J. Cent. South Univ. (2012) 19: 777-782
DOI: 10.1007/s11771-012-1071-9
Sky luminance distribution types in China
HE Ying(何荥)1, 2, 3, MAO Hua-song(毛华松)1, 2, QUAN Li(全利)4
1. College of Architecture and Urban Planning, Chongqing University, Chongqing 400045, China;
2. Key Lab of Chinese Education Ministry for Construction and New Technology of Mountain Cities,Chongqing University, Chongqing 400045, China;
3. State Key Laboratory of Subtropical Building Science, South China University of Technology,Guangzhou 510641, China;
4. Chongqing Vocational of Architecture Engineering, Chongqing 400039, China
? Central South University Press and Springer-Verlag Berlin Heidelberg 2012
Abstract: According to Chinese daylighting climate data, combined with the standard of Commission Inernationale de l’Eclairgae (CIE) general sky luminance distribution (SLD), the sky luminance distribution of typical daylighting climate zones in China was studied. Through the research on reference SLD of overcast, intermediate and clear types, the standards of clear and overcast SLD are consistent with CIE general sky, and the intermediate SLD has greater error with CIE general sky. The main types of Chinese sky luminance in different daylighting climate zones were obtained, which provide the basic data for simulation and forecast of China natural light, and provide reference for using the standard amendments of China future natural light.
Key words: sky luminance distribution; sky type; daylighting
1 Introduction
Sky luminance distribution (SLD) is constantly changed, not only associated with the local geographical latitude and topography, but also influenced by climatic conditions, solar elevation and atmospheric transparency. For example, in North China, climate is less sunny and more cloudy, and in south China, it is less cloudy and more sunny especially city of Chongqing, Sichuan province and Guizhou Province. Therefore, different areas have different sky luminance distributions [1]. It is difficult to describe the practical sky luminance distribution. For solving the problem, the reference sky was used as a representative of the true sky, and treated as the basis for daylighting design and calculation [2].
Commission Internationale de l’Eclairage (CIE) general sky standard was proposed for the standardization of exterior daylighting conditions for the worldwide application. The new concept of sky luminance is modeling skies under a wide range of occurrences from overcast sky to cloudless situations without or with sunlight, respectively. Based on the model, fifteen kinds of sky types were proposed and five kinds of them were overcast, five kinds of them were clear and five kinds of them were intermediate [3].
Recently, the local sky luminance distribution was studied through the daylighting climate observation data in England [4-5], Greece [6], Hong Kong [7-8], Thailand [9], Singapore [10], Japan [11] and China [12], and the local sky luminance distribution of common model was predicted for the local use of natural light to provide a good foundation.
Architectural daylighting design standards of China still use the model of overcast sky luminance as a standard [13]. The researches on SLD in Chinese daylighting climate zones were rarely reported after the general sky standard was issued. As the vast geography and changing climate, the study of sky luminance distribution in Chinese typical daylighting climate district can provide more accurate forecast and simulation of the basic data, and at the same time, it can be available for future natural light to use the standard amendments to provide relevant information.
2 CIE general sky standard
The CIE general sky standard was adopted and published as ISO 15469/CIE S003 standard in 2004. Compared to the previous standards, the general standard is more detailed and covers the whole occurrence spectrum considering different diffusion
scattering by the atmosphere and effects of direct sunlight.
The ratio of the luminance Lγ in an arbitrary sky element to the zenith luminance LZ is expressed in a functional formula as
(1)
The luminance gradation function φ is related the luminance of a sky element to zenith angle:
φ(Z)=1+aexp(b/cosZ), 0≤Z≤π/2 (2)
And at the horizon it is φ(π/2)=1; at the zenith,
φ(0)=1+aexpb (3)
The function f expresses the scattering indicatrix which is related the relative luminance of sky element to its angular distance from the sun:
(4)
Its value at the zenith is expressed as
(5)
The position of sun and that of the arbitrary sky element as well as parameters a, b, c, d and e (see Table 1) which describe atmospheric condition are input calculation quantities. The position of the arbitrary sky element is defined by zenith angle Z and the azimuthal difference Az between the element and the solar meridian as
χ=arcos(cosZs·cosZ+sinZs·sinZ·cosAZ) (6)
where AZ=|α-αs|.
CIE general sky standard has classified sky condition into three groups, i.e. overcast, clear and intermediate. Each group contains five different types of the sky, covering most of the sky type.
