New Simulation Model Advances EV Charging Infrastructure Analysis
As the global electric vehicle (EV) market continues its rapid expansion, the demand for reliable, efficient, and grid-friendly charging infrastructure has never been more critical. With millions of new EVs hitting the roads each year, understanding the complex interaction between charging stations and the power grid is essential for ensuring grid stability, optimizing energy use, and supporting sustainable transportation. In a significant contribution to this field, researchers Yang Chuangchuang and Yu Bo from the State Grid Electric Power Research Institute have developed a highly accurate simulation model for direct current (DC) charging piles based on a two-stage circuit architecture. Their work, published in the Journal of Power Supply, offers a robust tool for engineers, policymakers, and energy planners to evaluate the impact of widespread EV adoption on power systems.
The surge in EV adoption is reshaping the energy landscape. According to recent industry data, sales of battery electric vehicles have grown exponentially, with China leading the charge in both production and deployment. This shift toward electrified mobility brings with it a fundamental change in how energy is consumed. Unlike traditional vehicles that refuel at gas stations, EVs draw power directly from the electrical grid, often at high rates during charging sessions. This introduces new challenges, including increased peak demand, voltage fluctuations, and harmonic distortions that can degrade power quality. To address these issues, it is imperative to develop advanced models that accurately represent the behavior of EV charging equipment under various operating conditions.
At the heart of modern DC fast charging stations lies a two-stage power conversion system. The first stage, known as the AC/DC rectifier, converts alternating current from the grid into a stable direct current. The second stage, a DC/DC converter, then adjusts the voltage level to match the requirements of the vehicle’s battery. The performance of this two-stage system directly influences charging efficiency, power quality, and compatibility with different battery types. However, modeling such systems with high fidelity has proven challenging due to the nonlinear dynamics of power electronics and the variability of real-world loads.
Yang and Yu’s research addresses this challenge by proposing a comprehensive simulation framework that captures the essential characteristics of a DC charging pile. Their model is built on the widely used MATLAB/Simulink platform, making it accessible to a broad range of researchers and industry professionals. The key innovation lies in the careful selection of circuit topologies for each stage of the conversion process. For the front-end AC/DC rectifier, the authors chose the Vienna rectifier—a three-level topology known for its high efficiency, low harmonic distortion, and excellent power factor correction capabilities. For the back-end DC/DC stage, they selected the Buck-Boost converter, which offers the flexibility to both step up and step down voltage, making it ideal for charging batteries with varying voltage requirements.
The choice of the Vienna rectifier is particularly noteworthy. Unlike conventional two-level rectifiers, the Vienna topology reduces the voltage stress on switching devices by half, which enhances reliability and allows for higher power density. Additionally, it eliminates the risk of shoot-through faults—a common issue in bridge circuits—thereby simplifying control and improving safety. These advantages make it well-suited for medium-power applications such as EV charging stations. However, the Vienna rectifier does present a challenge: maintaining balance between the two DC-link capacitors. An imbalance can lead to uneven voltage distribution and reduced performance. To mitigate this, the researchers implemented a sophisticated control strategy that ensures stable operation under dynamic load conditions.
For the DC/DC stage, the selection of the Buck-Boost converter reflects a practical engineering trade-off. While isolated converters such as full-bridge topologies offer galvanic isolation and enhanced safety, they come with higher costs, greater complexity, and larger physical footprints. In contrast, the non-isolated Buck-Boost circuit is simpler, more compact, and less expensive, making it a cost-effective solution for many charging applications. Its ability to operate in both buck (step-down) and boost (step-up) modes allows it to accommodate a wide range of battery voltages, from low-voltage urban EVs to high-voltage long-range models. This versatility is crucial as the EV market diversifies and charging stations must support multiple vehicle types.
To ensure the model’s accuracy and responsiveness, the researchers employed a dual-loop control strategy based on proportional-integral (PI) controllers. This approach combines an outer voltage loop with an inner current loop, enabling precise regulation of the output voltage while maintaining high power quality on the input side. The voltage loop monitors the DC output and adjusts the reference current accordingly, while the current loop ensures that the input current follows a sinusoidal waveform synchronized with the grid voltage—achieving near-unity power factor. Furthermore, the model incorporates feedforward decoupling techniques to handle the inherent coupling between the d-axis and q-axis components in the synchronous reference frame, resulting in faster dynamic response and improved stability.
One of the most compelling aspects of the study is its rigorous validation process. The researchers conducted extensive simulations under various load conditions to verify the model’s performance against industry standards. In tests with purely resistive loads, the system demonstrated excellent voltage regulation, with steady-state output voltages remaining within ±0.5% of the setpoint across different input voltage levels (85%, 100%, and 115% of nominal). This level of precision is essential for ensuring consistent charging performance and protecting sensitive battery management systems.
Equally important is the model’s ability to minimize output voltage ripple—a key indicator of power quality. In all test scenarios, the voltage ripple coefficient remained below 1%, well within acceptable limits for EV charging applications. Excessive ripple can lead to inefficient charging, increased heat generation, and accelerated battery degradation. By maintaining low ripple, the proposed model contributes to longer battery life and higher overall system efficiency.
