Development of a centralized building management system

• Autor(a) principal: Ferreira, João Miguel Soares
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/31332
• topic: Internet of things (IoT); Industrial automation; Building automation; Building management system

A building management system has user confort and comodity, as well as reduction of energy consumption, as its main goals. To accomplish this, it is necessary to integrate sensors and actuators as to control and retrieve information about the physical processes of a building. These processes include control over illumination and temperature of a room, and even access control. The information, after processed, allows a more intelligent and efficient way of controlling electronic and mechanical systems of a building, such as HVAC and illumination, while also trying to reduce energy expenditure. The emergence of IoT allowed to increment the number of low level devices on these systems, thanks to their cost reduction, increased performance and improved connectivity. To better make use of the new paradigm, it is required a modern system with multi-protocol capabilities, as well as tools for data processing and presentation. Therefore, the most relevant industrial and building automation technologies were studied, as to define a modern, IoT compatible, architecture and choose its constituting software platforms. InfluxDB, EdgeX Foundry and Node-Red were the selected technologies for the database, gateway and dashboard, respectively, as they closely align with the requirements set. This way, a demonstrator was developed in order to assess a systems’s operation, using these technologies, as well as to evaluate EdgeX’s performance for jitter and latency. From the obtained results, it was verified that, although versatile and complete, this platform underperforms for real-time applications and high reading rate workloads.