%%class \documentclass{iaesarticle3} %%required package. add for your convenient, but do not remove the initial \usepackage{amsmath, amsfonts, amssymb, float, fancyhdr} \usepackage[scaled]{} \usepackage[figuresright]{rotating} \usepackage{authblk, graphicx, indentfirst, lastpage, lipsum} \setlength{\affilsep}{0cm} \renewcommand\Authfont{\normalsize} \renewcommand\Affilfont{\normalfont\small} \usepackage{subfig, caption, epstopdf} \usepackage[left=3cm, right=3cm, top=3cm, bottom=3cm, includefoot]{geometry} \usepackage{caption} \captionsetup{labelsep=period} \usepackage{titlesec} \titleformat{\section} {\normalfont\normalsize\bfseries}{\thesection}{1em}{} \titlespacing*{\section}{0cm}{0.7cm}{0cm} \titlespacing*{\subsection}{0cm}{0.5cm}{0cm} %%leave copyright info to the editor \CopyrightLine[Copyright]{2013}{Universitas Ahmad Dahlan. All rights reserved.} %%author \author[*]{\bfseries Badri Narayan Mohapatra} \author[ ]{\bfseries Dr. M. Ravi Kumar} \author[ ]{\bfseries Dr.Sushant Kumar Mandal} %%author's affiliation \affil[ ]{Centurian University Odisha,MNIT Jaipur, Centurian University Odisha} \affil[ ]{Paralakhemundi,Odisha,INDIA, telp/fax 06815 222150, Centurian University} \affil[*]{corresponding author, e-mail: badri1.mohapatra@gmail.com} %%title and shortitle (for footer) \title{Positioning of Light Shelves to Enhance Daylight Illuminance In Office Rooms} \shorttitle{Title of manuscript is short and clear, implies research results (First Author)} %%starting \begin{document} %%indentation. do not change \setlength{\parindent}{1.27cm} %%header and footer setting. do not change \pagestyle{fancy} \fancyhfoffset{0cm} %%journal info \journalname{Indonesian Journal of Electrical Engineering and Computer Science} \journalshortname{IJECECS} \revhistory{Received May 9, 201x; Revised August 3, 201x; Accepted August 16, 201x} \vol{x} \no{x} \doi{10.11591/telkomnika.vXiY.abcd} \months{April} \years{2013} \issn{2502-4752} %%build title \maketitle \begin{abstract} \textit{\indent %% Text of abstract This paper aims to improve daylighting by using light shelves. Horizontal light shelf is found to increase the illuminance in the interior of office rooms. By tilting external light shelf more illuminance can be achieved. For uniformity of illuminance tiltable light shelf is the best choice instead of horizontal light shelf. The performance of light shelves was examined through simulations on DIALux and compared with experimental values obtained from a prototype. Substantial improvement in illuminance is obtained in the experimental studies with tilted light shelves on the prototype. %% } \end{abstract} \begin{keyword} \textit{ %%write keyword here. separate by comma (,) Daylighting \sep Light shelf \sep DIALux \sep Illuminance \sep Sunlight \sep Daylight simulation. %% } \end{keyword} %% main text \section{INTRODUCTION}\label{intro} Daylight can replace electric lighting for most of the typical working day in most building types if the building is designed to allow daylight to reach most of the interior. However, daylighting is hard to achieve in most office buildings by simply applying windows around the perimeter of the building for side-lighting. Even when these windows are from wall to wall, this approach is unable to supply daylight to large portions of a building especially deeper into rear spaces away from the windows. Light tubes, pipes and fibers can be used to transport light to interior areas. Another way of enhancing daylight can be with the help of light shelves. Light shelves are typically placed above the eye level and can be internal and/ or external. The internal portion is designed to block direct sunlight from the window area above the shelf while the exterior portion shades the surface area closer to the window. Daylighting devices including light shelves have the potential to reduce the energy consumption of buildings. In general two approaches were considered, first by conducting empirical study of daylighting performance using physical scaled model with field measurement \cite{lim2012building, lim2015effects}, and second by conducting simulations using different types of software packages \cite{joarder2009simulation, safa2013study}. A light shelf is one of the passive systems for daylight control. It is a horizontal plate made of light-colored and reflective materials that reflects daylight and is placed above human eye level in the upper half of the window. It decreases the light severity near the window and increases the light penetration depth. Most of the time, the light shelf is accompanied with shading or an external shelf to perform better \cite{moazzeni2016investigating}. Meresi focuses on evaluating daylight performance of light shelves combined with external blinds in south-facing classrooms of Athens \cite{meresi2016evaluating}. A light shelf to provide shading to the lower part of the windows leading to reduced cooling loads in summer is discussed in \cite{brotas2013parametric}. According to \cite{freewan2010maximizing}, curved light shelf could improve the daylight level by 10\% compared to a horizontal lightshelf. To eliminate the effect of absorbing solar heat, shading systems should be located in the external part of the window. Both internal and external light shelves with slats are equipped to improve daylighting performance while allowing direct sun \cite{kim2012comparative}. Side-lighting systems