A Study on Solar Radiation Calculation for Istanbul with Measured Data İ. Ekmekci1., Y. Şen2
1Department of Industrial Engineering, İstanbul Commerce University, İstanbul, Turkey;
ismail.ekmekci@gmail.com
2Duzce University, Engineering Faculty, Duzce, Turkey
ABSTRACT: In this study, horizontal global solar radiation estimation models and estimated values are reviewed and compared with measured national database for Istanbul. The mathematical models used are based on the sunshine duration data as common in the world. For comparisons, the most common statistical methods (MBE, MPE and RMSE) are also used. It can be briefly said that solar energy measurements in our country are not perfect and also are not reliable for any feasibility study then the results show that measurements cannot be used directly for any analysis. Those measurements also include wrong and missing data parts. Investors should have their own measurement station at the best location for the planned action, and the measured data also should be analyzed by specialists.
KEYWORDS: Solar Energy, Solar radiaton Models, Istanbul, GEPA, Angström Model INTRODUCTION
In solar energy studies, solar radiation and its components for a specific location are the main necessities. The solar data should be sufficient, reliable and contemporary. Solar energy data measurement is a difficult process: measurement devices need to calibration and maintenance continuously. Due to the difficulties in radiation measurements, scientists are developed many mathematical models to estimate the solar energy data. Menges et al. studied the 50 models [1] and Bakirci reviewed 60 models [2] for solar radiation estimation. Radiation can also be estimated by satellite images. Turkey has a high potential on solar energy. The yearly average solar radiation is about 4.18 kWh/(m2.day), and the daily average sunshine duration is about 7.50 h [3]. This potential is one of the best of Europe.
Authority (EIE - General Directorate of Electrical Power Resources Survey and Development Administration) calculates that Turkey can produce 380 GWh electricity per year on 4600 m2 area which has a potential approximately of 4.52 kWh/(m2.day). In this
study, we tried to evaluate the twelve solar radiation estimation models available in the literature for Istanbul. Models are based on different regression methods that use the sunshine duration data. Two different solar radiation and sunshine duration data are used for analysis: first one is ten years meteorological data taken from State Meteorological Affairs, known as DMI in Turkey, and second one is database of Solar Energy Potential Atlas, known as GEPA in Turkey. Statistical methods are used for comparisons of the results and selection of the best models.
METHODS AND MATERIALS
Solar Radiation Estimation Models and Data Data and Methods of Comparison
In this study, the hourly measured irradiation and sunshine duration data on horizontal surface between 1997 and 2006 of Istanbul are used to calculate the contemporary average values.
Basic data of the meteorological station are
presented in Table 1.
Table.1. Basic information for local meteorological station
City Station Elevation
(m) Latitude Longitude
Average Temperature (°C)a
Average Wind Speed (m/s) b
Average Total Radiation (cal/cm2.day)a
Average Sunshine Duration
(h) a
Istan bul
Gozte
pe 33 40.97’N 29.8’E 14.2 2.8 307 5.9
Monthly averages of two databases are presented in Table 2. As seen from the Table, there are high differences between two databases. These differences are not scope of this study and are examined in another study [4].
The performances of the models were evaluated by the statistical error analysis methods: mean bias error (MBE), mean percentage error (MPE) and root mean square error (RMSE).
Table 2. Monthly average daily radiation and sunshine data
TYM a GEPA b
Horizontal global solar radiation
Sunsh ine durati on
Horizontal global solar radiation
Sunsh ine durati on
kWh/m2.day H kWh/m2.day h
JANUARY 1,42 2,18 2,00 3,46
FEBRUARY 2,28 3,70 2,57 4,43
MARCH 3,46 5,00 4,20 5,32
APRIL 4,46 5,85 5,28 6,85
MAY 5,90 8,38 6,30 8,61
JUNE 6,49 10,05 6,79 10,5
JULY 6,49 11,06 6,79 1 11,1
AUGUST 5,38 9,44 6,07 7 10,1
SEPTEMBE 4 R
4,33 7,56 5,09 7,83
OCTOBER 2,88 5,23 3,74 5,22
NOVEMBE R
1,77 3,29 2,37 3,85
DECEMBE R
1,24 2,02 1,80 2,96
MEAN 3,84 6,15 4,42 6,70
a: Ten years database monthly average (TYM) b: Solar energy potential atlas database (GEPA)
These are the most common statistical methods in solar radiation estimation models comparisons [1, 5, 6, 7 and 8]. MBE, MPE and RMSE are formulas are given below:
(1)
(2) (3) where N is the total number of observations, Hi,m is the ith measured value and Hi,c is the ith calculated value.
