Institutional Repository
Technical University of Crete
EN  |  EL



My Space

Smart natural lighting systems. Development & optimization of light pipes with integrated low energy consumption artificial lighting, managed by smart controls

Vasilakopoulou Konstantina

Full record

Year 2021
Type of Item Doctoral Dissertation
Bibliographic Citation Konstantina Vasilakopoulou, "Smart natural lighting systems. Development & optimization of light pipes with integrated low energy consumption artificial lighting, managed by smart controls", Doctoral Dissertation, School of Environmental Engineering, Technical University of Crete, Chania, Greece, 2021
Appears in Collections


The present Thesis examines the lighting and energy performance of light pipes. Light pipes are innovative daylight systems that can guide daylight and sunlight from the roof or the facades of buildings, for long distances, through very reflective tubes and provide natural light to spaces that would not otherwise have access to it.Based on the existing literature and the identified gaps, the present Thesis aimed to contribute to the knowledge on the technology of light pipes by experimentally investigating their performance in the Mediterranean - Greek environment and sky conditions; study the already available theoretical tools and/or simulation methods and whether they are able to predict the light pipe performance in the Greek environment and sky conditions; develop methodology/ies for the prediction of the light pipe lighting performance in the specific context; explore how light pipes can be efficiently combined with artificial lighting and controls and what is the magnitude of the energy savings achieved.Initially, two theoretical methodologies, the Luxplot method and the TTE method and simulations were analysed, tested, and compared. The results from the three methodologies may differ significantly. Equations providing the average illuminance from one light pipe in a rectangular space were developed for all the tested methodologies, for easier and faster application.Meanwhile, an experiment was set up to test the performance of a light pipe in the weather/sky conditions of Athens, Greece. A lighting system, consisting of a light pipe, LED lamps and daylight linked controls was designed and constructed in a test cell, located at the campus of the Kapodistrian University of Athens, Greece. The parameters initially monitored were the exterior and interior illuminance. The testing period lasted for approximately 7 months, from November 28th, 2014 until the beginning of August 2015 (no data was recorded in January 2015). The data acquired from the experiment show that the light pipe installed in the test cell offers 100 lux of interior illuminance on average, while the average Daylight Penetration Factor is around 0.15%. The maximum indoor illuminance recorded by a sensor throughout the testing period was 8.4 klux, on 26/04/2015, 12:15 pm. At the same moment, the minimum illuminance was 38 lux, the average illuminance in the space was 1,043 lux, while the exterior illuminance 120 klux. The bright patches of light on the reference plane during clear sky days, decrease the uniformity, which on average is 0.26. Uniformity is higher when the sky is relatively clear and when the reference plane is limited to the area under the light pipe diffuser. The indoor illuminance values were found to present a strong temporal and spatial variability. The analysis showed that the average indoor illuminance varies strongly as a function of the sky clearness and of the exterior illuminance levels. A significant reduction of the average indoor illuminance is observed for increasing Kd values, i.e., for cloudy skies. As measured, under clear sky conditions (0.07 < Kd < 0.14), the average indoor illuminance was about five times higher than during the almost fully cloudy period (0.84 < Kd < 0.93). It was also found that the relation between the average indoor and outdoor illuminance is almost exponential. The illuminance delivered by the light pipe on the reference plane was expressed as formulae, with the independent variable being the exterior illuminance. The sky clearness was also considered, as each of the 10 developed formulae corresponds to one of 10 Kd clusters. The R-squared for the developed equations ranges between 0.90 and 0.99.The experimental interior illuminance data was compared to the results calculated and simulated with the equations developed from the theoretical methodologies and the simulations. The TTE method was found to give results that are considerably different than the experimental data. On the other hand, the equations developed from the Luxplot method and the simulations emulating the light pipe as a luminaire of cosine luminous intensity, as well as the forward ray tracing simulations, provide comparable results with an error between 38 and 43%. The error for each methodology does not change significantly based on the exterior illumination.Since it was not possible to use experimental data for the lighting or energy consumption of the artificial lights installed in the test cell, simulations were used for estimating the used power for artificial lighting systems used in conjunction with light pipes. The artificial lighting systems that were simulated were two: a system like the one installed in the test cell and an “office system”, i.e., LED lamps, commonly used in office spaces, installed in an ortho-canonical grid, while the room modelled had the characteristics of the test cell (size, reflectance values). All the systems were dimmed according to the available daylight levels in the space. Equations for the calculation of the used power of the artificial lighting systems considered were developed, for the whole experimental period and for each month within that period, with independent variables being the exterior illuminance (Eex), the sun azimuth (α) and altitude (γ) and the sky diffuse coefficient (Kd). The four tested systems were found to consume 10 to 38% less energy than if the respective artificial lighting system was the only light source in the space. Arranging the luminaires in an ortho-canonical grid provides greater energy savings and better uniformity levels and should be preferred for environments with increased lighting quality requirements.The results presented in this Thesis apply to the specific environment and lighting systems used and simulated and are intended to showcase methodologies, rather than provide formulas for universal use.

Available Files