| Summary: | With the growing importance of sustainability, green technology is becoming popular to make buildings energy efficient. Various researches are ongoing to reduce both energy consumption and energy wastage significantly. As a result, smart technologies in buildings are in demand to resolve this issue. This can be done by accurately capturing the occupancy in the room based on the environmental parameters. In this paper, time constant of two environmental parameters, carbon dioxide and temperature were determined to provide a more accurate estimation of occupancy. Machine learning techniques were used to develop a model to analyze the relationship between these parameters and occupancy of buildings. A suitable sampling time was then determined to provide a more accurate estimation. The model started by training their respective class of Artificial Neural Network (ANN), simulation was made and the best performance data were chosen. MATLAB was used to determine the time constant and training of data for the optimal sampling time of the HVAC system. Using the derived time constant and appropriate sampling time, there will be lesser discrepancies in the occupancy estimation, this allows future improvement methods to be deployed based on these findings. This provides a universally acceptable method of occupancy estimation with the use of time constant. Furthermore, the building owner will be able to provide the optimal thermal comfort based on the occupancy level, which in turn saves energy consumption without compromising thermal comfort.
|
|---|