3 Chinese sky luminance distribution types
3.1 Data acquisition
In order to study Chinese sky luminance distribution types, two groups of daylighting climate observation data were used. One is year test statistics of fourteen daylighting weather observation stations in the typical daylighting climate zones of China during 1983-1984. The other is the day-by-minute observation data of eleven meteorological parameters during 1991-1993, such as illuminance, radiance and zenith luminance.
The original test data of two groups have already tested and verified by the technical requirements of daylighting measurement by CIE. So, the data have high credibility.
3.2 Types of sky
As described in the definition of reference sky, reference sky approximately corresponds to any of the most instantaneous real sky, and it cannot describe directly any of the most instantaneous sky. The real sky is quite different all over the world. It will be affected by geographic latitude, topography and climatic conditions, such as different climatic conditions in different parts of China. The proportion of all the sky is not the same type and the sky luminance distribution corresponding to different types of reference sky is also different.
Table1 Standard parameters
According to the typical daylighting climate observation data in China, we have a statistics of percentage for clear sky, overcast and intermediate sky in 1984 [2]. As listed in Table 2, the proportion of the intermediate type is more than 50% in different typical daylighting climate zones, thus the intermediate type is the most common climate. On the contrary, the proportion of clear and overcast is relatively low.
Table 2 Ratio (%) of overcast, clear and intermediate sky in China
The ratio of the zenith luminance to diffusion sky illuminance LZ/DV is suitable for expressing the type of sky luminance distribution. The values of zenith luminance LZ are different in different sky models, and LZ significantly affects the value of the horizontal diffusion illuminance under the sky. Therefore, the ratio of LZ/DV in the same sky kind comes directly from the integral combination of scattering and absorption functions. In theory, it represents the sky model. If the solar altitude angle is under 30°, LZ/DV can express the sky type better. When the value of LZ/DV is over 0.3, the corresponding sky type is overcast. When the value of LZ/DV is between 0.2 and 0.3, the corresponding sky type is intermediate sky. When the value of LZ/DV is less than 0.2, the corresponding sky type is usually clear [14]. Therefore, according to the division of CIE general sky standard, the sky luminance distribution of Chinese daylighting climates was studied from overcast, intermediate and clear sky types.
3.3 Overcast
Overcast is the real sky that can be expressed by the meteorological parameters of surface of the sun status Π, with total cloud amount of 10. The sky luminance distribution of overcast is stable because of the chance, which is less affected by weather condition.
We selected 2 d daylighting climates observation values as analysis data, as shown in Fig. 1. The value of LZ/DV is stable and tends to a fixed value in a stable overcast situation, which is usually more than 0.3. It matches with the overcast of CIE general sky.
The overcast is covered by clouds completely and no direct sunlight. According to the different position of the sun and the thickness of cloud, the sky type can be divided into two major categories: one is that surface of the sun is completely invisible; the other is that surface of the sun is visible. The two major categories are widespread in actual observation. The first sky type in CIE general sky is the same as CIE standard overcast sky. The third sky type in CIE general sky is similar to CIE standard overcast sky. And the latter sky luminance gradient is gentler than the first. The two sky types are widespread in China [15]. In the second and fourth sky type in CIE general sky we can observe the brightening towards the sun. This sky luminance distribution is similar to CIE standard clear sky. So, the second and fourth sky type in CIE general sky corresponds to the similar overcast sky luminance distribution in China [16]. The fifth sky type in CIE general sky is the sky of uniform luminance, and such sky type will appear in extreme conditions, for example, the dark sky before the storm. Such type sky is not the typical sky and seldom used in reality. So, the overcast reference sky which is recommended by CIE general sky can meet the daylighting demand of application and simulation in Chinese overcast climates.
Fig. 1 LZ/DV values during overcast sky varied by solar altitude in December 20th, 1991 (a), and July 1st, 1991 (b)
3.4 Clear
The clear sky refers to the cloudless sky in CIE general sky. It is the real sky that can be expressed by the meteorological parameters of surface of the sun status ☉2 (cloudless on the surface of the sun), with total cloud amount 0. Prom Fig. 2, the value of LZ/DV is almost constant and under 0.2 while sun height angler is low because it is less influenced by weather condition.
