid Tianjin, recognized that the key to solving this problem lies in connectivity. “Traditional distribution networks lack the flexibility to respond dynamically to fluctuating loads,” Li explained. “With EVs, we’re no longer dealing with a steady, predictable demand pattern. We need a system that can adapt in real time.”
Their solution centers on the deployment of SOPs—advanced power electronics devices that act as intelligent switches between distribution zones. Unlike conventional switches that simply open or close circuits, SOPs can precisely control the flow of active and reactive power between zones. This capability allows them to transfer excess load from an overloaded transformer to a nearby one with available capacity, effectively turning isolated zones into a coordinated network.
The researchers designed a comprehensive optimization model that coordinates EV charging schedules with SOP operations. The model considers multiple factors, including user charging preferences, electricity pricing, grid safety constraints, and overall system economics. It aims to minimize total operational costs while ensuring reliable service for all users.
One of the most innovative aspects of the model is its integration of user behavior. Previous studies often assumed that EV owners would follow centralized charging instructions without question. However, in reality, user compliance depends on satisfaction levels, charging costs, and convenience. To account for this, the team developed a fuzzy inference-based user participation model that predicts how likely drivers are to follow optimized charging schedules.
The model evaluates two key variables: the difference in state of charge (SOC) between uncontrolled and optimized charging scenarios, and the ratio of charging costs under both conditions. If the optimized plan delivers a higher final battery level at a lower cost, users are more inclined to participate. The system uses this insight to adjust its strategy, striking a balance between grid efficiency and user satisfaction.
To validate their approach, the researchers applied it to a real-world low-voltage distribution network in Tianjin, China. The test site included four distinct transformer zones with varying load profiles. Zones 1 and 3 primarily served residential areas with slow-charging EVs, while Zones 2 and 4 supported commercial loads and fast-charging stations. SOPs were installed between Zones 1–2 and Zones 3–4, enabling bidirectional power exchange.
Three scenarios were compared. In the first, EVs charged without any coordination, simulating typical user behavior. In the second, an optimization algorithm managed charging times but without SOP support. The third scenario combined charging optimization with SOP-based load sharing.
The results were striking. In the uncoordinated scenario, transformer load rates in Zones 1 and 3 exceeded safe thresholds for several hours each day, triggering high penalty costs in the simulation. System losses and electricity expenses were also elevated. When optimization was introduced without SOPs, load peaks were reduced, and costs decreased significantly. However, some transformers still experienced overload due to localized demand concentration.
The full integration of SOPs and optimized scheduling eliminated these issues entirely. Transformer load rates remained within safe limits across all zones, system losses dropped by nearly 18% compared to the base case, and total operational costs fell by 87%. Even more impressively, the system achieved zero penalties for load violations and ensured 100% charging satisfaction for users who followed the recommended schedule.
A critical advantage of the SOP-based approach is its ability to manage three-phase imbalance—a common problem in low-voltage networks. In Zone 3, where most EVs were connected to a single phase, uncontrolled charging caused voltage imbalance to exceed acceptable limits (0.5%). Both the base and optimization-only scenarios failed to correct this, with imbalance levels reaching 0.55% and 0.57%, respectively. However, with SOP coordination, the imbalance was reduced to 0.25%, well within the safe range.
The system also demonstrated exceptional responsiveness. During periods of high fast-charging demand in Zone 2, the SOP diverted power from Zone 1, which had surplus capacity. Similarly, when Zone 4 faced heavy loads, the SOP between Zones 3 and 4 redistributed power to prevent overloading. This dynamic load balancing not only improved reliability but also extended the useful life of transformers by reducing thermal stress.
From a user perspective, the system proved highly effective at promoting off-peak charging. By aligning charging times with lower electricity tariffs, the optimized scenarios reduced average charging costs by over 35% compared to uncontrolled charging. The fuzzy logic model ensured that only feasible schedules were proposed, increasing user trust and participation rates.
The economic implications are substantial. For utility operators, the ability to defer costly transformer upgrades by better utilizing existing capacity represents a major financial benefit. For city planners, the technology supports higher EV penetration without requiring extensive grid reinforcement. And for consumers, it enables more affordable and reliable charging.
