Using a fuzzy black-box model to estimate the indoor illuminance in buildings

• A fuzzy model able to estimate indoor illuminance using measured or simulated external conditions.
• The model shows good illuminance-estimation levels in comparison to indoor illuminance measurements.
• To obtain the model, knowledge about the room geometry, surfaces, windows and furniture is not required.
• The model can be used for simulation studies on living comfort, energy conservation and control design.
• The model is simple to implement in any programming language by using a set of fuzzy rules.

Indoor lighting conditions and the efficient use of solar energy are becoming increasingly important, both in terms of overall living comfort and energy-efficient buildings. Thus, the methods of achieving adequate indoor lighting conditions, by making use of daylight to save energy, while still maintaining other indoor conditions at an acceptable level, and the corresponding control algorithms, have lately become the subject of considerable attention. For this reason, a fuzzy black-box model of indoor illuminance levels is proposed. The model is able to estimate the indoor illuminance levels as its outputs, by using real external conditions as its inputs. In general, the indoor illuminance is a combination of daylighting and electric lighting. Our study has shown that by using a fuzzy illuminance model, with the solar radiation, the external illuminance, the position of the blinds and the status of the lights as its inputs, an indoor illuminance estimation comparable to the measured data can be obtained. Furthermore, the sufficiently small error measures indicate that the presented modelling approach can be incorporated into larger test environments and used for studies on indoor living comfort, energy conservation, control design or even model-based control. Whenever needed, several fuzzy-illuminance models can be used in parallel in order to obtain the illuminance levels for more than one indoor position.