Fig. 2 LZ/DV values during clear sky varied by solar altitude in December 4th, 1991 (a), and December 6th, 1991 (b)
According to the factor of pollution, CIE general sky divides the clear type into five categories, which represent the sky luminance distribution model in difference climates. After the mass tested data analysis of clear sky luminance distribution, it is found that the measured value of sky luminance distribution is consistent with the calculated value of the 12th type and 13th type in CIE general sky [16]. The daylighting climates as studying sample in July 14th, 1991 were selected. The day is cloudless turbid sky with broad solar corona which is similar to the 14th type of CIE general sky. After removing some invalid data, the sky diffusing illuminance and zenith luminance under the 14th type in CIE general sky were calculated. The variation law was compared between measurements and calculated values with time, as shown in Fig. 3. From Fig. 3, it can be seen that the difference between test and calculated values is small. The calculated value can reflect the actual changes of diffusion illumination. Compared the measured value of zenith luminance (LVZ) and calculated value of zenith luminance (LZ) with the variation of solar altitude angle in that day are compared, as shown in Fig. 4, calculated values ??also reflect the variation of zenith luminance by the solar elevation angle. Since the 11th type and 15th type are seldom observed, it is not discussed here. Overall, clear reference sky model is consistent with Chinese daylighting climates and can meet the sunlight requirements of simulation and daylighting calculations.
Fig. 3 Comparison of calculated DV and measured data Evd varied by time
Fig. 4 Comparison of calculated LZ and measured data LVZ varied by solar altitude
3.5 Intermediate
The intermediate sky is a sky type between the clear and overcast, and it is the real sky that can be expressed by the meteorological parameters of surface of sun status Π, ☉2, ☉, ☉0, with total cloud amount between 0/0 and 10/10. Since the intermediate sky varies widely with the surface of sun status, the cloud shape and the cloud amount, the intermediate sky luminance distribution is more complex than clear and overcast sky. The intermediate sky LZ/DV is clearly varied by solar altitude, as shown in Fig. 5. The intermediate sky is difficult to be described by using few kinds of reference sky. The CIE general sky divides the clear type into five categories. Except for the surface of the sun status is not visible by the cover of cloud in the 6th type, and the meteorological parameter is Π, in the other fourth kinds of type, the sun status is visible, and the meteorological parameters are ☉2,☉ and ☉0, respectively.
The traditional daylighting climates observation data record the meteorological parameter about the amount of cloud, the shape of cloud and the surface of the sun status. But it does not record the variation of sky luminance. So, it is difficult to distinguish the intermediate type in CIE general sky by test data which is available. For verifying, suppose that five categories of reference sky of CIE general sky can describe the sky luminance distribution of intermediate sky. The intermediate sky luminance was researched by using partial data of yearly tested statistics of fourteen daylighting weather observation stations in the typical light climate zone of China during 1983-1984. The sky type as research object was selected where we can observe the sun, and the cloud amounts are between 1 to 9 in Guangzhou and Chongqing. In accordance with measured value of the parallel beam/direct solar illuminance on horizontal plan (PV), luminous extinction coefficient (aV) and optical mass (m), the luminous turbidity (TV) can be calculated [17]. The value of sky zenith luminance and diffusion illuminance in different solar altitudes can be calculated and compared. The 6th type sky luminance standard of intermediate is ignored because of no direct sunlight. The value of LZ and DV of the 7th, 8th, 9th and 10th type sky luminance standard of intermediate type can be calculated. Because the limited space, the value of 10th type is not listed in Table 3.
There are large differences between the measured value and the theoretical value of intermediate sky luminance distribution in Tables 3 and 4. In order to further explain the difference between the measured value and the theoretical value of intermediate sky luminance distribution, the relationship between the diffusion illuminance and solar altitude of Chongqing and Guangzhou in 1984 was analyzed. As shown in Fig. 6. Under the influence of atmospheric conditions, solar conditions and solar altitude, the diffusion illuminance distribution scatters disorderly. The intermediate reference sky of CIE genera sky cannot cover most of the daylighting climate which occurs in intermediate sky situation. And the continuity is also poor. So, it is still needed to further study the reference sky luminance in intermediate sky daylighting climate.
