Adaptive DC bus signalling based control strategy for DC microgrids with multiple Boost converters

The pursuit of climate neutrality has driven the widespread adoption of renewable energy sources (RESs), such as solar photovoltaics (PVs) and wind power generation. Microgrids have emerged as a reliable and efficient solution for integrating these variable and intermittent energy sources. Among them, DC microgrids offer several advantages over AC microgrids, including reduced conversion losses, simpler control mechanisms, no reactive power synchronization issues, and more effective integration of RESs and modern electronic loads.

However, despite these advantages, DC microgrids also face several challenges. Voltage regulation and stability management are more complex due to the absence of frequency-based control mechanisms found in AC systems. Additionally, protection and fault detection in DC networks are challenging, as conventional circuit breakers designed for AC systems may not be effective in DC environments. Furthermore, integration with existing AC infrastructure requires additional power conversion stages, increasing system complexity and potential efficiency losses. These challenges must be carefully addressed to ensure the reliable and scalable deployment of DC microgrids. Control strategies are crucial for optimizing resource utilization and enhancing the performance, stability, and efficiency of microgrids. Currently, centralized control strategies are prevalent, offering coordinated resource management. However, they are prone to significant drawbacks, such as a single point of failure, compromising the entire system's stability and reliability. Additionally, centralized systems often require extensive communication infrastructure, leading to increased costs and complexity.

To address these limitations, distributed control and decentralized control strategies are gaining interest. These strategies distribute decision-making across multiple controllers rather than relying on a single central system, reducing the risk of a single point of failure and enhancing system resilience. Among these techniques, DC bus signaling (DBS) control method, which utilizing the DC bus voltage as a communication medium, presents an intriguing alternative to traditional communication methods. This approach enables efficient coordination between components without the complexity and cost of conventional communication networks.

Based on traditional DBS control, several adaptive DBS control methods have been proposed to ensure DC microgrids can respond effectively to a wide range of operational challenges and environmental changes. The research into DBS and improved methods highlights their potential to support decentralized control strategies, providing a promising avenue for enhancing microgrid resilience and efficiency. However, these methods may lack a dynamic prioritization scheme for managing energy flows among key components like photovoltaic (PV) systems, battery energy storage units (BESUs), and the utility grid (UG). This deficiency can lead to suboptimal energy utilization, increasing reliance on the utility

grid and diminishing the microgrid's self-sufficiency. Meanwhile, applying these methods in a DC microgrid with multiple parallel-connected power converters can generate beat frequency oscillations, causing DC bus voltage fluctuations and disrupting stable microgrid operation.

To address these challenges, this thesis focuses on optimizing DBS control strategies and supressing beat frequency oscillations, ensuring a stable DC bus voltage and efficient, reliable, and sustainable operation of DC microgrids.

Based on the mode-adaptive DBS control strategy, a modified method for DC microgrids with multiple Boost converters is proposed. This strategy introduces a dynamic prioritization mechanism that adaptively adjusts the charging and discharging operations of battery energy storage units (BESUs) based on real-time power availability, load demand, and state-of-charge (SoC) conditions. Unlike conventional approaches with fixed priority rules, the proposed method continuously optimizes energy distribution to enhance microgrid efficiency and resilience.

When PV systems generate surplus power, the strategy prioritizes BESU charging before exporting to the utility grid (UG). During power deficits, it prioritizes BESUs for load supply over sourcing power from the UG. Additionally, an adaptive SoC balancing algorithm ensures uniform charge distribution among BESUs, preventing overcharging or deep discharging, which extends battery lifespan and maintains optimal storage levels.

Furthermore, the proposed strategy incorporates enhanced coordination among multiple Boost converters, dynamically adjusting their operation to optimize power flow and minimize losses. This design not only reduces dependence on the UG but also improves microgrid self-sufficiency and cost-effectiveness while ensuring seamless mode transitions under varying operating conditions.

To suppress beat frequency oscillations, a hybrid method to reduce the beat frequency oscillations is proposed in this thesis. This method combines additional filter inductors with the introduction of switching frequency channels among different converters. Detailed simulation and experiment investigation have been done to study the influence of switching frequency channel setting and inductor selection, improving the flexibility and cost-effectiveness of the solution.