New Control Strategy Enhances Fast-Charging Station Stability with Flywheel Energy Storage
As electric vehicles (EVs) gain momentum across urban centers worldwide, the infrastructure supporting their rapid adoption must evolve in parallel. One of the most pressing challenges in modern EV ecosystems is the strain placed on power grids by fast-charging stations, particularly during peak demand periods when multiple vehicles initiate high-power charging simultaneously. These transient load surges can lead to significant voltage fluctuations on the DC bus, degrade power quality, and accelerate wear on grid components. In response, researchers are exploring advanced energy storage integration and next-generation control methodologies to stabilize these dynamic systems.
A groundbreaking study recently published in the Journal of Power Supply introduces a novel nonlinear control strategy designed to mitigate the adverse effects of sudden EV charging loads on fast-charging station (FCS) operations. The research, conducted by Han Lei, Wang Yufei, and Xue Hua from the College of Electrical Engineering at Shanghai University of Electric Power, leverages the principles of immersion and invariance (I&I) theory to enhance the performance of flywheel energy storage systems (FESS) integrated within DC fast-charging stations.
Flywheel energy storage has emerged as a promising solution for managing short-term power imbalances due to its high power density, rapid response time, and long operational lifespan. Unlike chemical batteries that degrade with repeated charge-discharge cycles, FESS stores energy kinetically in a rotating mass, enabling it to deliver or absorb large bursts of power within seconds. This makes it particularly well-suited for smoothing the transient power demands associated with EV fast charging, especially during the initial current ramp-up phase when load impact is most severe.
However, traditional control approaches for FESS—typically based on proportional-integral (PI) controllers—struggle under large-signal disturbances. While effective for small perturbations around a steady-state operating point, PI-based strategies often fail to maintain optimal performance when subjected to abrupt load changes, such as those caused by multiple EVs plugging in within a short timeframe. The limitations of classical control methods stem from their reliance on linearized models, which do not adequately capture the nonlinear dynamics inherent in real-world grid interactions.
The research team recognized this gap and sought to develop a control framework capable of handling large-signal transients while ensuring robust stability and fast dynamic response. Their approach centers on the application of immersion and invariance theory—a sophisticated nonlinear control methodology that allows engineers to design control laws by embedding a desired lower-dimensional dynamic behavior into the full system.
In practical terms, the I&I-based strategy constructs a virtual manifold—a mathematical surface in the system’s state space—toward which the system’s trajectory is driven asymptotically. By carefully shaping this manifold, the researchers ensure that the DC bus voltage converges quickly to its reference value, even in the face of significant disturbances. The control law is derived to guarantee that deviations from this manifold decay exponentially, thereby achieving global asymptotic stability.
What sets this strategy apart is its ability to directly link the flywheel’s output current to the dynamics of the DC bus voltage and the mechanical speed of the flywheel rotor. This integration allows the system to respond proactively to both external load variations and internal state changes, such as fluctuations in rotational speed due to energy discharge or recharge cycles. By embedding these physical relationships into the control structure, the method reduces reliance on multiple cascaded control loops and tuning parameters, simplifying the overall architecture while enhancing performance.
The proposed control system was tested in a comprehensive simulation environment using MATLAB/Simulink, modeling a typical DC fast-charging station equipped with a permanent magnet synchronous motor (PMSM)-based FESS. The station configuration included a three-phase AC grid connection, a PWM rectifier for grid-side power conversion, and multiple EV charging points represented as constant resistive loads to simulate realistic fast-charging profiles.
Two distinct test scenarios were evaluated to validate the effectiveness of the new control strategy. In the first, a single EV initiated charging at the 0.5-second mark, simulating a typical user plugging in after arriving at a charging hub. Under conventional PI control, the system exhibited a noticeable voltage sag of nearly 49 volts on the DC bus, accompanied by a sharp spike in grid power demand—reaching a maximum ramp rate of 700 kW/s. In contrast, the I&I-controlled FESS reduced the voltage dip to just 3.1 volts and limited the grid power ramp to only 16.3 kW/s, representing an improvement of over 97% in power smoothing capability.
Moreover, the recovery time for the DC bus voltage was reduced to under 10 milliseconds, demonstrating exceptional dynamic responsiveness. This rapid stabilization ensures that other connected devices and charging units remain unaffected by the transient event, maintaining overall power quality and reliability.
