A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Improving residential building energy simulations through occupancy data derived from commercial off-the-shelf Wi-Fi sensing technology
https://orcid.org/0000-0001-7224-7940
Abstract One of the main required actions for moving forward in zero-carbon cities’ mission is to increase building energy efficiency. Building energy simulations are key tools for HVAC control, building energy management, testing different building energy efficiency scenarios, and providing feedback to the occupants. Unfortunately, these tools have a significant weakness in occupant behavior modeling and prediction. Several recent studies used different occupancy measurements and methods to predict occupant behavior. However, they suffer from threatening the occupant's privacy and the high cost of their infrastructure and deployment. This paper aims to leverage passive WiFi sensing methods for occupant behavior estimation using commercial off-the-shelf WiFi devices, for which the infrastructure is already available in many buildings. Passive and device-free WiFi sensing does not threaten the occupant's privacy, and its deployment cost is low. Therefore, this method of occupancy measurement has the potential to be used on an urban scale and to play an essential role in zero carbon and smart cities in the future. Different machine learning algorithms have been applied to the collected, denoised, and preprocessed Channel State Information (CSI) data to estimate the occupant behavior. Five schedules of occupancy, occupant activity, lighting, electrical equipment usage, and thermostat set point temperature could be extracted and estimated from CSI data. In the last step, all estimated schedules were fed into EnergyPlus, an energy simulation software, along with other required data for accurate energy demand calculation to estimate heating and cooling demand, which can be used for HVAC control and occupancy feedback.
Improving residential building energy simulations through occupancy data derived from commercial off-the-shelf Wi-Fi sensing technology
https://orcid.org/0000-0001-7224-7940
Abstract One of the main required actions for moving forward in zero-carbon cities’ mission is to increase building energy efficiency. Building energy simulations are key tools for HVAC control, building energy management, testing different building energy efficiency scenarios, and providing feedback to the occupants. Unfortunately, these tools have a significant weakness in occupant behavior modeling and prediction. Several recent studies used different occupancy measurements and methods to predict occupant behavior. However, they suffer from threatening the occupant's privacy and the high cost of their infrastructure and deployment. This paper aims to leverage passive WiFi sensing methods for occupant behavior estimation using commercial off-the-shelf WiFi devices, for which the infrastructure is already available in many buildings. Passive and device-free WiFi sensing does not threaten the occupant's privacy, and its deployment cost is low. Therefore, this method of occupancy measurement has the potential to be used on an urban scale and to play an essential role in zero carbon and smart cities in the future. Different machine learning algorithms have been applied to the collected, denoised, and preprocessed Channel State Information (CSI) data to estimate the occupant behavior. Five schedules of occupancy, occupant activity, lighting, electrical equipment usage, and thermostat set point temperature could be extracted and estimated from CSI data. In the last step, all estimated schedules were fed into EnergyPlus, an energy simulation software, along with other required data for accurate energy demand calculation to estimate heating and cooling demand, which can be used for HVAC control and occupancy feedback.
Improving residential building energy simulations through occupancy data derived from commercial off-the-shelf Wi-Fi sensing technology
https://orcid.org/0000-0001-7224-7940
Abstract One of the main required actions for moving forward in zero-carbon cities’ mission is to increase building energy efficiency. Building energy simulations are key tools for HVAC control, building energy management, testing different building energy efficiency scenarios, and providing feedback to the occupants. Unfortunately, these tools have a significant weakness in occupant behavior modeling and prediction. Several recent studies used different occupancy measurements and methods to predict occupant behavior. However, they suffer from threatening the occupant's privacy and the high cost of their infrastructure and deployment. This paper aims to leverage passive WiFi sensing methods for occupant behavior estimation using commercial off-the-shelf WiFi devices, for which the infrastructure is already available in many buildings. Passive and device-free WiFi sensing does not threaten the occupant's privacy, and its deployment cost is low. Therefore, this method of occupancy measurement has the potential to be used on an urban scale and to play an essential role in zero carbon and smart cities in the future. Different machine learning algorithms have been applied to the collected, denoised, and preprocessed Channel State Information (CSI) data to estimate the occupant behavior. Five schedules of occupancy, occupant activity, lighting, electrical equipment usage, and thermostat set point temperature could be extracted and estimated from CSI data. In the last step, all estimated schedules were fed into EnergyPlus, an energy simulation software, along with other required data for accurate energy demand calculation to estimate heating and cooling demand, which can be used for HVAC control and occupancy feedback.
Improving residential building energy simulations through occupancy data derived from commercial off-the-shelf Wi-Fi sensing technology
Samareh Abolhassani, Soroush (author) / Zandifar, Azar (author) / Ghourchian, Negar (author) / Amayri, Manar (author) / Bouguila, Nizar (author) / Eicker, Ursula (author)
Energy and Buildings ; 272
2022-07-30
Article (Journal)
Electronic Resource
English
Occupancy Data Sensing, Collection, and Modeling for Residential Buildings
| Springer Verlag | 2022
Estimating building energy ratings for the residential building stock: Location and occupancy
| BASE | 2014
Post-occupancy life cycle energy assessment of a residential building in Australia
| Taylor & Francis Verlag | 2014