Abstract
Climatic variables are frequently used as weighting factors to indicate the degree of clearness for interpreting sky patterns. However, such important parameters are not always widely available and their criteria to define a sky condition are not clear-cut. In addition, certain variables may be more effective than the others in terms of sky identification. This paper studies the capability of various daylight parameters, namely zenith luminance, global, direct-beam and sky-diffuse illuminance, and solar altitude for categorizing the 15 International Commission on Illumination (CIE) standard skies. A new form of artificial neural networks called probabilistic neural network (PNN) which is a powerful technique for pattern recognition was used for the analysis. The findings suggested that the PNN is an appropriate tool when a number of climatic parameters of various criteria for differentiating sky standards are employed, and the ratio of zenith luminance to diffuse illuminance (L z/D v) and solar altitude (α s) are respectively the most and the least significant input parameters for discriminating between the 15 CIE skies.
Similar content being viewed by others
References
Assunção HF, Escobedo JF, Oliveira AP (2007) A new algorithm to estimate sky condition based on 5 minutes-averaged values of clearness index and relative optical air mass. Theor Appl Climatol 90(3–4):235–248
Bartzokas A, Darula S, Kambezidis HD, Kittler R (2003) Sky luminance distribution in Central Europe and the Mediterranean area during the winter period. J Atmos Sol Terr Phys 65(1):113–119
Bartzokas A, Kambezidis HD, Darula S, Kittler R (2005) Comparison between winter and summer sky-luminance distribution in Central Europe and in the Eastern Mediterranean. J Atmos Sol Terr Phys 67(7):709–718
Basheer IA, Hajmeer MN (2000) Artificial neural networks: fundamentals, computation, design and application. J Microbiol Methods 43(1):3–31
Cacoullos T (1966) Estimation of multivariate density. Ann Inst Stat Math 18(2):179–189
Chen WC, Hsu SW (2007) A neural network approach for an automatic LED inspection system. Expert Syst Appl 33(2):531–537
Chirarattananon S, Chaiwiwatworakul P (2007) Distributions of sky luminance and radiance of North Bangkok under standard distributions. Renewable Energy 32(8):1328–1345
CIE (1994) Guide to recommended practice of daylight measurement. CIE 109. CIE, Vienna
Darula S, Kittler R (2005) New trends in daylight theory based on the new ISO/CIE sky standard 3 zenith luminance formula verified by measurement data under cloudless skies. Build Res J 53(1):9–31
Enarun D, Littlefair PJ (1995) Luminance models for overcast skies: assessment using measured data. Light Res Technol 27(1):53–58
González PA, Zamarreño JM (2005) Prediction of hourly energy consumption in building based on a feedback artificial neural network. Energy Build 37(6):595–601
Hajmeer M, Basheer I (2002) A probabilistic neural network approach for modeling and classification of bacterial growth/no-growth data. J Microbiol Meth 51(2):217–226
ISO (2004) 15469:2004(E)/CIE S 011/E:2003 spatial distribution of daylight—CIE standard general sky. ISO, Geneva
Janjai S, Masiri I, Nunez M, Laksanaboonsong J (2008) Modeling sky luminance using satellite data to classify sky conditions. Build Environ 43(12):2059–2073
Kasten F, Young AT (1989) Revised optical air mass tables and approximation formula. Appl Opt 28(22):4735–4738
Kim D, Kim DH, Chang S (2008) Application of probabilistic neural network to design breakwater armor blocks. Ocean Eng 35(3–4):294–300
Kittler R, Darula S (2002) Parametric definition of the daylight climate. Renewable Energy 26(2):177–187
Kittler R, Perez R, Darula S (1997) Sky classification respecting energy-efficient lighting, glare and control needs. J Illum Eng Soc 26(1):57–68
Kittler R, Darula S, Perez R (1998) A set of standard skies. Polygrafia, Bratislava
Lam JC, Li DHW (1996) Daylight availability in Hong Kong and energy implications. Int J Ambient Energy 17(2):79–88
Lebaron BA, Michalsky JJ, Perez R (1990) A simple procedure for correcting shadowband data for all sky conditions. Sol Energy 44(5):249–256
