A Scalable Microservices Architecture for Condition Monitoring and State-of-Health Tracking in Power Conversion...

Area of Science:

• Electrical Engineering
• Computer Science
• Materials Science

Background:

• Power converters are critical components in modern electrical infrastructure, including electric vehicle charging, battery energy storage, and photovoltaic systems.
• High reliability demands continuous condition monitoring for predictive maintenance, a task challenging for traditional SCADA and HMI systems due to scalability and data handling limitations.

Purpose of the Study:

• To propose a scalable, containerized microservices-based architecture for degradation tracking and State-of-Health (SoH) monitoring in power conversion systems.
• To address the scalability and data aggregation bottlenecks of traditional systems for long-term predictive maintenance.

Main Methods:

• A decoupled four-layer architecture was designed, employing UDP servers for data ingestion, RabbitMQ for message routing, and MongoDB for data storage with a FastAPI interface.
• Validation was performed using a Hardware-in-the-Loop (HiL) setup with a Typhoon HIL606 simulator monitoring an Active Neutral Point Clamped (ANPC) power converter.

Main Results:

• The system achieved a Packet Delivery Ratio (PDR) of 1.0 at ingestion rates up to 100 messages per second per node.
• Transmission and processing overheads remained consistently below 5 ms, significantly exceeding the nominal requirement of 2 msgs/s.
• Demonstrated robust data integrity and timely data availability for tracking thermal dynamics and aging trends.

Conclusions:

• The proposed modular architecture offers horizontal scalability essential for Industry 4.0 integration.
• This high-performance framework enables robust, long-term health monitoring for modern power electronics, enhancing reliability and predictive maintenance capabilities.