Method of Predicting the Energy Consumption of a Building

Method for forecasting the energy consumption of a building, implemented by a computer and using a physical model taking into account the heating and cooling (Pc) of the building by a heating and/or air-conditioning device, the passive solar inputs (Pasp), the inputs by internal gains (Pi) such as the heating by people and household appliances, and the heat losses (heat_losses) of heat convection and/or conduction type between the building and the outside environment, wherein it comprises a learning step to deduce therefrom the value of the parameters of the physical model based on measurements performed on the building in the past, comprising:

• the determination of at least one parameter of the physical model relating to the internal gains and/or to the heat losses based on past measurements during periods of weak sunlight;
• the determination of at least one parameter of the physical model relating to the passive solar inputs based on past measurements during sunny periods.