Predictive optimization of heat demand utilizing heat storage capacity of buildings

Stricter energy efficiency requirements and the increasing production of renewable energy pose new challenges for energy systems. Storage capacity and flexibility are required so that the availability of the energy can always be secured.
One source of flexibility in energy systems is building’s mass whose heat storage capacity enables them to be utilized as short-term energy storages. The efficient utilization of this flexibility potential requires predicting the thermal response of the buildings at feasible time horizon. However, developing prediction models for individual buildings is typically time-consuming which hinders the utilization of the heat storage capacity of buildings at large scale.
In this study, new mathematical modelling methods were developed to predict the indoor temperature and heat demand of buildings. The developed models were then applied to the predictive optimization of the heat demand utilizing the heat storage capacity of buildings. The research showed that the developed modelling methods are suitable for application in predictive optimization of the heat demand. The generalizability and ease of implementation enable the prediction models to be utilized in individual buildings and at city-level predictive optimization.
Furthermore, it was shown that by utilizing the prediction models and the flexibility of the buildings, it is possible to significantly schedule the heat demand and decrease the peak demand. The results of this dissertation can be applied to develop city-level demand side management schemes that aim to increase the energy efficiency and flexibility of energy systems.