English

Recent computational improvements allow for wind and thermal simulations on more complex urban configurations. Their thermo-aeraulic features can now be investigated by more sophisticated CFD models, coupled with energy ones. By assessing more accurately micro-climatic conditions, their suitability for both human comfort and building energy consumption prediction is increased.  Such coupled studies already exist but are still scarce. They highlight the impact of urban morphology and its complexity on induced flow phenomena and radiative exchanges.

Based on observations conducted in an office building, we apply advanced statistical analysis methods, leading to the formulation of stochastic models for the prediction of buildings occupants’ actions on window openings and shading devices.  The statistical analysis method – based on generalised linear mixed models – enables a correct treatment of the longitudinal nature of the datasets, an accurate estimation of the calibration parameters’ uncertainty and a detailed study of the differences between the occupants surveyed.

In order to inform the design of a building or a group of buildings in relation to their potential energy efficiency, the main impact will be at the initial concept design stage. Variations and interactions of parameters need to be considered quickly as the design develops. In addition to the variation and interrelation of parameters associated with individual buildings, the design should consider the influence, both from and on, neighbouring buildings and landscape features.

The availability of input data and appropriate computing times are two major challenges when simulating entire city districts. For dynamic heat demand simulations, we contribute to these tasks by developing an integral tool chain. To reduce the amount of input data, we automatically create typical datasets of building setups for office buildings according to their year of construction and basic geometric data. We use these datasets to automatically parameterize a dynamic building model.