Lee EWM, Yuen RKK, Lo SM, Lam KC, Yeoh GH (2004) A novel artificial neural network fire model for prediction of thermal interface location in single compartment fire. Fire Saf J 39(1):67–87
Li DHW, Lam JC (2001) An analysis of climatic parameters and sky condition classification. Build Environ 36(4):435–445
Li DHW, Lam JC (2003) An analysis of all-sky zenith luminance data for Hong Kong. Build Environ 38(5):739–744
Li DHW, Lau CCS (2007) An analysis of non-overcast sky luminance models against Hong Kong data. J Sol Energy Eng 129(4):486–493
Li DHW, Tang HL (2008) Standard skies classification in Hong Kong. J Atmos Sol Terr Phys 70(8–9):1222–1230
Li DHW, Lam JC, Lau CCS (2002) A study of solar radiation daylight illuminance and sky luminance data measurements for Hong Kong. Archit Sci Rev 45(1):21–30
Li DHW, Lau CCS, Lam JC (2003) A study of 15 sky luminance patterns against Hong Kong data. Archit Sci Rev 46(1):61–68
Li DHW, Lau CCS, Lam JC (2004a) Overcast sky conditions and luminance distribution in Hong Kong. Build Environ 39(1):101–108
Li DHW, Lau CCS, Lam JC (2004b) Standard skies classification using common climatic parameters. J Sol Energy Eng 126(3):957–964
Li DHW, Tang HL, Lee EWM, Muneer T (2010) Classification of CIE standard skies using probabilistic neural network. Int J Climatol 30(2):305–315
López G, Gueymard CA (2007) Clear-sky solar luminous efficacy determination using artificial neural networks. Sol Energy 81(7):929–939
Markou MT, Bartzokas A, Kambezidis HD (2007) Generation of daylight reference years for two European cities with different climate: Athens, Greece and Bratislava, Slovakia. Atmos Res 86(3–4):315–329
Mood AM, Graybill FA (1962) Introduction to the theory of statistics. McGraw-Hill, New York
Navarro J, Sendra JJ (2008) Determination of the origin of the illumination vector due to vertical windows under Moon-Spencer sky conditions (uniformly overcast). Renewable Energy 33(1):168–172
Navvab M, Karayel M, Ne’eman E, Selkovitz S (1984) Analysis of atmospheric turbidity for daylight calculations. Energy Build 6(2–4):293–303
Parzen E (1962) On estimation of a probability density function and mode. Ann Math Stat 33(3):1065–1076
Quteishat A, Lim CP (2008) A modified fuzzy min-max neural network with rule extraction and its application to fault detection and classification. Appl Soft Comput 8(2):985–995
Roisin B, Bodart M, Deneyer A, Herdt PD (2008) Lighting energy savings in offices using different control systems and their real consumption. Energy Build 40(4):514–523
Rossini EG, Krenzinger A (2007) Maps of sky relative radiance and luminance distributions acquired with a monochromatic CCD camera. Sol Energy 81(11):1323–1332
Soler A, Robledo L (2005) Investigation of the overcast skies luminance distribution using 35 sensors fixed on a dome. Energy Convers Manag 46(17):2739–2747
Specht D (1990) Probabilistic neural networks. Neural Netw 3(1):109–118
Tregenza PR (1987) Subdivision of the sky hemisphere for luminance measurements. Light Res Technol 19(1):13–14
Tregenza PR (2004) Analysing sky luminance scans to obtain frequency distributions of CIE standard general skies. Light Res Technol 36(4):271–281
Tregenza PR, Sharples S (1995) New daylight algorithm. University of Sheffield, Sheffield
Ward System Group (2000) NeuroShell 2—release 4.0. Ward System Group, Frederick
Wittkopf SK, Soon LK (2007) Analysing sky luminance scans and predicting frequent sky patterns in Singapore. Light Res Technol 39(1):31–51
Younes S, Muneer T (2007) Clear-sky classification procedures and models using a world-wide data-base. Appl Energy 84(6):623–645
Acknowledgments
The work described in this paper was fully supported by a grant from the City University of Hong Kong (Project no. 7002284), and K. L. Cheung is supported by a City University of Hong Kong studentship.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, D.H.W., Tang, H.L., Cheung, K.L. et al. Sensitivity analysis of climatic parameters for sky classification. Theor Appl Climatol 105, 297–309 (2011). https://doi.org/10.1007/s00704-010-0392-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00704-010-0392-6