A step-by-step application of Sandia method in developing typical meteorological years for different locations in Oman

A step-by-step application of Sandia method in developing typical meteorological years for different locations in Oman
of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  INTERNATIONAL JOURNAL OF ENERGY RESEARCH Int. J. Energy Res.  2005;  29 :723–737Published online 1 April 2005 in Wiley InterScience ( DOI: 10.1002/er.1078 A step-by-step application of Sandia method in developingtypical meteorological years for different locations in Oman Naseem M. Sawaqed 1 , Yousef H. Zurigat 2, n , y and Hilal Al-Hinai 3 1 Mechanical Engineering Department, Mu ’ tah University, Jordan 2 Mechanical Engineering Department, University of Jordan, Amman, Jordan 3 Department of Mechanical and Industrial Engineering, Sultan Qaboos University, Oman SUMMARYThis paper reports on the development of typical meteorological years (TMYs) for seven different locationsin Oman based on measured meteorological data. Depending on the availability of data the TMYsdeveloped using Sandia method used data covering 7–17 years. The method as implemented here in a step-by-step procedure with illustrations is made simple. The procedure described herein is computerized andcan handle any number of data sets in an easy-to-use manner. This should facilitate the development of TMYs for any location where enough data is available. Sensitivity analysis of different weights assigned todifferent weather parameters shows that Sandia method is highly affected by solar flux even if its weight isreduced by half while the weights of other parameters such as temperature, wind, and relative humidityhave less impact on the selection of TMY. Copyright # 2005 John Wiley & Sons, Ltd. KEY WORDS : Sandia method; meteorology; TMY; energy systems; modelling; simulation 1. INTRODUCTIONThe typical meteorological year (TMY) for a given location is a data set of actual hourlymeteorological parameters for 1 year constructed from measured data of several past years.A TMY month is selected from the months of the period considered based on statistical analysisand the months so selected are assembled to form the typical year. TMY is used for comparativeperformance of alternative systems whose performance is dependent on weather conditions.Thus, the TMY is commonly used in building energy systems simulations and in assessments of wind and solar energy systems performance including PV systems. Also it is used in modelling of agricultural systems and the dispersion of air pollutants and in simulations of microenvironmentgreenhouses in agriculture and solar desalination systems. Several methods for generatingtypical data have been developed. These include the Synthesized Data Technique (Hall, 1978; Received 18 May 2004Accepted 2 September 2004 Copyright # 2005 John Wiley & Sons, Ltd. y E-mail: sponsor: Petroleum Development of Oman (PDO) n Correspondence to: Yousef H. Zurigat, Mechanical Engineering Department, University of Jordan, Amman, Jordan.  Siurna  et al  ., 1984), the Design Year (Klein  et al  ., 1976), the Weather Year for EnergyCalculations (WYEC) (Crow, 1981), and the TMY (Hall, 1978).The Synthesized Data Technique (Hall, 1978; Siurna  et al  ., 1984) is based on time-seriesanalysis. While this method is good for single variable it has been shown to be inapplicable forsynthesizing multi-variable meteorological data Siurna  et al  . (1984). The other methods citedabove are based on selecting typical short-term data from a multi-year data base. The DesignYear (Klein  et al  ., 1976) was developed for a 9-month heating season. Each of the 9 months wasselected from an 8-year data base using mean monthly insulation as a primary criterion andmean monthly temperature as a secondary one. The WYEC (Crow, 1981) consists of 12 monthseach being selected individually from a 30-year data base.Based on Sandia method (Hall  et al  ., 1978) developed by Sandia Laboratories, in U.S.A.,Pissimanis  et al  . (1988) used an available 17 years weather data base for the city of Athens. Theselection criteria were based on 13 weather parameters namely; dry bulb temperature (max, min,and mean), dew point (max, min, and mean), wind velocity (max and mean), and daily values of global radiation. Most recently, using the Sandia method Petrakis  et al  . (1998) and Kalogirou(2003) developed TMYs for the city of Nicosia, Cyprus. The study of Kalogirou (2003) includedadditional variables such as illuminance, visibility, preipitation, and snow fall data. Fordescription of and performance comparison between different methods for generating TMYsthe reader is referred to Argiriou  et al  . (1999).The literature review conducted in this work shows that Sandia method developed by SandiaNational Laboratories (Hall  et al  ., 1978) is the most widely used method. Although the methodwas published and described in several sources the procedure is not quite clear to simplify directimplementation. Furthermore, the characteristics of the method in terms of the dependence of the resulting TMY on the weights given to different weather parameters are not completelyaddressed. In this paper the Sandia method was used to develop TMYs for seven climatic regionsin Oman, namely: Seeb, Marmul, Salalah, Masirah, Majis, Sur, and Fahud (see Figure 1).The main aim of this paper is to report on the development of TMYs for seven differentlocations in Oman using Sandia method and to present method in a simplified manner based ona step-by-step procedure so that any interested reader can easily follow and apply. Sensitivityanalysis is also conducted to gage the effect of weights on the TMY selection. This work shouldaid in making the development of TMYs using this method easy and straight forward.2. DATA FORMATMeteorological data recorded at seven weather stations in Oman were obtained from theDirectorate General of Civil Aviation and Metrology (DGCAM) at Seeb Airport. These dataare processed and put by DGCAM in two files; hourly and daily data files. The hourly data donot contain hourly solar radiation data and hourly solar flux data were generated from the dailyglobal values using the methodology described in the next section.3. ESTIMATION OF HOURLY SOLAR FLUX DATA FROM DAILY GLOBALRADIATION DATAThe total hourly solar radiation falling on a horizontal surface,  I  th  in Wm  2 , could be estimatedfrom the daily global radiation in accordance with the methodology given by Duffie and Copyright # 2005 John Wiley & Sons, Ltd.  Int. J. Energy Res . 2005;  29 :723–737N. M. SAWAQED, Y. H. ZURIGAT AND H. AL-HINAI 724  Beckman (1982) as presented below: I  th  ¼  rH  3 : 6where  H   is the daily global radiation for the day and location under consideration(kJm  2 day  1 ) and  r  is the ratio of the total radiation for the hour to that of the whole dayand is given by (Duffie and Beckman, 1982) r  ¼  p 24 ð a  þ  b  cos ð W  ÞÞ  cos ð W  Þ   cos ð W  s Þ sin ð W  s Þ   p 180 W  s  cos ð W  s Þ 264375 where  a =0.4090+0.5016sin( W  s  60) and  b =0.6609–0.4767sin( W  s  60).Here,  W   is the hour angle of the sun (deg) given as: W   ¼  36024   ð h    12 Þ Figure 1. Map of Sultanate of Oman. Copyright # 2005 John Wiley & Sons, Ltd.  Int. J. Energy Res . 2005;  29 :723–737STEP-BY-STEP APPLICATION OF SANDIA METHOD  725  where  h  is the time of the day (solar time) and  W  s  is the sun sit hour angle (deg), given as: W  s  ¼  a  cos ð tan ð f Þ  tan ð d ÞÞ and, here  f  is the latitude of the location (deg) and  d  is the declination angle of the sun (deg)given by: d  ¼  23 : 45 sin 360 284  þ  day365   where day is the day of the year under consideration (1–365).The total solar radiation consists of the  direct  or beam radiation coming directly from the sunand the diffuse component scattered to the ground from the sky dome. The latter depends on theclarity of the sky and could be estimated from the correlation (Collares-Pereira and Rabl, 1979)which gives the daily average diffuse radiation,  H  d , as: H  d  ¼  H   0 : 775  þ 0 : 00653 ð W  s   90 Þ  ½ 0 : 505  þ 0 : 00455 ð W  s   90 Þ  cos ð 115 K  T     103 Þf g where  K  T   is the clearness index for the day, defined as the ratio of the daily radiation on ahorizontal surface to the daily extraterrestrial radiation ( H  o ), that is: K  T   ¼  H H  o and H  o  ¼  24   3600 I  sc p  1  þ  0 : 033 cos 360    day365    cos  f  cos  d  sin  W  s  þ  p W  s 180 sin  f  sin  d    where  I  sc  is the solar constant=1367Wm  2 .The hourly values of the diffuse solar radiation,  I  d ;  can be estimated from the equation by Luiand Jordan (1960) which gives: I  d  ¼  H  d 24 p cos  W    cos  W  s sin  W  s    p W  s 180 cos  W  s The