BASICS OF ESTIMATION MODELS The simplest and most known model developed for estimation of monthly average daily global solar radiation on horizontal surface is the Angström-type regression model. After this model, dozens of models developed on this base.
Today, the basic model is based on the fraction of the measured and clear day monthly average sunshine hours at the location in question [9]:
(4) where H is the monthly average daily global solar radiation, H0 is the monthly average daily extraterrestrial solar radiation, S is the monthly average daily bright sunshine hours, S0 is the monthly average day length, and a and b are empirical coefficients. The monthly average extraterrestrial daily radiation on a horizontal surface, H0, can be calculated by the following equation [9]:
(5) where Isc is the solar constant (Isc=1353 W/m2), f is the eccentricity correction factor, φ is the latitude of the location, δ is the solar declination angle, ws is the mean monthly sunrise hour angle. The eccentricity correction factor, solar declination and sunrise hour angle can be computed by the following Equations (6)-(7)-(8), respectively [9]:
(6) (7) (8) where D is the number of the days of the year starting from the first of January. The clear day
maximum possible sunshine duration) can be calculated by the following equation:
(9) Model 1:
Bakirci proposed a linear exponential model for Istanbul [7]:
(10 ) Model 2:
Bakirci also suggested the following exponential equation for Istanbul [7]:
(11) Model 3:
Ulgen and Hepbasli proposed the following linear equation for Ankara, Istanbul and Izmir [6]:
(12) Model 4:
Ulgen and Hepbasli also suggested the following third order polynomial equation for Ankara, Istanbul and Izmir [10]:
(13)
( 1 3 )
Model 5:
Tiris et al. obtained the following correlation from the experimental data measured in Gebze, Kocaeli [11]:
(14) Model 6:
Yildiz and Oz proposed the following equation for Turkey [12]:
(15) Model 7:
Tasdemiroglu and Server developed the following correlation for the six locations (
Ankara, Antalya, Diyarbakir, Gebze, Izmir and Samsun) in Turkey [13]:
(16 ) Model 8:
Akinoglu and Ecevit proposed the following equation for Turkey [14]:
(17) Model 9:
Kilic and Ozturk have determined the regression coefficients a and b as a function of the solar declination angle, latitude and elevation (km) [15]:
(18a) (18b) b= 0.533 – 0.165 Cos(φ-δ) (18c)
Model 10:
Glover and McCulloch suggested the following model that depends on the latitude [16]:
(19) Model 11:
Page proposed the following equation [17]:
(20) Model 12:
Gopinathan has derived following correlation [18]:
(21)
RESULTS AND DISCUSSION
Twelve mathematical models have used for estimation of the monthly average daily global solar radiation data. Validation of the models has been performed by using the basic statistical methods (MBE, MPE and RMSE). As mentioned in the previous section we have used two different databases. So we have two different global solar radiation and sunshine duration data. Therefore we have made four different evaluations:
1. By using TYM sunshine duration data, we estimated the solar radiation data and compare with
a. TYM solar radiation data, b. GEPA solar radiation data.
2. By using GEPA sunshine duration data, we estimated the solar radiation data and compared with
a. TYM solar radiation data, b. GEPA solar radiation data.
