Development of Long-Range Wireless Sensor Network for Monitoring Historical Buildings
摘要
The phenomenon of uneven performance of mechanical heating circles in historical buildings, resulting in significant differences of energy demand and occupant comfort is present within several buildings of BME (Budapest University of Technology and Economics), To help identify the issues and understand the operation of a building with several heating circles and various profiles in each, while educational, laboratory and administrative tasks run uninterrupted, a need for monitoring systems arise. In these buildings, it is rarely possible to implement wire powered monitoring systems, while wi-fi coverage may also not be sufficient. The aim was to implement a wireless monitoring system whose endpoints are accumulator powered and use a communication protocol which does not require internet connection, while being able to send and receive data through dense constructions within a district area of examination. A wireless monitoring system was developed to use in BME Building F, using Long Range Wide Area Networking (LoRaWAN) technology applied on Raspberry Pi Pico 2 microcontrollers with various appliances to examine connection between hygrothermal conditions, CO2 concentration and occupancy patterns. Data collection was managed by a LoRaWAN Gateway operated by a Raspberry Pi 4 Model B logging data locally and online, providing remote log availability. To provide portability and reliable conveyance of the endpoint devices, sensor boxes were modelled and 3D printed. The data collected so far has already helped identify trends in indoor comfort and usage. This information can also support energy-saving strategies and building energy simulations. In the future, the system may help detect summer overheating using Overheating Degree Hours above 26 ℃ (ODH26) method. The developed system is applicable to any site, where wire powered monitoring is not possible, while a need for long time logging is present. Finite element dynamic energy and comfort simulations might be also possible based on the data collected.