Data Mining of Smart WiFi Thermostats to Develop Multiple Zonal Dynamic Energy and Comfort Models of a Residential Building

摘要

Smart WiFi thermostats have gained an increasing foothold in the residential building market. The data emerging from these thermostats is transmitted to the cloud. Companies are attempting to use this data to add value to their customers. This overarching theme establishes the foundation for this research, which seeks to utilize smart WiFi thermostat data from individual residences to develop a dynamic model to predict real time cooling demand and then apply this model to 'what-if' thermostat scheduling scenarios. The ultimate goals of these efforts are to reduce energy use in the residence and/or demonstrate the ability to respond to utility peak demand events. A regression tree approach (Random Forest) was used to develop models to predict the room temperature as measured by each thermostat and the cooling status. The models developed, when applied to validation data (e.g., data not employed in training the model) yielded R-squared values of greater than 0.98. The results from the 'what if' scenarios show a huge opportunity for quantifying cooling energy consumption reduction through the use of more aggressive non-occupied temperature setpoint schedules, as well as the total time that cooling/heating could be interrupted in responding to a high demand event while maintaining thermal comfort within acceptable ranges.