are designed to avoid an unequal distribution of natural light which may occur through the use of traditional lateral windows \cite{gago2015natural}. Light shelves affect the architectural and structural design of a building and must be considered at the start of the design phase as they require a relatively high roof in order to function efficiently. This article focuses on light shelves which guide daylight uniformly to the room. This work mainly relates to investigate light shelf influence of selected design parameters based on simulation as well as experiment on prototype model. The paper outline is given as follows. Section \ref{intro} presents the introduction to light shelves as a passive architectural element. Section \ref{obj} describes the main objective and the context of this research. Section \ref{ls-sp} describes about adaption system of light shelf. Section \ref{dia} describes the model parameters value used in simulation as well as simulation settings and it's process. Different cases on variation with lightshelf is discussed in section \ref{res}. The article concludes with brief concluding remarks and future proposed work in section \ref{conc}. \section{Objective}\label{obj} Due to increasing awareness about environmental issues, studies focusing on reducing and saving energy are being carried out in many fields. To address this issue, the light shelf system is one among various systems that is being studied actively because it reduces the indoor lighting energy consumption in buildings and brings natural light deep into indoors, creating a balance between indoor illumination and the saving of lighting energy. Light shelves are natural lighting systems, the use of which has been highlighted in various studies as a suitable approach for reducing lighting energy consumption. Daylighting would be effective when daylight falls on windows throughout the day. Utilizing a light shelf is a common strategy for enabling daylight transmission while controlling direct sun and discomfort glare to maintain occupant comfort. It is difficult for most high-rise office buildings to achieve daylight by simply using windows around the perimeter of the side lit building. Even when these side windows are from wall to wall, this strategy can hardly supply sufficient sunlight deep into the space. New daylighting strategies are needed for better daylight utilization in office buildings, such as increased daylight penetration and uniformity. Utilizing a light shelf is such a strategy for enabling daylight transmission while controlling direct sun and discomfort glare to maintain occupant comfort. %~ \begin{figure} %~ \centering %~ \includegraphics[width=10cm]{./1eps/r} %~ \caption{Light shelf redirect light into the analysis room} \label{r} %~ \end{figure} \begin{figure} \centering \includegraphics[width=15cm]{./1eps/room} \caption{Different possible positions of light shelf for redirecting light into a room} \label{r} \end{figure} This light shelf creates one strong horizontal line, breaking the building into two parts. The lower surface of the light shelf is made of a matte finish while the upper surface is composed of specular mirror panels, reflecting sunshine to building. A light shelf is an easy solution. Installing a light weight light shelf has minimal requirements/ effects on the building’s structure and existing envelope. %One of the most significant benefits of using a light shelf is to improve the illuminance value of areas deep in rooms by using redirected daylight. A light shelf is not always a simple horizontal slab. It could be of different shapes, sizes, materials, and mounted at different positions. Traditional windows, as the major source of daylight, have a common problem which is uneven distribution of daylight in the room. Several innovative daylighting systems such as light shelves, fixed and movable reflective louvres, reflective sills, prismatic glazing, light pipes, etc., have been developed to address this problem \cite{hashemi2014daylighting}. Consequently, light shelves are often proposed to help reduce glare issues, providing better illumination distribution, and increasing the homogeneity of daylight distribution into the spaces \cite{berardi2015analysis}. Light shelves reduce glare issues while increasing the daylighting penetration into the space \cite{berardi2016benefits}. and \cite{oh2014sa}. Exterior shelves are more effective shading devices than interior shelves. A combination of exterior and interior shading devices will work best in providing an even illumination gradient \cite{parise2013combined}. Light shelves can be divided into superior and inferior sections. Their job is to reflect the light which shines on them towards the surface of the ceiling in order to achieve a better penetration and more uniform distribution of light, while decreasing the electrical energy consumption for lighting \cite{gago2015natural}. Although light shelves are only effective during the seasons of the year where light falls directly onto them, they help reduce glare. As they reduce levels of illumination they are not always apt for rooms with north exposure \cite{gago2015natural}. %The light shelf system can bring natural light deep into indoors and make indoor illumination distribution uniform to provide pleasant indoor environments to occupants by reflecting natural light from the outside with the reflective