measurement and comparison of solar radiation estimation models ...

More documents

Recommendations

Info

horizontal surface by the DMI. The yearly and monthly optimum tilt angles were determined by converting available data. It has been concluded that the yearly total optimal tilt angle was 39.40° for the year 2000. The smallest optimal tilt angle was 6.7° in June and the largest was 65.2° in November. Ulgen and Hepbasli (2004) reviewed solar radiation models for Turkey in general and some of its provinces. 41 models used to estimate the monthly average daily global solar radiation on a horizontal surface were categorized in four groups namely; lineer models, polynomial or quadratic models, angular models, and modified Angstrom-type models. They concluded that most of solar radiation models developed for Turkey over a 19 year period was in the polynomial forms and the models gave reasonably well results for Turkey or elsewhere with similar climatic conditions. Aras et al. (2006) developed empirical models to predict the monthly-average daily global solar radiation over twelve provinces in the Central Anatolia Region (CAR) of Turkey and to compare calculated values obtained from developed models with data measured by the Turkish State Meteorological Service (DMI) in the period from January 1990 to December 1996 based on the various statistical methods. The general linear, quadratic, and cubic polynomial models were derived for the region. Values for the maximum monthly average daily measured global solar radiation ranged from 24 (for Cankiri, Turkey) and 27.10 (for Nevsehir, Turkey) to 25.85 MJ/m2. Maximum average daily measured sunshine duration values were in the range of 9.6 (for Cankiri, Turkey) and 11.90 (for Nevsehir and Sivas, Turkey) to 11.12 h. Ulgen and Hepbasli (2004) compared some existing models used for estimating the monthly-average daily global solar radiation for Istanbul, Ankara and Izmir of Turkey. They also developed some empirical models for these cities. The MBE, RSME, MPE, and t-statistic methods were used to evaluate the performance of the models. They concluded that the 2 new empirical models were found to be reasonably good for all the test methods. A new correlation for the estimation of monthly average daily global solar radiation was developed by Akinoglu and Ecevit (1990) and they compared with the correlations of Rietveld, Benson et al., Ogelman et al. The overall results show that the quadratic form gives better performance in terms of global applicability. The new quadratic model should be preferred for the monthly average global solar radiation estimation when the data for bright sunshine hours are available. 11
Togrul et al. (2002) developed some statistical relations in order to estimate monthly mean daily global solar radiation in Turkey. The global solar radiation was measured by Kipp pyranometer in 6 observation stations. Various regression analyses were applied to estimate monthly mean solar radiation in Turkey, by using the daily fraction of possible hours, (n/N), where n is the measured sunshine duration in a day and N is the theoretical day length of that day. Three statistical tests, root mean square error (RMSE) and mean bias error (MBE), and t-statistic were used to evaluate the accuracy of the correlations. It was seen that the equations which include the summer and winter periods gave better results than the others in all of the developed equations. 2.2. Diffuse Solar Radiation Ulgen and Hepbasli (2001) showed the monthly-average global and diffuse solar radiation data. The theoretical analysis of the monthly-average clearness index was defined. As a result, the value of the monthly-average global radiation varied from 5964 kJ/m² in December to 27.154kJ/m² in June. The values of the monthly-average daily clearness index raged from 0.45 to 0.66. The developed models were found to be suitable and reliable. Ulgen and Hepbasli (2002) used hourly global and diffuse radiation measurements over a 5 year period to establish a relationship between the daily diffuse fraction and the daily clearness index for Izmir, Turkey. The comparisons of the results were done with other correlations available in the literature. After generating the results from 16 models, the best results were obtained from the higher order polynomial models. Aras et al. (2005) developed 12 new models for estimating monthly average daily diffuse solar radiation on a horizontal surface in the CAR and the diffuse solar radiation models in the literature were analyzed in detail. In conclusion, the provinces in the CAR have almost the same diffuse solar radiation values. Tarhan and Sari (2004) analyzed global and diffuse solar radiation in 5 cities in the CBSR. A quadratic polynomial equation was empirically developed to predict the monthly average daily global radiation. A hybrid model was also developed based on the predictions of six existing models. As a result, the quadratic model has been chosen as the best model for the solar radiation data of the CBSR region. 12
Page 1 and 2: MEASUREMENT AND COMPARISON OF SOLAR
Page 3 and 4: ACKNOWLEDGEMENTS First of all, the
Page 5 and 6: ÖZET İZMİR/TÜRKİYE İÇİN GÜ
Page 7 and 8: 4.8. Ulgen and Hepbasli’s Model
Page 9 and 10: Table LIST OF TABLES Table 1. Turke
Page 11 and 12: DMI EIEI IYTE MBE RMSE e t-stat MPE
Page 13 and 14: development. Globally, an estimated
Page 15 and 16: eceived directly (direct radiation)
Page 17 and 18: According to these data, Turkey's a
Page 19 and 20: 2.1. Global Solar Radiation CHAPTER
Page 21: Electrical Works Research Directora
Page 25 and 26: so widely throughout China that the
Page 27 and 28: Figure 2. Approximate construction
Page 29 and 30: Meteorological data including tempe
Page 31 and 32: CHAPTER 4 ANALYSIS OF GLOBAL RADIAT
Page 33 and 34: The equation was in the Group 2. H
Page 35 and 36: CHAPTER 5 ANALYSIS OF DIFFUSE RADIA
Page 37 and 38: 5.3. 퐇 퐝 as a Function of 퐇
Page 39 and 40: CHAPTER 6 RESULTS 6.1. Results for
Page 41 and 42: Table 5. Hₒ values for Izmir Mont
Page 43 and 44: 1000 900 800 700 600 500 400 300 20
Page 45 and 46: According to Table 9 the maximum va
Page 47 and 48: 36 Table 10. Hourly average global
Page 49 and 50: Evaluation of Group 1: There is onl
Page 51 and 52: 2000 1800 1600 1400 1200 1000 800 6
Page 53 and 54: The following main results were obt
Page 55 and 56: CHAPTER 7 CONCLUSION In this thesis
Page 57 and 58: Haydar A.; Balli O.; Hepbasli A. (2
Page 59 and 60: Yildiz, M. and Oz, S. (1994), Evalu
Page 61 and 62: 700 600 500 400 300 200 100 0 900 8
Page 63 and 64: 800 700 600 500 400 300 200 100 0 6

measurement and comparison of solar radiation estimation models ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?