Fig. 5 Intermediate sky LZ/DV varied by solar altitude in July 12th, 1991 (a), and July 15th, 1991 (b)
Table 3 Comparison of calculated and measured data of scatter lighting in Guangzhou
Table 4 Comparison of calculated and measured data of scatter lighting in Chongqing
Fig. 6 Comparison of calculated DV and measured data Evd of intermediate sky in Guangzhou (a) and Chongqing (b)
4 Conclusions
1) Five types of clear sky luminance distribution model meet the needs of daylighting design and calculation in difference daylighting climate zones in China.
2) Five types of overcast sky luminance distribution model meet the needs of daylighting design and calculation in difference daylighting climate zones in China.
3) Five types of intermediate sky luminance distribution model don’t meet the needs of daylighting design and calculation in difference daylighting climate zones in China. And the of model more types is needed to meet the requirements.
References
[1] CHEN Zhong-lin, TANG Ming-fang. Building physics (graphic version) [M]. China Architecture & Building Press, 2009: 215-222. (in Chinese)
[2] HE Ying. Study on the sky luminance distribution model by information methods [D]. Chongqing: Chongqing University (College of Agriculture and Urban Planning), 2008. (in Chinese)
[3] S011/E-2003. Commission Internationale de l?Eclairage (CIE): Spatial distribution of daylight - CIE standard general sky [S].
[4] BARTZOKAS A, KAMBEZIDIS H D, DARULA S, KITTLER R. Comparison between winter and summer sky-luminance distribution in central Europe and in the eastern Mediterranean [J]. Journal of Atmospheric and Solar-Terrestrial Physics, 2005(67): 709-718.
[5] EDWARD N, CHENG V, GADI A, MU J, LEE M, GADI A. Defining standard skies for Hong Kong [J]. Building and Environment, 2007(42): 866-876.
[6] D H W LI, C C S LAU, J C LAM. Overcast sky condition and luminance distribution in Hong Kong [J]. Build Environ, 2004(39): 101-108.
[7] IGAWA N, KOGA Y, MATSUZAWA T, NAKAMURA H. Models of sky radiance distribution and sky luminance distribution [J]. Solar Energy, 2004(77): 137-157.
[8] CHIRARATTANANON S, CHAIWIWATWORAKUL P. Distributions of sky luminance and radiance of North Bangkok under standard distributions [J]. Renewable Energy, 2007(32): 1328-1345.
[9] TREGENZA P R. Analysing sky luminance scans to obtain frequency distributions of CIE standard general skies [J]. Lighting Research and Technology, 2004, 36(4): 271-281.
[10] WITTKOPF S K, SOON L K. Analysing sky luminance scans and predicting frequent sky patterns in Singapore [J]. Lighting Res Technol, 2007, 39(1): 31–51.
[11] MARKOU M T, KAMBEZIDIS H D, BARTZOKAS A, KATSOULIS B D, MUNEER T. Sky type classification in central England during winter [J]. Energy, 2005, 30: 1667–1674.
[12] HE Ying, LIN Yong, WENG Ji. Sky luminance distribution model by ant colony systems [J]. Journal of Chongqing University, 2009, 32(7): 747-751. (in Chinese)
[13] GB/T 50033-2001. Standard for daylighting design of building [S]. (in Chinese)
[14] KITTLER R, DARULA S. Prevailing sky conditions: Identifying simple parameters for definition [J]. Light Res and Technol, 1997, 29(2): 63-68.
[15] CHEN Zhong-lin, Guo Ping. Study on sky luminance distribution with fractal theory [J]. Journal of Chongqing Jianzhu University, 1998, 20(1): 74-80. (in Chinese)
[16] XU Yu-cheng. The study of sky luminance distributions model on pseudo overcast sky [D]. Chongqing: Chongqing Jianzhu University (College of Agriculture and Urban Planning), 1995. (in Chinese)
[17] DARULA S, KITTLER R, KAMBEZIDIS H D. Guidelines for more realistic daylight exterior conditions in energy conscious designs [M]. 1998: 2-12.
(Edited by DENG Lü-xiang)
Foundation item: Project(50908239) supported by National Natural Science Foundation of China; Project(2010KB14) supported by State Key Laboratory of Subtropical Building Science of South China University of Technology, China; Project(CDJZR10190006) supported by the Fundamental Research Funds for the Central Universities, China
Received date: 2011-07-26; Accepted date: 2011-11-14
Corresponding author: HE Ying; Assistant professor; PhD; Tel: +86-13320269588; E-mail: he_ying@cqu.edu.cn