surface of light shelf and reflecting light again with the ceiling surface. So it is effective for energy saving \cite{oh2014sa}. The objectives of this article can be summarized in the following sentences. (i) To look at different configurations of lightshelves as an effective technique that can be used to enhance natural lighting in offices. (ii) Assess the energy savings (direct and indirect) that could be achieved by employing lightshelves into different deep plan office alternatives. (iii) Investigate into what makes a good daylight design, and what could be reliable. (iv) Suggest better ways to input an interface between daylighting and artificial lighting during non-occupancy hours when lights will be switched off completely while keeping in mind energy consumption levels. \section{Light shelf system parameters} \label{ls-sp} Light shelves position and placement may vary depending on the required demands, in which it may be external or internal, or even placed at both areas. The ultimate goal is to use lightshelves as a strategy which can help illuminate deeper into office spaces, yet at the same time at a controlled brightness level, within the occupants fields of vision. It is important to consider various implications that can have a great significance on daylighting design strategies. Utilizing all-glass facades in buildings is a way to help improve the quality of lighting in the indoor environment, however varying temperatures from the exterior environment limits such use. Having attained a deeper understanding of light shelves, it is important to summarize the advantages and disadvantages of using light shelves before considering their best and suitable ways of use. Assessing light shelves as part of innovative systems could decrease energy loads, as well as enhance lighting deeper into the office spaces. Light shelves are placed internally and externally, to make an effort to penetrate most light into spaces without glare and discomfort. \subsection{Variables Affecting Light Shelf Performance} Because a light shelf can function to block the direct sun and help increase daylight levels deep within a room, there is a need to find a balance between how much light to block at the front of room and how much light to direct into the rear of room. Variables affecting the performance include light shelf mounting height, geometry, building orientation, location, material and the climate conditions effect on light shelf. \subsubsection{Light Shelf Mounting Position} External light shelves could redirect more daylight and work as overhang devices. Light shelves could locate inside or outside or on both sides of a building facade. Internal light shelf blocks less daylight, and it usually redirects less daylight into buildings. Combined light shelves could utilize the advantages of both and provide the most evenly distributed illumination. Light shelves often require high ceilings in a room. This is because a light shelf mounted below eye-level will reflect sunshine directly to people’s eyes. Lower light shelf could also block people’s visibility of exterior view, reduce ceiling height visually and make people near window uncomfortable. \begin{figure} \centering % \begin{subfigure}[t]{6.5cm} % \centering % \includegraphics[width=6.5cm]{./1eps/2nd} % \caption{Case 1: Internal \& external light shelf}\label{st1} % \end{subfigure} % \quad \includegraphics[width=6.5cm]{./1eps/3rd} \caption{Case 1: Internal light shelf}\label{st2} \end{figure} % \quad \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/external} \caption{Case 2: External light shelf}\label{st3} \end{figure} % \quad \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/angular} \caption{Case 3: Internal \& angled external light shelf}\label{st4} \end{figure} % \quad \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/5th} \caption{Case 4: External flat \& angled light shelf}\label{st5} \end{figure} % \quad \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/6th} \caption{Case 5: Externally hanging angled light shelf}\label{st6} \end{figure} % \caption{Different possible cases to analyse suitable configuration of light shelves}\label{c1-4} %\end{figure} Figure \ref{st2} to \ref{st6} shows different possible models of light shelves which can be used to provide uniform daylight in a room. The different models are indicated from Case 1 %Case 1 to Case 5 and are discussed below. %Case 1 represents a flat light shelf in which half of the length is inside a room and the other half is outside the room. % It is named as `Internal \& external light shelf' in Fig. \ref{st1}. The same is also indicated as `Int+Ext' in Results section of this article. Case 1 represents a flat light shelf which is completely inside a room starting at a window. It is named as `Internal light shelf' in Fig. \ref{st2}. The same is also indicated as `Int' in Results section of this article. Case 2 represents a flat light shelf which is completely outside a room starting at a window. It is named as `External light shelf' in Fig. \ref{st3}. The same is also indicated as `Ext' in Results section of this article. Case 3 represents a light shelf in which half of the length is flat and inside a room. The other half is outside the room and tilted at an angle from the inside half. It is named as `Internal \& angled external light shelf' in Fig. \ref{st4}. The same is also indicated as `Int+Ext at $\theta^\circ$' in Results section of this article, where $\theta^\circ$ represents the tilt angle of the light shelf in the outside with respect to the flat light shelf plane. Case 4 represents a light shelf in which half of the length is flat and outside a room. The other half is also outside the room but is tilted at an angle from the other half. It is named as `External flat \& angled light shelf' in Fig. \ref{st5}. The same is also indicated as `Ext half at $\theta^\circ$' in Results section of this article. Case 5 represents an angled light shelf which is at a distance from a window. It is named as `Externally hanging angled light shelf' in Fig. \ref{st6}. The same is also indicated as `Hang at $\theta^\circ$, d $mt$' in Results section of this article, where d $mt$ is the distance of the light shelf from the window in meters. \subsubsection{Ceiling Effects} The ceiling slope is another important characteristic. Freewan et. al. \cite{freewan2010maximizing} tested a room with a fixed light shelf and compared the four different ceiling geometries: flat ceiling, curved ceiling, chamfered and sloped ceiling on their effect of daylight performance of the test room. %Findings of the test need to be written. \subsubsection{Material, Climate, Building Orientations and Location} Most light shelf products are made of opaque materials. The solid geometry blocks direct sunshine and prevents glare in areas near windows while the high reflectance surface reflect lights into deeper part of rooms. Climate is a key question when considering light shelf applications. Light shelves may not be suitable for all climates. To make a light shelf function properly, it should be designed specifically for each climate, latitude and window orientation. %A light shelf does not necessarily improve the total illuminance level of the whole room, but it can transfer the light from the front to the back of the room and make the light more even. Light shelf position and configuration will have significant effects on light shelf performance. But more research on the combined effects of these factors should be conducted. The characteristics of a light shelf that affect its lighting performance include its type, angle, height, and width. %An external light shelf has excellent lighting performance; however, it is also affected by factors such as climate and weather. \section{DIALux}\label{dia} Recent studies on light shelf have incorporated various technologies to improve efficiency. Among those technologies, the angle of the light shelf is an important element that determines efficiency. Thus, a detailed verification is performed on an office building by using DIALux lighting simulation tool. Performance is evaluated and its effectiveness is verified through a scaled prototype. \subsection{Model parameters and material specifications}\label{pns} Simulation parameters in DIALux analysis were set to specific values as shown in Table \ref{DIA}, and other study space dimensions of the room under consideration are shown in Table \ref{Stu}. \begin{table} \centering \caption{DIALux analysis parameters}\label{DIA} \begin{tabular}{|c|c@{}|c@{}|c|c@{}|c|} \hline Location & Bhubaneswar latitude, 20.25 and longitude, -85.83 \\ \hline Orientation & South facing \\ \hline Room Dimension & 3$mt$ x 5$mt$ x 2.8$mt$ (W x D x H) \\ \hline Time & March 21 \\ \hline Condition & CIE Overcast/ Clear Sky \\ \hline Type & Illuminance (Lux) \\ \hline \end{tabular} \end{table} \begin{table} \centering \caption{Study space dimensions}\label{Stu} \begin{tabular}{|c|c@{}|c@{}|c|c@{}|c|} \hline Type & Dimensions \\ \hline Floor area & 15 ($mt^2$) \\ \hline Light shelf area & 0.9 ($mt^2$) \\ \hline Working plane level & 0.85 ($mt$) \\ \hline Lintel level & 2.1 ($mt$) \\ \hline Window height & 1.25 ($mt$) \\ \hline Window width & 1.8 ($mt$) \\ \hline \end{tabular} \end{table} Various types of materials were used in the study space model in DIALux analysis. Table \ref{refl} shows the list of elements, area covered by the elements, and their reflectance values that are analyzed in this study. \begin{table} \centering \caption{Study space dimensions and its reflectance}\label{refl} \begin{tabular}{|c|c@{}|c@{}|c|c@{}|c|} \hline Element & Area & Reflectance \\ \hline Wall & 15 ($mt^2$) & 50 (\%) \\ \hline Ceiling & 0.9 ($mt^2$) & 70 (\%) \\ \hline Floor & 0.85 ($mt$) & 20 (\%) \\ \hline Light Shelf & 1.8($mt$)x 0.5($mt$) (W x D)=0.9 ($mt^2$) & 90 (\%) \\ \hline \end{tabular} \end{table} \begin{figure} \centering \includegraphics[width=12cm]{./1eps/st} \caption{Window size and representation of the room under consideration} \label{st} \end{figure} \section{Results and Discussion} \label{res} Based on the parameters and specifications given in section \ref{pns}, DIALux simulations are conducted for different possible cases discussed in section \ref{ls-sp} to analyse suitable configurations. Illuminance values for some of the cases are measured from a developed prototype. Table \ref{lshelf} shows the simulation results for the different cases indicated in Fig. \ref{c1-4}. Simulations convey the daylight values ($DL$) with units in lux at 9 am, noon and 3 pm. Simulation results of Table \ref{lshelf} indicate that maximum daylight is available without light shelves. Daylight values in afternoon are lower than those in morning and noon because daylight does not fall on the window directly