The study also highlights the importance of modeling accuracy. The original optimization problem involved complex nonlinear equations, which are computationally intensive to solve. To improve efficiency, the researchers applied second-order cone programming (SOCP) techniques to linearize key components of the model. This transformation allowed for faster computation while maintaining high solution accuracy, making the method suitable for real-time applications.
Another strength of the approach is its scalability. While tested on a four-zone network, the underlying principles can be extended to larger systems with multiple SOPs and diverse load types. As urban grids evolve into more interconnected, power-electronics-rich environments, such coordination mechanisms will become essential.
The findings have broader implications for smart city development. As cities integrate more distributed energy resources—such as rooftop solar, battery storage, and EVs—the traditional radial distribution model is becoming obsolete. The future grid must be bidirectional, adaptive, and intelligent. Technologies like SOPs serve as foundational building blocks for this transformation.
Moreover, the research underscores the need for holistic planning. Grid operators cannot treat EVs as isolated loads. Instead, they must consider them as mobile energy assets that can both consume and, in some cases, supply power. The integration of charging optimization with flexible interconnection brings this vision closer to reality.
The environmental benefits are equally significant. By reducing system losses and enabling higher renewable energy utilization, the method contributes to lower carbon emissions. Efficient load management also reduces the need for peaking power plants, which are often fossil-fuel-based and less efficient.
While the technology is promising, challenges remain. The upfront cost of SOP installations is still relatively high, though decreasing as power electronics advance. Regulatory frameworks may need updating to support cross-zone power sharing and cost allocation. Additionally, cybersecurity becomes more critical as grid control systems become more interconnected.
Nevertheless, the path forward is clear. As EV adoption continues to accelerate, utilities must adopt smarter, more flexible approaches to grid management. The work by Li Juan and her colleagues provides a practical, data-driven blueprint for doing so. By combining advanced control algorithms with cutting-edge hardware, they have demonstrated a viable solution to one of the most pressing challenges in modern power systems.
The implications extend beyond China. Cities across North America, Europe, and Asia are grappling with similar issues as EV markets expand. The Tianjin case study offers valuable insights that can be adapted to different regulatory and technical environments. The core principle—using intelligent interconnection to turn isolated grid segments into a unified, responsive network—is universally applicable.
Future research could explore additional dimensions, such as vehicle-to-grid (V2G) integration, where EVs not only charge intelligently but also feed power back to the grid during peak periods. The current model focuses on unidirectional charging, but with minor modifications, it could accommodate bidirectional energy flows.
Another promising direction is the integration of artificial intelligence for predictive control. By analyzing historical charging patterns, weather data, and traffic information, AI models could forecast demand more accurately and proactively adjust SOP operations.
The success of this project also highlights the importance of collaboration between industry and academia. The team included engineers from State Grid Tianjin, one of China’s largest utility companies, and researchers from Tianjin University’s Key Laboratory of Smart Grid, supported by the Ministry of Education. This partnership ensured that the research addressed real-world operational challenges while maintaining scientific rigor.
In conclusion, the study presents a transformative approach to managing EV charging in urban distribution networks. By leveraging flexible interconnection through SOPs and combining it with user-aware optimization, the method significantly enhances grid resilience, economic efficiency, and user satisfaction. It represents a major advancement in smart grid technology and offers a scalable solution for cities aiming to support sustainable transportation.
As the world transitions to cleaner energy systems, innovations like this will play a crucial role in ensuring that the electricity grid evolves in step with changing demand patterns. The days of passive, rigid power distribution are ending. The future belongs to intelligent, adaptive networks capable of supporting the dynamic needs of modern society.
Li Juan, Zhang Liang, Zhang Xuefei, Liu Yingying, Wu Chang, Ruan Jiaao, Ji Haoran, State Grid Tianjin Electric Power Company and Tianjin University, Electric Power Information and Communication Technology, DOI: 10.16543/j.2095-641x.electric.power.ict.2024.10.02