According to the results, best models are defined as Model 10 for 1.a, Model 6 for 1.b, Model 1 for 2.a and again Model 1 for 2.b. Results are
presented in Figure 1 and Table 3. In this study, it can be briefly said that solar energy measurements in our country are not perfect and also are reliable for any feasibility study then the results show that measurements cannot be used directly for any analysis. Those measurements also include wrong and missing data parts. The solar radiometry science has some developments to increase the measurement quality and reliability [19]. However, today radiometric devices are not also reliable.
Therefore researchers should notice the uncertainty level of measurements and make the energy calculations concerning with the inaccuracies. The solar energy investments highly
depend on the radiometric measurements. For this kind of energy investments, which are so expensive today, the feasibility studies should be done by professional persons. In this study we used the measured data from DMI station and GEPA database. Investors should have their own measurement station at the best location for the planned action, and the measured data also should be analyzed by specialists.
Table 3. Comparison of the results
Fig. 1. Comparisons of the results
ACKNOWLEDGEMENTS
The Turkish State Meteorological Service (DMI) is acknowledged for the supply of data.
REFERENCES
[1]. Menges H. O., Ertekin C. and Sonmete M.
H., Evaluation of global solar radiation models for Konya, Turkey, Energy Conversion and Management 47 (18–19) (2006), pp. 3149–3173.
[2]. Bakirci K., Models of Solar Radiation with Hours of Bright Sunshine: A Review, Renewable and Sustainable Energy Reviews 13, (2009), pp. 2580–2588.
[3]. EIE, www.eie.gov.tr
[4]. Uzman Enerji, www.uzmanenerji.com [5]. Ulgen K, Hepbasli A. Comparison of diffuse fraction of daily and monthly global radiation for Izmir, Turkey. Energy Source 2003; 25:637–
49.
[6]. Ulgen K, Hepbasli A. Solar radiation models. Part 2: Comparison and developing new models. Energy Source 2004; 26:521–30.
[7]. Bakirci K., Correlations for estimation of daily global solar radiation with hours of bright sunshine in Turkey, Energy 34 (4) (2009), pp.
485–501.
[8]. Bulut H., Büyükalaca O., Simple model for the generation of daily global solar-radiation data in Turkey, Appl Energy 84 (2007), pp. 477–
491.
[9]. Duffie JA, Beckman WA. Solar engineering of thermal processes. New York: Wiley; 1991.
[10]. Ulgen K, Hepbasli A. Comparison of solar radiation correlations for Izmir, Turkey.
International Journal of Energy Research 2002;
26:413–30.
[11]. Tiris M, Tiris C, Ture IE. Correlations of monthly-average daily global, diffuse and beam radiations with hours of bright sunshine in
Gebze, Turkey. Energy Conversion and Management 1996; 37:1417– 21.
[12]. Yildiz M, Oz S. Evaluation of the solar energy potential of Turkey. In: Proceedings of the 6th National Energy Congress; 1994. p. 250–
60 [in Turkish].
[13]. Tasdemiroglu E, Sever R. An improved correlation for estimating solar radiation from bright sunshine data for Turkey. Energy Conversion and Management 1991; 31(6):599–
600.
[14]. Akinoglu BG, Ecevit A. A further comparison and discussion of sunshine based models to estimate global solar radiation. Solar Energy 1990; 15:865–72.
[15]. Dogniaux R, Lemoine M. Classification of radiation sites in terms of different indices of atmospheric transparency. Solar energy research and developmentin the European Community, Series F, vol. 2. Dordrecht, Holland: Reidel;
1983.
[16]. Glover J, McGulloch JDG. The empirical relation between solar radiation and hours of sunshine. Quarterly Journal of the Royal Meterological Society 1958; 84:172–5.
[17]. Page JK. The estimation of monthly mean values of daily total short wave radiation on vertical and inclined surface from sunshine records for latitudes 40N–40S. Proceedings of UN Conference on New Sources of Energy 1961; 4(598):378–90.
[18]. Gopinathan KK. A general formula for computing the coefficients of the correlations connecting global solar radiation to sunshine duration. Solar Energy 1988; 41:499–502.
[19]. Gueymard C. A., Myers D. R., Modeling Solar Radiation at the Earth’s Surface, Springer, 2008
