Daylight control modelling with a multi-screen shading device

Abstract

The museum Boijmans van Beuningen in Rotterdam is currently closed for restoration. The architect Van der Steur made an extensive study of lighting in museum rooms (1930). This resulted in a pitched roof with glass and a layer of horizontal glass panes (legramen) that form the ceiling of the exhibition rooms. Below the glass panes timber blades (schoepen) are placed under an angle to disperse the light.
The museum board wants to preserve the vision on daylight of Van der Steur and is now considering options for sun shading above the ceiling. This addition will allow them to better control the illumination in the museum and protect the artwork.

This Master thesis describes a method to assess the daylight exposure in the museum Boijmans van Beuningen. It can be used for a combined lighting design (daylight and artificial). The purpose of the method is to compare different sun shading solutions and their effects on the illuminance. A toolbox is made that can be used on other museums.

The method starts with a 3D model of one of the museum rooms. Alterations had to be made to accommodate the temporary renovation state. For the validation of the model, HDR images were made on-site. These results were compared with the calculation made with the 3D model. Iteratively the best possible fit was made.

With the validated model an hourly daylight simulation was done using the EPW climate-based weather data. The resulting illuminance exceeded the desired level at several points in time. Three different types of sunscreens were chosen. These were schematised as a continuous layer on top of the legramen. Based on the behaviour of the sun shading on three specific days, a daylight factor of 2% was found between the illuminance on the wall and the outside illuminance. With this relation and the EPW weather data the desired sun shading states for an entire year are predicted. A control mechanism is designed for the opening and closing of the sun shading based on local measurements. When the sun shading state changes, the illuminance can fall below the desired level of 125 lx. In that case, additional artificial lighting is needed. The calculations show that over an entire year, the total exposure is 469.286 lx·hr, this is a reduction of 82.2 % compared to the museum without any sun shading.

With the developed approach the user can gather building-specific characteristics and geometries, that include some form of ‘device’ that controls the top light entering a room. If the first part of the modelling process is done, and the hourly data is collected, the model behaves within the practically accepted limits that were set. This version can assess what selection of screens is effective, that will reduce the incoming light, without unnecessarily over-reducing it, and so limits the need for additional diffuse artificial lighting.