Control of HVAC systems may reduce congestion of the electricity grid on district level by shifting energy demand of buildings and increase the self-consumption of local photovoltaic energy. To achieve an optimal control of ventilation, occupant behaviour should be taken into account. To describe occupant behaviour, usually black box models are used and typically need large amounts of high quality training data. Alternatively, use of physical relations allows for a good predictive power requiring less training data. TNO has developed a novel hybrid modelling approach: SirinE, combining data driven occupancy models with physics models including HVAC component models, a ventilation model and a heat transfer model. The SirinE model is calibrated with standardized monitor data. The SirinE simulation is an integrated modelling environment, enabling calibration of the parameters of both the ventilation and the heat transfer model to a real building. This paper focusses on the AirMaps ventilation model, and the results are compared with monitoring data of a field test of a dwelling.