Reducing the Parameter Count through a Sensitivity Analysis Performed on a Mathematical Model Used for Estimating Energy Consumption in a Passive House

Energy efficiency in the buildings sector is one of the main research areas on which the European Union has focused its efforts. Reducing energy consumption has become a priority, as well as its estimation and prediction. There are a number of different models used to achieve the latter, from white-box models (fully informed) to black-box models (all necessary information is inferred from data), as well as hybrid approaches. However, these existing gray-box models rely on mathematical models which can easily be over-parameterized. We have implemented a gray-box model and propose an improved Sensitivity Analysis method which we have used to identify the correlated parameters. Moreover, upon analyzing the results of this analysis, we propose a simplification applied on the mathematical model, without actually modifying it.