Prediction of window opening behavior and its impact on HVAC energy consumption at a residential dormitory using Deep Neural Network

•A novel method of predicting the window state in residential dorms using Deep Neural Network (DNN)•A new method of embedding the window state prediction DNN model into the EnergyPlus using Python API.•A comparison of air exchange rates from the blower door test and the mass-balance equation.•Assessing energy loss through windows using DNN-predicted window state values.