The second test scenario simulated a more demanding operational condition: three EVs connecting to the station at 0.5, 2.5, and 4.0 seconds, respectively. This sequence mimics real-world usage patterns in busy urban charging stations where vehicles arrive in quick succession. Once again, the I&I-based control strategy outperformed both the baseline system (without FESS) and the traditionally controlled FESS.
Under repeated load injections, the enhanced control method maintained a remarkably stable DC bus voltage, with minimal oscillation and no cumulative degradation in performance. The flywheel system not only provided immediate power compensation during each charging event but also efficiently managed its own energy state, transitioning into a recharge mode once the peak demand subsided. This self-regulating behavior prevents the FESS from entering low-speed regimes where its power delivery capability diminishes, ensuring sustained readiness for future load demands.
Perhaps one of the most significant advantages of the proposed strategy is its adaptability to varying operating conditions. Unlike fixed-gain PI controllers, which require retuning when system parameters change, the I&I approach inherently accommodates variations in grid impedance, load characteristics, and flywheel speed. This robustness makes it particularly suitable for deployment in diverse geographical and climatic environments, where ambient temperature, grid strength, and usage patterns can differ significantly.
From a system integration perspective, the control strategy also improves coordination between the FESS and the grid-side converter. By feeding flywheel speed information back into the grid controller, the system enables predictive adjustment of grid power delivery, further reducing stress on the utility network. This bidirectional awareness fosters a more intelligent and responsive energy ecosystem, aligning with broader trends toward smart grid technologies and distributed energy resource management.
The implications of this research extend beyond individual charging stations. As cities move toward electrified transportation networks, the scalability of such control solutions becomes critical. A fleet of fast-charging stations employing advanced FESS controls could collectively act as virtual power plants, providing ancillary services such as frequency regulation, peak shaving, and voltage support to the wider grid. This dual functionality—serving both transportation and grid stability needs—enhances the economic viability of charging infrastructure investments.
Furthermore, the environmental benefits are substantial. By minimizing grid stress and reducing the need for oversized transformers and cabling, the technology supports more sustainable urban development. It also enables greater integration of renewable energy sources, such as solar and wind, into charging station operations. When paired with photovoltaic arrays or wind turbines, the FESS can store excess renewable generation and release it during peak charging hours, reducing reliance on fossil-fuel-based grid power.
The research also highlights the importance of interdisciplinary collaboration in advancing clean energy technologies. The team combined expertise in electrical machine design, power electronics, control theory, and grid integration to develop a holistic solution. Their work exemplifies how theoretical advances in nonlinear dynamics can yield tangible improvements in real-world engineering applications.
From a policy standpoint, the findings support the case for incentivizing the adoption of intelligent energy storage in EV charging infrastructure. Municipalities and utility providers can leverage such technologies to defer costly grid upgrades, improve service reliability, and meet carbon reduction targets. Regulatory frameworks that encourage innovation in grid-edge technologies will be essential in accelerating the transition to a fully electrified transport sector.
Looking ahead, the research team plans to extend their work into experimental validation using physical prototypes. While simulation results are highly encouraging, real-world testing will be crucial to confirm performance under unpredictable environmental conditions and component aging. Future work may also explore hybrid configurations, combining flywheel storage with battery systems to optimize both energy and power delivery characteristics.
Another promising direction involves integrating artificial intelligence techniques to further refine control decisions. Machine learning algorithms could be trained to predict charging demand based on historical data, weather patterns, and traffic flow, enabling even more proactive energy management. Such adaptive systems would represent the next evolution in smart charging infrastructure.
The success of this study underscores the vital role of academic research in addressing complex engineering challenges. Funded by the Shanghai Science and Technology Innovation Action Plan, the project demonstrates how targeted public investment in R&D can yield innovations with broad societal impact. As global EV adoption continues to accelerate, solutions like the one developed by Han, Wang, and Xue will be essential in ensuring that the supporting infrastructure is resilient, efficient, and sustainable.
In conclusion, the introduction of an immersion and invariance-based control strategy for flywheel energy storage in fast-charging stations marks a significant step forward in power system stability and efficiency. By enabling faster, more accurate responses to load transients, the method enhances grid compatibility, improves user experience, and supports the broader goals of decarbonization and energy security. As the world moves toward a cleaner, electrified future, such innovations will play a pivotal role in shaping the infrastructure of tomorrow.
Han Lei, Wang Yufei, Xue Hua, College of Electrical Engineering, Shanghai University of Electric Power, Journal of Power Supply, DOI: 10.13234/j.issn.2095-2805.2024.6.260