TL;DR
This paper presents a comprehensive three-layer hierarchical control architecture for islanded DC microgrids (DCmGs) that addresses the challenge of coordinating distributed generation units (DGUs), loads, and flexible network structures. The proposed system ensures stable operation and optimal energy management by integrating a primary voltage control layer, a secondary power-to-voltage translation layer, and a tertiary MPC-based Energy Management System (EMS).
What Problem Does This Paper Address?
Islanded DC microgrids, while offering numerous advantages, present complex control challenges. Ensuring stable voltage, efficient energy distribution, and optimal resource utilization across various generation sources (like renewables and batteries) and diverse loads (including nonlinear ones) is difficult. Existing approaches often focus on individual control layers, leaving a gap in understanding how these layers interact and contribute to the overall system's performance. Specifically, translating high-level energy management decisions into actionable voltage references for primary controllers, especially in dynamic and potentially reconfigurable DCmG topologies, remains a significant hurdle. The paper aims to bridge this gap by providing a unified, top-to-bottom hierarchical control framework.
Methods
The paper proposes a three-layer hierarchical control architecture:
- Tertiary Layer (EMS): This layer, at the highest level, uses a Model Predictive Control (MPC) approach to optimize energy management. It generates optimal power references and operational modes (ON/OFF, operating modes) for DGUs, considering forecasts, DGU capabilities, and economic objectives. It simplifies network and DGU filter losses for computational efficiency.
- Secondary Layer: This is a novel intermediary layer that translates the EMS's power references into appropriate voltage references for the primary layer. It achieves this by solving an optimization problem that embeds DCmG power-flow equations, including network and DGU filter losses, and DGU capability limits. This layer operates at a faster time scale than the EMS to ensure quick response to changes. A key contribution is proving that the power-flow equations are always solvable and deriving a novel, local condition for the uniqueness of load voltages.
- Primary Layer: This layer consists of decentralized voltage controllers at each DGU. These controllers are assumed to be stable and capable of tracking voltage references provided by the secondary layer, ensuring overall voltage stability and enabling plug-and-play operation of DGUs.
The framework is designed to handle generic DCmG topologies and accommodate topological changes resulting from EMS commands.
Datasets
Not discussed in the paper. The paper uses a modified 16-bus DC system for simulation and validation.
Key Insights and Ideas
- Unified Hierarchical Control: The paper demonstrates the effectiveness of a tightly integrated three-layer hierarchical control structure for DC microgrids, addressing the often-missing comprehensive analysis of inter-layer coordination.
- Power-to-Voltage Translation: The secondary layer's role in translating EMS power references into voltage references is crucial for voltage-controlled primary layers, and the paper provides a robust method for this translation using power-flow equations.
- Guaranteed Solvability of Power Flow: A significant theoretical contribution is proving that the power-flow equations used in the secondary layer's optimization problem are always solvable, which is essential for the feasibility of the control strategy.
- Novel Uniqueness Condition for Load Voltages: The paper derives a new, local condition for the uniqueness of load voltages, which is critical for ensuring that DGUs can effectively meet the EMS's power references, especially in the presence of nonlinear loads.
- Topology Flexibility: The control framework is designed to accommodate dynamic changes in the DCmG topology, which can occur due to EMS commands to switch DGUs on or off.
Key Results
The paper validates its proposed hierarchical control scheme through simulations on a modified 16-bus DC system. The results demonstrate:
- Stable Operation: The system successfully maintains voltage stability and tracks power references from the EMS, even with dynamic load changes and DGU operational mode switches.
- Optimal Energy Management: The MPC-based EMS effectively optimizes DGU dispatch, battery charging/discharging, and PV generation utilization to minimize costs and ensure load satisfaction.
- Robustness to Topology Changes: The control architecture handles topological changes caused by DGU switching without compromising system stability.
- Uniqueness of Voltages: The derived condition for load voltage uniqueness is shown to hold in simulations, supporting the theoretical findings.
Limitations and Future Work
- Computational Complexity: While the paper states that high computational performance is not required, the EMS optimization problem, especially with mixed-integer programming, can become computationally intensive for larger or more complex DCmGs.
- Centralized EMS: The current EMS is centralized, which might be a single point of failure.
- Future Work: The authors suggest several avenues for future research:
- Developing an EMS that enables the DCmG to operate in grid-connected mode.
- Designing a secondary layer that does not require complete knowledge of the DCmG topology.
- Decentralizing the secondary and tertiary control layers to create a fully scalable DCmG control architecture.
Summary updated