Another critical performance metric is the total harmonic distortion (THD) of the input current. High harmonic content can cause overheating in transformers and cables, interfere with other equipment, and violate grid code requirements. The simulation results showed that the THD remained below 1.3% under full-load conditions, significantly lower than the 13% threshold specified in relevant standards. This outstanding performance is attributed to the Vienna rectifier’s inherent ability to shape the input current waveform and the effectiveness of the PI-based control algorithm in maintaining sinusoidal current injection.
Beyond static resistive loads, the researchers also tested the model using a dynamic battery representation—the PNGV (Partnership for a New Generation of Vehicles) model. This physics-based battery model captures the nonlinear voltage-current characteristics of lithium-ion cells, including internal resistance, open-circuit voltage, and state-of-charge dependencies. By integrating the PNGV model into the simulation, the team was able to assess the charger’s behavior under realistic charging scenarios, such as constant-current and constant-voltage phases.
The results were equally impressive. When connected to the PNGV battery model, the charging system reached steady state within 0.3 seconds, demonstrating fast transient response. The output voltage stabilized at 600.1 volts with a peak-to-peak ripple of just 10.52 volts, yielding a ripple coefficient of 0.9%—again, well within acceptable limits. The input current THD was measured at 1.1%, confirming that the charger maintains high power quality even when interacting with a dynamic, nonlinear load. These findings validate the model’s applicability to real-world EV charging environments.
The implications of this research extend far beyond the laboratory. Accurate simulation models are indispensable tools for power system planners and grid operators. As EV penetration increases, utilities must anticipate how charging loads will affect distribution networks, especially during evening peak hours when many drivers return home and plug in their vehicles. Without proper modeling, uncoordinated charging could lead to voltage sags, transformer overloads, and increased wear on grid infrastructure.
By providing a validated, high-fidelity model of a DC charging pile, Yang and Yu’s work enables more realistic load flow studies, harmonic analysis, and stability assessments. Engineers can use the model to simulate the impact of different charging strategies—such as time-of-use pricing, smart charging, or vehicle-to-grid (V2G) integration—on local and regional grids. Policymakers can leverage the insights to design incentives that encourage off-peak charging and reduce strain on the system. Moreover, equipment manufacturers can use the model to optimize the design of future charging stations, balancing performance, cost, and grid compatibility.
The research also highlights the importance of interdisciplinary collaboration in advancing sustainable transportation. The development of the model required expertise in power electronics, control theory, battery modeling, and grid integration—fields that are often siloed within academia and industry. By bridging these domains, the authors have created a holistic tool that reflects the complexity of modern energy systems. Their work exemplifies the kind of systems thinking needed to transition to a low-carbon future.
Looking ahead, several avenues for further research emerge from this study. While the current model focuses on steady-state and transient performance, future work could incorporate thermal modeling to assess the impact of temperature on component lifespan and efficiency. Additionally, extending the model to include multiple charging stations operating in parallel could help analyze network-level effects such as load aggregation and harmonic resonance. Integration with renewable energy sources—such as solar photovoltaic systems or wind turbines—would also be valuable, particularly for off-grid or microgrid applications.
Another promising direction is the incorporation of machine learning techniques to enhance control performance. While PI controllers are well-established and reliable, adaptive or predictive control algorithms could further improve efficiency and responsiveness, especially under rapidly changing load conditions. Furthermore, as battery technologies evolve—such as solid-state or lithium-sulfur cells—the model could be updated to reflect new charging characteristics and requirements.
From a practical standpoint, the model’s compatibility with MATLAB/Simulink makes it highly accessible for both academic and industrial users. The platform’s extensive library of blocks and tools allows for easy modification and extension, enabling researchers to tailor the model to specific use cases. Whether used for teaching, research, or product development, the model serves as a valuable resource for advancing the state of the art in EV charging technology.
In conclusion, the work of Yang Chuangchuang and Yu Bo represents a significant step forward in the modeling and analysis of EV charging infrastructure. By combining a well-chosen two-stage circuit topology with advanced control strategies and rigorous validation, they have created a simulation model that accurately captures the behavior of DC charging piles under diverse operating conditions. The model’s ability to meet stringent performance criteria—such as voltage regulation, low ripple, and minimal harmonic distortion—demonstrates its readiness for real-world applications.
As the world moves toward electrified transportation, tools like this will play a crucial role in ensuring that the transition is smooth, efficient, and sustainable. By enabling deeper insights into the interaction between EVs and the grid, this research supports the development of smarter, more resilient energy systems. It also underscores the importance of continued investment in power electronics research and development, which lies at the foundation of the clean energy revolution.
The findings not only advance technical knowledge but also provide actionable insights for stakeholders across the energy and automotive sectors. As governments set ambitious targets for EV adoption and utilities prepare for a more distributed and dynamic grid, studies like this one offer the analytical foundation needed to make informed decisions. In doing so, they help pave the way for a future where electric vehicles are not just a mode of transport, but an integral part of a smarter, cleaner, and more sustainable energy ecosystem.
Yang Chuangchuang, Yu Bo, State Grid Electric Power Research Institute, Journal of Power Supply, DOI: 10.13234/j.issn.2095-2805.2024.4.74