in the afternoon for the simulated room orientation and location. It is observed that at all times of day, light shelves as given in the four cases are able to reduce the amount of daylight entering the room. From Table \ref{lshelf}, comparison of %Case 1 and Case 4 shows that a tilt angle of 10$^\circ$, 15$^\circ$, 20$^\circ$ and 25$^\circ$ slightly increases the daylight illuminance in the room. \begin{table} \centering \caption{Simulation performance of light shelf on investigated room at G$_8$}\label{lshelf} \begin{tabular}{|c|c|c|c|c|c|c|} \hline Investigated & Morning (9 am) & Noon & Afternoon (3 pm) \\ \cline{2-4} Room & $DL$ (lux) & $DL$ (lux) & $DL$ (lux) \\ \hline Without Light shelf & 99 & 120 & 72 \\ \hline %Case 1 (Int+Ext) & 86 & 104 & 63 \\ \hline Case 1 (Int) & 86 & 105 & 63 \\ \hline Case 2 (Ext) & 82 & 100 & 60 \\ \hline Case 4 (Int+Ext at 10$^\circ$) & 88 & 109 & 65 \\ \hline Case 4 (Int+Ext at 15$^\circ$) & 69 & 109 & 65 \\ \hline Case 4 (Int+Ext at 20$^\circ$) & 89 & 108 & 65 \\ \hline Case 4 (Int+Ext at 25$^\circ$) & 87 & 106 & 64 \\ \hline Case 4 (Int+Ext at 30$^\circ$) & 86 & 104 & 63 \\ \hline \end{tabular} \end{table} \begin{figure} \centering \includegraphics[width=15cm]{./1eps/Gnormal} \caption{Simulated grid point display of the analysed room at 9 am for CIE overcast sky without light shelf} \label{Gn} \end{figure} Figure \ref{Gn} shows the simulation result obtained from DIALux for the investigated room without using a light shelf in case of CIE overcast sky at 9 am. Lux values corresponding to the middle row of Fig. \ref{Gn} are noted and analyzed in the tabulated results. Table \ref{grid} shows the daylight intensity at specific grid points (G$_1$ to G$_8$) of a room. It is observed that maximum daylight is available in the room for %Case 1 when light shelf is not used. In Case 4, internal with external light shelf at tilt angles of 10$^\circ$ or 20$^\circ$ gives better illuminance except at grid G$_1$. \begin{table} \centering \caption{Simulation Performance of light shelf on grid points of investigated room for CIE overcast sky}\label{grid} \begin{tabular}{|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{} |c@{}|c|} \hline Investigated & \multicolumn{8}{c|}{Morning Time $DL$(lux) on grid points} \\ \cline{2-9} Room & G$_1$ & G$_2$ & G$_3$ & G$_4$ & G$_5$ & G$_6$ & G$_7$ & G$_8$ \\ \hline Without Light shelf & 225 & 544 & 410 & 280 & 187 & 134 & 103 & 90 \\ \hline %Case 1 (Int+Ext) & 78 & 302 & 263 & 199 & 149 & 111 & 88 & 77 \\ \hline Case 1 (Int) & 230 & 506 & 373 & 240 & 155 & 112 & 87 & 88 \\ \hline Case 2 (Ext) & 189 & 283 & 271 & 204 & 146 & 108 & 85 & 74 \\ \hline Case 4 (Int+Ext at 10$^\circ$) & 88 & 372 & 279 & 240 & 154 & 116 & 91 & 80 \\ \hline %Case 4 (Int+Ext Angle 15 deg) & 93 & 346 & 288 & 214 & 153 & 115 & 91 & 80 \\ \hline Case 4 (Int+Ext at 20$^\circ$) & 98 & 357 & 281 & 214 & 156 & 117 & 92 & 81 \\ \hline %Case 4 (Int+Ext Angle 25 deg) & 103 & 365 & 304 & 224 & 161 & 116 & 90 & 78 \\ \hline %Case 4 (Int+Ext Angle 30 deg) & 142 & 373 & 310 & 226 & 160 & 115 & 89 & 77 \\ \hline \end{tabular} \end{table} A scaled prototype of the simulated room is setup using plywood. The prototype is scaled at a ratio of 1:15 with respect to the simulation model to validate the obtained simulation results. Considering the room dimensions given in Table \ref{DIA}, the prototype room dimensions are 0.2$mt$ x 0.33$mt$ x 0.19$mt$ (W x D x H). A luxmeter is used to measure the illuminance levels in the prototype. Measurements were taken in an intermediate sky condition. A photograph of the implemented prototype is shown upside down in Fig. \ref{extg} to indicate mirrored roof and light shelf. Figure \ref{exter} is photograph of implemented prototype using %internal and external light shelf (Case 1) for external light shelf (Case 2). In prototype, we are using a glass mirror as light shelf. One side of the glass is coated with red paint to restrict transmission of light to other side of the glass. Light rays falling on the mirror are reflected to desired area on the roof from where the light is reflected to the interior of the room. %Measurments are noted and shown in tables \ref{expg12} and \ref{expg3}. \begin{figure} \centering \includegraphics[width=8.5cm]{./1eps/opposite} \caption{Actual image of implemented prototype (upside down)}\label{extg} \end{figure} %\begin{figure} % \centering % \includegraphics[width=8.5cm]{./1eps/bothie} % \caption{Prototype with internal and external light shelf}\label{bothie} %\end{figure} \begin{figure} \centering \includegraphics[width=8.5cm]{./1eps/exter} \caption{Prototype with external light shelf}\label{exter} \end{figure} \subsection{External light shelf positioning} Assuming a height of 1.8 $mt$ for inhabitants of a room, mounting position of an external light shelf is varied from 1.5 $mt$ to 1.8 $mt$. Figures \ref{st5} and \ref{st6} show two different configurations of external light shelf experimented with varying angles to maximize daylight at G$_8$. Figure \ref{st5} indicates an external light shelf in which half of the shelf is connected perpendicular to the window, and the further half of the shelf is tilted at an angle with the perpendicular shelf. Figure \ref{st6} indicates a tilted external light shelf which is positioned at a distance from the window. %we can see the building room with different angles with external light shelf. It's