학술논문

MQTT Architecture for Stream Analytics of PMU Data
Document Type
Conference
Source
2021 32nd Irish Signals and Systems Conference (ISSC) Irish Signals and Systems Conference (ISSC), 2021 32nd. :1-7 Jun, 2021
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Protocols
Databases
Scalability
Taxonomy
Publish-subscribe
Phasor measurement units
Real-time systems
PMU
MQTT
Stream Analytics
Failover
Language
ISSN
2688-1454
Abstract
Synchrophasor data, derived from Phasor Measurement Units, offers great potential to provide insights into the operation of electrical power systems in the presence of high capacities of renewable generation. Accessing synchrophasor data is challenging since established protocols are rigidly hierarchical and ‘point-to-point’, based on serialized methodologies requiring ‘data concentrators’. This makes real-time and in-stream analysis of synchrophasor data very cumbersome. An alternative is explored in this paper. This novel method is database independent, and allows for synchrophasor data transport, analysis and archival. Using a publish-subscribe MQTT broker, real-time stream analytics are made available in real-time through republishing to new MQTT topics. The advantages of MQTT as a development and deployment platform are presented; in terms of scalability, computational efficiency, latency, modular development, data rate and failover. The dependency chain of an event detector is presented in detail, highlighting the JSON communication protocol used and the implemented topic structure. An intention of this paper is to present the topic name convention and JSON schema employed and invite debate on conventions.