Moving Microgrid Hierarchical Control to an SDN-Based Kubernetes Cluster: A Framework for Reliable and Flexible...

Area of Science:

• Electrical Engineering
• Computer Science
• Network Engineering

Background:

• Traditional microgrid hierarchical control faces challenges with centralized, monolithic Software Defined Networking (SDN) controllers, leading to latency, resiliency, and scalability issues.
• These drawbacks limit the effectiveness and robustness of microgrid communication networks.

Purpose of the Study:

• To propose and validate a novel, intelligent control network for microgrids that overcomes the limitations of monolithic SDN controllers.
• To enhance microgrid communication performance, scalability, and resilience through a distributed microservices architecture.

Main Methods:

• The study proposes segregating SDN controller functionalities into microservices groups.
• These microservices are distributed across a bare-metal Kubernetes cluster for enhanced control.
• Performance was validated using PLECS hardware-in-the-loop simulation.

Main Results:

• The microservices-based SDN controller demonstrated a 10.76% reduction in latency compared to monolithic architectures.
• Packet loss was decreased by 42.23%, and recovery time during failures was reduced by 53.41%.
• The approach eliminated the single point of failure inherent in monolithic designs.

Conclusions:

• Combining Kubernetes with SDN microservices offers a robust framework for microgrid control, improving performance and reliability.
• This distributed architecture enhances security, portability, and application recovery time.
• The proposed solution serves as a reference for future edge computing and intelligent control in networked microgrids.