Abstract
With the increase of dc based renewable energy generation and dc loads, the medium voltage dc (MVDC) distribution network is becoming a promising option for more efficient system integration. In particular, large-capacity photovoltaic (PV)-based power generation is growing rapidly, and a corresponding power conversion system is critical to integrate these large PV systems into MVDC power grid.
Introduction
As the global energy landscape shifts away from traditional synchronous generators, the modern power grid is evolving into what is known as a Power Electronics Dominated Power System (PEDPS). This transition, highlighted in the Chinese Journal of Electrical Engineering (Vol. 7, No. 4, 2021), brings forth significant technical challenges regarding stability, efficiency, and complex system control. This issue explores critical solutions for integrating inverter-based resources into the existing and future infrastructure.
The Shift to MVDC and Renewable Integration
A primary focus of current research is the integration of large-scale renewables, particularly Photovoltaics (PV), into medium-voltage DC (MVDC) grids. MVDC networks offer a more efficient path for power distribution compared to traditional AC grids by reducing conversion stages and improving power quality for DC-native loads and generation.
As shown in the schematic above, the integration of large-capacity PV systems (typically >1 MW) requires sophisticated DC/DC and AC/DC conversion stations to interface with MVDC distribution networks operating at voltages greater than 10 kV. The optimization of these ultra-high voltage step-up converters is essential for maintaining system efficiency.
Advanced Modeling and Control Methodologies
To address the inherent instability of inverter-based systems, researchers are employing several novel modeling and control techniques:
- Control-Loop Decomposition: This method analyzes grid-forming converters by isolating inner-loop voltage and outer-loop power controllers. This allows engineers to identify specific triggers for small-signal instability, ensuring grid-forming robustness across varying short-circuit ratios.
- Multi-Time Scale Modeling: For Electric Vehicle (EV) fast-charging stations, co-simulation techniques balance high-frequency switching accuracy with the need for long-term (day-long) harmonic evaluation.
- Golden Section Search (GSS): By combining GSS with traditional Perturb and Observe (P&O) methods, a faster and more stable Maximum Power Point Tracking (MPPT) algorithm has been developed for PV systems.
- Network-Connected Backbone Routers (NCBR): Prototypes for smart grid architectures in railway systems allow for better management of bidirectional power flows.
Key Results and Experimental Validation
The theoretical models presented in this issue are supported by rigorous experimental and simulation data. One of the standout achievements is a GSS-based MPPT algorithm that reached an overall efficiency of 98.99%, locating the maximum power point in a mere 36 ms.
Furthermore, the assessment of power quality in high-demand scenarios, such as EV fast-charging stations, has been streamlined through Python-supervised co-simulations. These simulations allow for the detailed tracking of harmonic emissions over extended periods.
The data visualized above demonstrates the complexity of harmonic management in modern grids. By simplifying EV chargers into Norton equivalent circuits, researchers can assess these long-term power quality impacts without the prohibitive computational burden of full switching models.
Conclusion and Future Outlook
While significant strides have been made in stabilizing and optimizing PEDPS, challenges remain. Future work must address the "curse of dimensionality" as these stability analyses scale from small test feeders to massive, continent-scale grids. Additionally, while MVDC architectures are theoretically superior for renewable integration, reducing the hardware costs and improving the reliability of high-voltage DC-DC converters remains a priority for the industry. The research presented here provides a foundational framework for the next generation of smart, efficient, and stable power grids.
Summary updated