Thesis
Thesis
Noureddine Zemmouri
February 2005
It is conventional in most countries to express the level of daylight illuminance within buildings
in relative rather than absolute terms as a fraction or percentage of the illuminance occurring
simultaneously on an external horizontal surface exposed to the whole sky. This fraction known
as the daylight factor, is a geometrical property of the room and window and, for a given sky
luminance distribution, it is theoretically a constant independent of changes in the sky
illuminance. Basically it acts as transfer factor whereby the internal illuminance at specific points
in a room can be determined by multiplication with an external horizontal illuminance reference
value. However, this method generally considers uniform and overcast sky conditions. Various
procedures for expanding the daylight factor method to include real skies have been experienced.
Unfortunately, these procedures were either incomplete, too complex, or involved long periods of
expensive design time. Therefore, it became necessary especially in the case of hot arid zones
like Algeria to develop a simplified daylighting design procedure which will expand the daylight
evaluation procedure to include all sky conditions.
Random processes such as daylight, has always been difficult to evaluate. Most experienced
designers will quickly suggest that experience is the best tool for evaluating such events. With the
rapidly evolving technologies in the field of daylighting, it is desirable to merge the experience of
many designers with algorithms that may aid in difficult situations. Cognitive computing has been
an emergent set of problem solving algorithms that attempt to imitate natural problem solving
techniques. One such method is called Fuzzy Logic.
The objective of these studies is to obtain a definitive description of the daylight climate of
Algeria by determining typical sky conditions from existing NASA satellite measured records of
cloud cover information. A Fuzzy sets method has been applied for the development of a
program called MatLight used for the estimation of the external and internal luminous
environment under actual sky conditions.
I
The thesis has therefore been divided into two main parts. Part one including chapters 1 2 and 3
concerns the state of the art of daylighting practice and the data on which it is based. Part two
contains chapters 4, 5, 6 and 7 and concerns the integrated modelling and the fuzzy approach.
II