Modelling.jpg
13.1.jpg
13.2.jpg
Fig 13.189

Fig. 13.290 

13.2 Computer Simulation
Modern computers found in architectural and engineering offices are significantly more powerful than a decade ago. They are therefore capable of running simulation programs that were often used solely by large research organisations with mainframe supercomputers.

Most simple simulation programs can handle daylight factor calculations in simple spaces, which are generally sufficient for sports halls. Complex programs are still in the developmental stage, with varying degrees of accuracy and complexity.

Radiance, for example, is a program that is constantly evolving. It is validated by a number of methods and can produce very realistic, rendered images, glare and daylight factor measurements. However, it is extremely complex to use and requires Unix computers to run it.

A program used by engineers for general lighting, heating and ventilation calculations is Cymap. Its daylight modelling capability is limited to simple daylight factor calculation capabilities. It uses the CIE standard overcast sky (5000 lux) as a basis for calculations.

Many designers, while finding computer simulation useful as a learning technique, and for developing a feel for the sensitivity of daylighting performance to parameters such as room configuration, window size, and surface reflectances, are unwilling to rely on it when making final decisions on the visual effectiveness and aesthetic quality of a particular design proposal. This is a reasonable stance, since most computer tools cannot yet simulate the complex conditions found in real buildings.

Simulating the performance of advanced daylighting components, or detailed assessment of fully integrated daylighting systems (including visual comfort, the luminous quality of interior spaces, thermal comfort and energy consumption) is at present outside the scope of the average professional practice and is more usually carried out by specialised design firms or laboratories.
Back to Top ^
BACK | HOME | NEXT
BACK | HOME | NEXT
13.0 Modelling

To calculate and judge the effectiveness of any daylighting design strategy it is first necessary to perform some form of modelling exercise. This can range from simple hand calculations of average daylight factors to the fully rendered computer images of simulation programs such as Radiance, described below.

The complexity of the modelling will depend on the requirements for accuracy, availability of expertise, cost and time. Often the more complex methods are not the most effective; financially and accuracy wise, given the frequent need for fast results and the high levels of expertise necessary to use the computer programs.

There were several problems found with daylight prediction tools, in a study for ETSU.87 The reasons stated for inadequacies in daylight tools included that:
•  they may be tedious or difficult to use;
•  they may not calculate the quantities designers need;
•  they may contain errors;
•  they may be limited to certain situations, for example rectangular rooms or clear glazing;
•  designers may be unaware of their existence;
•  designers may have insufficient training to use the tool;
•  computer tools may be too expensive or require different or superior hardware as well as users familiar with the technology.

A productive approach for designers is to observe and record the daylighting characteristics of existing buildings. Visual assessment of an interior, combined with the measurement of light levels with a luxmeter will provide a far more convincing and useful basis for daylighting design than a set of figures on a computer. This will enhance the designer’s ability to make a well-informed judgment on the significance of a design tool or the likely performance of a particular design proposal.88 

13.1 Description of Modelling
The method traditionally used for daylighting assessment is the scale model. Light is not dependent on scale, so a model of any scale can reasonably accurately show the effect of light through windows.

BRE’s survey, mentioned above, concluded that architects are more likely to use models for assessing lighting scenarios than engineers, as they are used to working with models and generally less reluctant to resort to computer simulation. Engineers may also use models as verification for their calculations and simulations on computers. Building services consultants and local authorities will use computers for energy and environmental analysis, whilst architects generally did not.30 

Full size mock-ups are generally only needed for complex lighting systems. It is usually sufficient to construct a model no larger than desktop, provided that the surfaces have the same reflectances and colours as in the completed room. A model large enough for someone to put their head inside will allow qualitative assessments to be made of the lighting, as if in the real room. The model should also be viewed under lighting conditions similar to the intended site. This can be done in an artificial sky, see fig. 13.1 (see Appendix H for a list of artificial skies available for use in the UK) or under a real sky.

Models can be expensive to make depending on complexity of the model and number of measurements to be made. However, constructing a model of a sports hall will be a relatively simple process compared to most building types. Internal fittings are also minimal, so there will be no requirement to build scale furniture or fittings.

Models should be fixed to a helidon (adjustable and calibrated multi-axis turntable), see fig. 13.2, so that daylight and sunlight penetration can be assessed for different times of the year.

Sensors mounted inside the model can allow the designer to take readings and assess the daylight factors, illuminance levels and luminances (brightness) of surfaces.

Pairs of models can also be constructed to allow comparative assessment of different shading or glazing combinations.