simulation result shown in figure \ref{st13-16} An iron frame is attached at the window to hold an external light shelf at various tilt angles as shown in Fig. \ref{st35} and \ref{st36}. Illumination measurements for different positions of light shelf at different time instants was noted. %The most relevant measured values are shown in Table \ref{expg3}. \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/frame} \caption{Metal frame to hold external light shelf at a height of 10.7$cm$ from ground}\label{st35} \end{figure} \begin{figure} \centering \includegraphics[width=6.5cm]{./1eps/framegl} \caption{External light shelf at a height of 10.7$cm$ with reflecting mirror }\label{st36} \end{figure} % \caption{Prototype to test external light shelf at different tilt angles}\label{st35-36} %\end{figure} During actual measurements it was found that the sky was clear for the day simulation results were obtained. Since the number of results obtained was large, only important data is retained in this article and results from simulation and actual measurements for similar cases are shown in same tables. Tables \ref{grid-cs-9}, \ref{grid-cs-1} and \ref{grid-cs-5} show the simulation and measured illumination values corresponding to 9 am, 1 pm and 5 pm respectively. In these tables, the distance of light shelf from the window for Case 5 is shown as $d1$, $d2$ and $d3$. Actual values of $d1$, $d2$ and $d3$ from the prototype are 0.1 $mt$, 0.15 $mt$ and 0.2 $mt$ respectively. Since the prototype is scaled down by a factor of 15, the values of $d1$, $d2$ and $d3$ in the simulations are taken as 1.5 $mt$, 2.25 $mt$ and 3 $mt$ respectively. \begin{table} \centering \caption{Performance of light shelf on grid points of investigated room at 9 am for CIE clear sky} \label{grid-cs-9} \begin{tabular}{|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{} |c@{}|c|} \hline Investigated & \multicolumn{4}{|c|}{Simulated $DL$(lux)} & \multicolumn{4}{|c|}{Measured $DL$(lux)} \\ \cline{2-9} Room & G$_1$ & G$_4$ & G$_7$ & G$_8$ & G$_1$ & G$_4$ & G$_7$ & G$_8$ \\ \hline Without Light shelf & 654 & 615 & 282 & 259 & 2800 & 900 & 200 & NA \\ \hline %Case 1 (Int+Ext) & 2490 & 536 & 222 & 215 & 300 & 800 & 400 & NA \\ \hline Case 1 (Int) & 4426 & 525 & 260 & 242 & 500 & 700 & 400 & NA \\ \hline Case 2 (Ext) & 3252 & 533 & 228 & 220 & 1200 & 1000 & 300 & NA \\ \hline Case 4 (Ext at 15$^\circ$) & 2915 & 526 & 224 & 213 & 1600 & 900 & 200 & NA \\ \hline Case 4 (Ext at 20$^\circ$) & 2976 & 528 & 222 & 210 & 1600 & 800 & 200 & NA \\ \hline Case 4 (Ext at 25$^\circ$) & 3269 & 560 & 223 & 207 & 1800 & 900 & 200 & NA \\ \hline Case 4 (Ext at 30$^\circ$) & 3099 & 525 & 216 & 202 & 1800 & 600 & 200 & NA \\ \hline Case 4 (Ext at 35$^\circ$) & 3170 & 522 & 211 & 197 & 1800 & 500 & 100 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d1$) & 4557 & 659 & 279 & 267 & 4700 & 1500 & 300 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d1$) & 4578 & 652 & 275 & 262 & 4600 & 1200 & 300 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d1$) & 4546 & 639 & 268 & 254 & 4500 & 1500 & 300 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d2$) & 4590 & 661 & 280 & 268 & 4700 & 1300 & 300 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d2$) & 4585 & 657 & 275 & 262 & 4600 & 1400 & 300 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d2$) & 4576 & 652 & 270 & 258 & 4500 & 1200 & 300 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d3$) & 4595 & 661 & 280 & 268 & 4400 & 1400 & 300 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d3$) & 4591 & 657 & 277 & 264 & 4400 & 1500 & 300 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d3$) & 4586 & 654 & 273 & 260 & 4600 & 1400 & 300 & NA \\ \hline \end{tabular} \end{table} Table \ref{grid-cs-9} shows the illumination values from simulation results corresponding to CIE clear sky at 9 am for different cases of light shelf positioning. Table \ref{grid-cs-9} also shows actual measurements for CIE clear sky at 9 am. It is observed that at G$_1$, maximum daylight of 4700 lux is obtained with an external light shelf (Case 5) inclined at 10$^\circ$ with the horizontal and placed at a distance of 0.1 to 0.15 $mt$ from the window. It is seen that correlation between simulation and actual results is very low, except for lux values corresponding to Case 5 at G$_1$. The large variation in the two results is attributed to difference in parameters employed. We used mirrors in the prototype which have a reflectivity close to 99\%, whereas the reflectivity used in the simulations was 90\% (maximum allowed by simulator) for light shelf and 70\% for roof. Another factor beyond our control were passing clouds which caused very low lux values for some of the actual measurements (lux values at G$_1$ for %Case 1 and Case 1). \begin{table} \centering \caption{Performance of light shelf on grid points of investigated room at 1 pm for CIE clear sky} \label{grid-cs-1} \begin{tabular}{|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{} |c@{}|c|} \hline Investigated & \multicolumn{4}{|c|}{Simulated $DL$(lux)} & \multicolumn{4}{|c|}{Measured $DL$(lux)} \\ \cline{2-9} Room & G$_1$ & G$_4$ & G$_7$ & G$_8$ & G$_1$ & G$_4$ & G$_7$ & G$_8$ \\ \hline Without Light shelf & 25878 & 668 & 311 & 275 & 84000 & 9800 & 1100 & NA \\ \hline %Case 1 (Int+Ext) & 2594 & 560 & 232 & 227 & 1600 & 12000 & 1400 & NA \\ \hline Case 1 (Int) & 8666 & 