CONSTRUCTION OF TEMPORAL DATA-BASED FILES FOR BUILDING ENERGY SIMULATIONS: THE MISSING DATA ISSUE

To quantify the energy performance of a given building, the most uncertain and probably the most influent parameters are the dynamical parameters known as the occupancy rate, the weather data and the temperature set point. Users sometimes encounter gaps in these data, and techniques are needed to estimate variables when data are missing. To obtain most of the dynamic data, measurements are made at least during one year. These measurements are expensive and time consuming. In this paper, we proposed a methodology by means of statistical approaches to build a full year dynamic data files in order to fill gaps according to the measured data features. The methodology is based on statistical approaches, (Iman and Conover 1982) and on the methodology used by J. Goffart (2013) to generate weather data. Thus performance indicators are used to evaluate the methodology and its consequences on energy consumption estimation. An office building has been monitored and multiple sensors have been mounted on candidate locations to get needed data. In this paper a methodology to adress the problem of missing data in measurement data such as temperature set point data is proposed.