In this article, the main research goal is to present a model for rapid assessment of specific heating energy consumption of residential buildings. In the first part of this research paper the main inputs that could influence the building heating demand are identified.The next step in the development of the prediction model was to conduct dynamic simulations with different combinations of the inputs and to obtain a database of results.  With this database, a multiple regression was applied and a simple (3 inputs) and accurate (R2=0.987) model was obtained.  

This paper presents the first step modelling process of a smart building  in order to ensure better energy management and human comfort in buildings. It consists on combining physical models and experimental measurements in order to have more adapted models for virtual simulation and optimal control. Advantages and difficulties related to this process will be detailed through a defined use case of a smart building: PREDIS.   The paper starts, by introducing the studied building and its different components: HVAC system and thermal envelope.

Increasing concern about building sector energy consumption and the simultaneous need for an acceptable thermal environment makes it necessary to estimate in advance what effect different thermal factors will have on occupants. So far most human thermal comfort models are based on estimates assuming steady-state conditions. However, this often leads to underestimations of local cold or hot surfaces. These kinds of models does not take into account variable conditions.

To study the impact variations in input on output vari-ation (Sobol index) of building model, a stochastic model (time series) of variables is introduced. To build the time series we used data measured in a building for each hour for a duration of one month.
The external temperature does not appear directly in the model but affects all the input. So after model it as Auto-Regressive process we propose to model input, for example the heating flux, by an ARMAX process.

For the purpose of revitalizing traditional architecture, we considered the characteristics of a traditional architecture cluster with regard to cross ventilation, and from an environmental engineering point of view, we proposed techniques to improve the surrounding environment. In this report we make use of and improve the urban area model used in a prior report, as well as use CFD to take into account the effects of surrounding topography. To improve the urban area model, we make detailed computations of the wind pressure coefficient distribution.