Publication
Real-time indoor air quality (IAQ) monitoring system for smart buildings
| datacite.subject.fos | Engenharia e Tecnologia::Outras Engenharias e Tecnologias | pt_PT |
| dc.contributor.advisor | Lima, José | |
| dc.contributor.advisor | Brito, Thadeu | |
| dc.contributor.advisor | Nakano, Alberto Yoshihiro | |
| dc.contributor.author | Biondo, Elias Junior | |
| dc.date.accessioned | 2023-05-19T13:51:28Z | |
| dc.date.available | 2023-05-19T13:51:28Z | |
| dc.date.issued | 2023 | |
| dc.description | Mestrado de dupla diplomação com a UTFPR - Universidade Tecnológica Federal do Paraná | pt_PT |
| dc.description.abstract | Indoor air quality (IAQ) is a term describing the air quality of a room, it refers to the health and comfort of the occupants. Normally, people spend around 90% of their time in indoor environments where the concentration of air pollutants, such CO, CO2, VOCs, SO2, O3 and NOx, may be two to five times — and occasionally, more than 100 times — higher than outdoor levels. According to the World Health Organization (WHO), the indoor air pollution is responsible for the deaths of 3.8 million people annually. It has been indicated that IAQ in residential areas or buildings is significantly affected by three primary factors: (i) Outdoor air quality, (ii) human activity in buildings, and (iii) building and construction materials, equipment, and furniture. In this contest, this work consist in a real time IAQ system to monitoring and control thermal comfort and gas concentration. The system has a data acquisition stage, where the data is measured by a set of sensors and then stored on InfluxDB database and displayed in Grafana. To track the behavior of the measured parameters, two machine learning algorithms are developed, a mathematical model linear regression, and an artificial intelligence model neural network. In a test made to see how precise were the prediction of the two models, linear regression model performed better then neural network, presenting cases of up to 99.7% and 98.1% of score prediction, respectively. After that, a test with smoke was done to validate the models where the results shows that both learning models can detect adverse cases. Finally, prediction data are storage on InfluxDB and displayed on Grafana to monitoring in real-time measured data and prediction data. | pt_PT |
| dc.identifier.tid | 203299701 | pt_PT |
| dc.identifier.uri | http://hdl.handle.net/10198/28332 | |
| dc.language.iso | eng | pt_PT |
| dc.rights.uri | http://creativecommons.org/licenses/by-nc/4.0/ | pt_PT |
| dc.subject | Indoor air quality | pt_PT |
| dc.subject | Monitoring system | pt_PT |
| dc.subject | Machine learning | pt_PT |
| dc.subject | Artificial intelligence | pt_PT |
| dc.title | Real-time indoor air quality (IAQ) monitoring system for smart buildings | pt_PT |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| rcaap.rights | openAccess | pt_PT |
| rcaap.type | masterThesis | pt_PT |
| thesis.degree.name | Engenharia Industrial | pt_PT |