545 & 266 & 261 & 31700 & 12000 & 1400 & NA \\ \hline Case 2 (Ext) & 7372 & 542 & 227 & 229 & 2500 & 14100 & 1400 & NA \\ \hline Case 4 (Ext at 15$^\circ$) & 7018 & 534 & 222 & 220 & 3600 & 2900 & 600 & NA \\ \hline Case 4 (Ext at 20$^\circ$) & 8090 & 570 & 233 & 231 & 3600 & 2800 & 600 & NA \\ \hline Case 4 (Ext at 25$^\circ$) & 7149 & 535 & 218 & 213 & 4200 & 3900 & 700 & NA \\ \hline Case 4 (Ext at 30$^\circ$) & 7217 & 534 & 214 & 209 & 4200 & 3800 & 700 & NA \\ \hline Case 4 (Ext at 35$^\circ$) & 7294 & 531 & 209 & 204 & 5700 & 3300 & 600 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d1$) & 8808 & 693 & 290 & 276 & 77100 & 4600 & 900 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d1$) & 8835 & 686 & 285 & 271 & 82100 & 4700 & 950 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d1$) & 8798 & 672 & 278 & 263 & 82300 & 4800 & 1000 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d2$) & 8842 & 695 & 290 & 277 & 77000 & 5200 & 800 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d2$) & 8840 & 690 & 285 & 271 & 82000 & 4600 & 900 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d2$) & 8825 & 685 & 281 & 266 & 82000 & 4900 & 1200 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d3$) & 8848 & 695 & 291 & 278 & 77400 & 4500 & 800 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d3$) & 8846 & 691 & 287 & 273 & 5500 & 4500 & 800 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d3$) & 8844 & 688 & 283 & 269 & 76200 & 4500 & 700 & NA \\ \hline \end{tabular} \end{table} Table \ref{grid-cs-1} shows simulation results and actual measurements of daylight penetration in a room corresponding to CIE clear sky at 1 pm for different cases of light shelf positioning. In order to find the best possible angle of light shelf, the tilt angle in simulations is varied from 10$^\circ$ to 35$^\circ$. It is observed that with external light shelf at the window, daylight at grid points G$_1$, G$_2$ \& G$_3$ is less than without a light shelf. This is attributed to light blocked by the light shelf. For the instances when the light shelf is at a distance from the window, daylight values are higher than when the light shelf is at the window. The daylight values are higher at grid points deeper in the room than without a light shelf. It is observed that at G$_1$, maximum daylight of 4700 lux is obtained with an external light shelf inclined at 10$^\circ$ with the horizontal and placed at a distance of 0.1 to 0.15 $mt$ from the window. The table also shows actual measurements from the prototype for CIE clear sky at 9 am. It is observed that correlation between simulation and actual results is very low, except for lux values corresponding to Case 5 at G$_1$. The large variation in the two results is attributed to difference in parameters employed. We used a mirror in the prototype which has a reflectivity close to 99\%, whereas the reflectivity used in the simulations was 90\% and 70\% for light shelf and roof respectively. Passing clouds have caused very low lux values for some of the actual measurements %(Case 1 G$_1$ lux value???). In Table \ref{grid-cs-5} for Case 4 simulation results, a 20$^\circ$ tilt gives better illumination towards the interior of the room compared with other tilt angles. As we increase the distance of light shelf from window from 0.4$mt$ to 0.8$mt$, the lux value increases inside each grid point (except G$_1$) of the investigated room. It is observed that increasing the distance of the hanging light shelf from window beyond 0.7 $mt$ does not increase the daylight penetration into the room. \begin{table} \centering \caption{Performance of light shelf on grid points of investigated room at 5 pm for CIE clear sky} \label{grid-cs-5} \begin{tabular}{|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{} |c@{}|c|} \hline Investigated & \multicolumn{4}{|c|}{Simulated $DL$(lux)} & \multicolumn{4}{|c|}{Measured $DL$(lux)} \\ \cline{2-9} Room & G$_1$ & G$_4$ & G$_7$ & G$_8$ & G$_1$ & G$_4$ & G$_7$ & G$_8$ \\ \hline Without Light shelf & 70 & 90 & 49 & 44 & 800 & 410 & 80 & NA \\ \hline %Case 1 (Int+Ext) & 395 & 96 & 42 & 43 & 70 & 270 & 100 & NA \\ \hline Case 1 (Int) & 510 & 90 & 43 & 43 & 250 & 125 & 90 & NA \\ \hline Case 2 (Ext) & 389 & 94 & 41 & 42 & 250 & 350 & 90 & NA \\ \hline Case 4 (Ext at 15$^\circ$) & 344 & 93 & 40 & 41 & 300 & 340 & 100 & NA \\ \hline Case 4 (Ext at 20$^\circ$) & 389 & 94 & 41 & 42 & 300 & 340 & 100 & NA \\ \hline Case 4 (Ext at 25$^\circ$) & 363 & 93 & 39 & 39 & 300 & 350 & 100 & NA \\ \hline Case 4 (Ext at 30$^\circ$) & 370 & 92 & 38 & 38 & 340 & 390 & 80 & NA \\ \hline Case 4 (Ext at 35$^\circ$) & 380 & 92 & 37 & 37 & 320 & 330 & 70 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d1$) & 517 & 107 & 45 & 45 & 470 & 180 & 50 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d1$) & 522 & 105 & 44 & 44 & 490 & 170 & 50 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d1$) & 514 & 103 & 42 & 42 & 500 & 170 & 40 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d2$) & 523 & 107 & 45 & 45 & 620 & 250 & 50 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d2$) & 522 & 106 & 44 & 44 & 540 & 240 & 50 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d2$) & 519 & 105 & 43 & 43 & 510 & 210 & 50 & NA \\ \hline Case 5 (Ext at 10$^\circ$, $d3$) & 524 & 107 & 45 & 45 & 610 & 280 & 50 & NA \\ \hline Case 5 (Ext at 20$^\circ$, $d3$) & 524 & 106 & 44 & 44 & 620 & 260 & 50 & NA \\ \hline Case 5 (Ext at 30$^\circ$, $d3$) & 522 & 106 & 43 & 43 & 630 & 330 & 50 & NA \\ \hline \end{tabular} \end{table} Table \ref{grid-cs-5} shows simulation and actual measurements of illumination for CIE clear sky at 9 am corresponding to different cases of light shelf positioning. It is observed that at G$_1$, maximum daylight of 4700 lux is obtained with an external light shelf inclined at 10$^\circ$ with the horizontal and placed at a distance of 0.1 to 0.15 $mt$ from the window. It is observed that correlation between simulation and actual results is low, except for lux values corresponding to Case 5 at G$_1$. The variation in the two results is attributed to difference in parameters employed. We used a mirror in the prototype which has a reflectivity close to 99\%, whereas the reflectivity used in the simulations was 90\% and 70\% for light shelf and roof respectively. Passing clouds caused very low lux values for some of the actual measurements %(Case 1 G$_1$ lux value???). \begin{figure} \centering \includegraphics[width=12cm]{./1eps/ls} \caption{Mirror representation in front of prototype window} \label{mirror} \end{figure} In Tables \ref{grid-cs-9}, \ref{grid-cs-1} and \ref{grid-cs-5}, we observe that there is small enhancement in illumination in some grid points. We have attempted another configuration of hanging light shelf to get substantial illumination enhancement in the interiors. Figure \ref{mirror} is a crude attempt to enhance the daylight illumination in the interiors. Table \ref{expg3} shows the measured illumination at all grid points obtained at different times of the day. As given in the table, we have obtained substantial enhancement of illumination at grid points G$_6$ \& G$_7$ and significant enhancement in all the other grid points by this directed reflection approach. \begin{table} \centering \caption{Experiment Performance of light shelf on grid points of investigated room}\label{expg3} \begin{tabular}{|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{}|c@{} |c@{}|c|} \hline Investigated & \multicolumn{7}{c|}{ Time(9 am) $DL$(lux) on grid points} \\ \cline{2-8} Prototype Room & G$_1$ & G$_2$ & G$_3$ & G$_4$ & G$_5$ & G$_6$ & G$_7$ \\ \hline Without light shelf & 2600 & 2200 & 1600 & 1200 & 900 & 700 & 500 \\ \hline With ext light shelf & 4100 & 3100 & 2300 & 2200 & 2700 & 10000 & 22000 \\ \hline Investigated & \multicolumn{7}{c|}{ Time(11 am) $DL$(lux) on grid points} \\ \cline{2-8} Prototype Room & G$_1$ & G$_2$ & G$_3$ & G$_4$ & G$_5$ & G$_6$ & G$_7$ \\ \hline Without light shelf & 5700 & 3700 & 2200 & 1700 & 1200 & 900 & 700 \\ \hline With ext light shelf & 6300 & 4300 & 3400 & 2700 & 9100 & 12400 & 24400 \\ \hline Investigated & \multicolumn{7}{c|}{ Time(1 pm) $DL$(lux) on grid points} \\ \cline{2-8} Prototype Room & G$_1$ & G$_2$ & G$_3$ & G$_4$ & G$_5$ & G$_6$ & G$_7$ \\ \hline Without light shelf & 6900 & 4700 & 2400 & 1800 & 1300 & 900 & 800 \\ \hline With ext light shelf & 7300 & 5400 & 3100 & 2900 & 9600 & 25400 & 22700 \\ \hline Investigated & \multicolumn{7}{c|}{ Time(3 pm) $DL$(lux) on grid points} \\ \cline{2-8} Prototype Room & G$_1$ & G$_2$ & G$_3$ & G$_4$ & G$_5$ & G$_6$ & G$_7$ \\ \hline Without light shelf & 4000 & 2700 & 1500 & 1200 & 800 & 500 & 400 \\ \hline With ext light shelf & 4400 & 2700 & 1700 & 1600 & 13900 & 17500 & 16200 \\ \hline \end{tabular} \end{table} \section{Conclusion}\label{conc} We have used simulations and measurements from a prototype to verify illumination performance with a light shelf in indoor environments. The results of this article is summarised in the following sentences. Light shelves improve daylight penetration with some configurations. Light shelves provide shading near the window and eliminate glare. Light shelves can improve uniformity of daylighting. This article confirms that a lightshelf can ensure great promise in terms of energy efficiency and daylight penetration into deep areas. Some aspects on which further work can be carried out are given below. Even though the prototype gives very good results, we need to develop models to find exact positioning of light shelf which can give best interior environment. Throughout the study, the interior environment was considered to be a vacant office for the simulation and modeling process. However, the addition of furniture and arrangements in the space could have a different effect on the output of the study. This research could also continue in the path of testing the effect of internal materials and finishes, ceiling heights, glass material on daylight penetration. The study should look at most different reflective materials, as well as compare between them. %% The Appendices part is started with the command \appendix; %% appendix sections are then done as normal sections %% \appendix %% \section{} %% \label{} %% References %% %% Following citation commands can be used in the body text: %% Usage of \cite is as follows: %% \cite{key} ==>> [#] %% \cite[chap. 2]{key} ==>> [#, chap. 2] %% %% References with BibTeX database: \bibliographystyle{IEEEtran} \bibliography{mybibfile} \end{document} %\bibliographystyle{IEEEtran} %\bibliography{} %% Authors are advised to use a BibTeX database file for their reference list. %% The provided style IEEEtran.bst formats references is generally used. %% For references without a BibTeX database: % %% %% End of file `telkomnika.tex'.