Thibault Charles

Lighting modelling is used by architects, designers and engineers to test a visual effect, experiment the access to daylight, estimate lighting metrics, etc. The simplest model can result in a hand drawing showing the sun beams transmitted through a window or reflected on an opaque surface. This kind of model is often used and really powerful at early design stages. It allows giving directions to the designs but is usually not accurate enough to provide useful metrics and visual validations… Voir plus

Lighting modelling is used by architects, designers and engineers to test a visual effect, experiment the access to daylight, estimate lighting metrics, etc. The simplest model can result in a hand drawing showing the sun beams transmitted through a window or reflected on an opaque surface. This kind of model is often used and really powerful at early design stages. It allows giving directions to the designs but is usually not accurate enough to provide useful metrics and visual validations. The next step in lighting modelling is often a computer based simulation where the architect or engineer set up a 3D model of the building to study. This model can provide a lot more information about the real behaviour of light but is much more complex to set up and require a lot of inputs which can be difficult to assess at early design stages.
In this report, we focus on the different possibilities to model a window in Radiance. We start by giving a brief introduction of the concepts governing the lighting physic and also explain the basis of the Radiance numerical approach. The main materials “primitives” available in Radiance to model a transparent or translucent material are then introduced and two base cases are presented to allow the reader understanding these different materials. A brief introduction of an alternative lighting modelling approach is finally given. Voir moins

This study focuses on sensitivity and uncertainty analysis in the context of energy conservation calculation for office buildings. The energy analysis tool resorted to is an advanced version of the detailed dynamic thermal model from the French Building Code. Within the probabilistic frame adopted in the study energy conservation calculation no more provides a single point estimate but rather a prediction interval around the expected value for a given confidence level.
- Firstly the… Voir plus

This study focuses on sensitivity and uncertainty analysis in the context of energy conservation calculation for office buildings. The energy analysis tool resorted to is an advanced version of the detailed dynamic thermal model from the French Building Code. Within the probabilistic frame adopted in the study energy conservation calculation no more provides a single point estimate but rather a prediction interval around the expected value for a given confidence level.
- Firstly the independent variables domain is mapped through global sensitivity analysis giving a clear and thorough view on the main contributions to the energy consumption variance. For that purpose Morris method is used.
- Then the sources of uncertainty are modeled considering the specific process of energy auditing suitable for commercial buildings.
- Eventually uncertainties are combined through Monte Carlo simulation and a non-parametric method (Wilks) is used for estimating the quantiles of interest.
The whole process was applied to three case studies: two real office building and one virtual test case. The paper details the sensitivity and uncertainty analysis methods and the procedure for quantifying the sources of uncertainty as they impact upon building simulation. The case studies, the simulation tool and the method implementation for analyzing the effect on simulation outputs are all described. Voir moins