engleski

Prediction of User-defined Room Temperature Setpoints by LSTM Neural Networks

Buildings are one of the largest consumers of energy in the world and as such an opportunity for significant energy savings. Modelling of user behavior is a prerequisite for predicting the building energy consumption that offers a possibility to adapt the building operation to best suit the user requirements while achieving minimal energy consumption through room air conditioning. In this paper, a deep learning model for prediction of an individual room temperature setpoint provided by the user is presented. Model learning was conducted on the available dataset containing external weather and internal climate measurements. External weather conditions consist of air temperature, relative humidity and direct, diffuse and global solar irradiation. The internal conditions refer to four offices for which the air temperature, the mode of operation of the fan coil controller and the temperature setpoint are available. The model was built using a neural network with Long Short- Term Memory. The highest accuracy was achieved by the model in which the variables on the old setpoint, system working, day of the week and relative humidity were used. Presented results show high model accuracy, applicable for optimization of real time building operation.

user behavior modelling ; room temperature reference prediction ; deep learning ; LSTM ; neural networks

nije evidentirano