The present paper describes an approach to develop a neural network model approach to characterize the thermal and electrical relationship in a climatized building. The approach is fundamental for the inclusion of such building in an grid with demand-response strategy, such as a micro-grid or a renewable energy community. The approach is based on the simulation of different buildings with variable boundary conditions in the EnergyPlus environment. This simulation is used to create a database of electrical and thermal profiles, which is then used to create direct (electrical to thermal) and inverse (thermal to electrical) models of the building. Both models were validated against test data to assess the accuracy of their predictions.
Belloni, E., riganti fulginei, F., Lozito, G.M., Poli, D. (2023). Direct and Inverse Neural Modelling of Buildings HVAC Systems. In EUROCON 2023 - 20th International Conference on Smart Technologies, Proceedings (pp.269-274). Institute of Electrical and Electronics Engineers Inc. [10.1109/EUROCON56442.2023.10198977].
Direct and Inverse Neural Modelling of Buildings HVAC Systems
riganti fulginei f.;
2023-01-01
Abstract
The present paper describes an approach to develop a neural network model approach to characterize the thermal and electrical relationship in a climatized building. The approach is fundamental for the inclusion of such building in an grid with demand-response strategy, such as a micro-grid or a renewable energy community. The approach is based on the simulation of different buildings with variable boundary conditions in the EnergyPlus environment. This simulation is used to create a database of electrical and thermal profiles, which is then used to create direct (electrical to thermal) and inverse (thermal to electrical) models of the building. Both models were validated against test data to assess the accuracy of their predictions.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