direct component of the solar radiation on a horizontal surface,  I  bh  (beam horizontal) isthen obtained from: I  bh  ¼  I  th    I  d The hourly value of the direct solar radiation on a surface normal to the direction of the beam, I  bn  (important in the design of solar systems that track the sun), can be calculated from: I  bn  ¼  I  bh cos  y z where  y z  is the solar zenith angle calculated by Duffie and Beckman (1982) as:cos  y z  ¼  cos  f  cos  d  cos  W   þ sin  f  sin  d 4. TMY DEVELOPMENT PROCEDURE USING SANDIA METHODFor ease of presentation the procedure is applied here for one of the seven locations considered;namely Seebl with 17 years of relatively complete weather data. Missing data were replaced Copyright # 2005 John Wiley & Sons, Ltd.  Int. J. Energy Res . 2005;  29 :723–737N. M. SAWAQED, Y. H. ZURIGAT AND H. AL-HINAI 726  using interpolation techniques. Due to sun cycle it is recommended that 12 or 24 years be used.Four of the TMYs developed in this work were based on at least 12 years of data (Seeb, Sur,Salalah, and Majis) while the three remaining (Fahud, Marmul, and Masira) were based on, atleast, 7 years. With the tool developed in this work, more years can be incorporated as moredata become available. With the automatic data collection systems employed nowadays in mostmeteorological stations it is anticipated that in the future more complete data will be availableto allow for revisions of the developed TMYs.In Sandia method, individual months of the TMY are selected from different years of theavailable meteorological data. That is, if data record contains 15 years of data, all 15 Januarysare examined, and the one judged to be most typical is selected as the January of the TMY. Theother 11 months of the year are selected in a similar manner. The months selected may belong todifferent years and thus, adjacent months in the TMY may have discontinuities at the monthsinterfaces. As recommended by the method these are smoothed out for a total period of 12, 6hon each side.The procedural steps of the method are outlined below considering each month in a year at atime. Step  1: Construct the cumulative distribution functions (CDFs) for both the ‘ short-term ’and ‘ long-term ’ daily mean values for each of the parameters (elements or indices) considered.These parameters are the dry bulb temperature (mean, min, max, and range), relativehumidity (mean, min, max, and range), wind velocity (mean, max gust, and max sustainable)and global solar flux data. The daily mean values for a given month in a given year aretermed ‘ short-term ’ daily averages. When these are averaged over the years for each dayin a given month the resulting daily averages for that month are termed ‘ long-term ’ dailymeans. The CDF for a given parameter for a given month data is calculated as follows:let  n  be the number of days in a given month, then for the given parameter we have  n values in the month. Therefore, the probability that the parameter assumes any given dailyvalue is 1/ n . The first step in calculating the CDF is to sort the data in ascending order. Thenusing  j   as the rank index of the sorted data the CDF values are obtained using the followingformula:CDF  j   ¼  1 n j  ;  j   ¼  1 ; 2 ;  . . .  ; n For the month of January Figure 2 presents the short- and long-term CDFs for the mean dailytemperature parameter for a number of selected years for Seeb region. The presentation shownin Figure 2 is useful as it gives a quick view of the occurrence of certain values. For example, forJanuary, 1991 the mean temperature is above 22 8 C for about 60% of the time while for 2000 themean daily temperature exceeds 21 8 C for over 85% of the time. Of course, this presentationbecomes more informative when hourly data is used instead of daily data. Step  2: For each month of the year,  five candidate months  are selected from the correspondingmonths of the years considered in the development of the TMY. These candidate months arethose having the closest short- to the long-term CDFs over all parameters CDFs. Closeness of the short- to long-term CDF is based on the absolute difference between them.The selection of the five candidate months is done as follows:(a) For the month under consideration, calculate the Finkelstein–Schafer (FS) statistic(Finkelstein and Schafer, 1971) which represents the monthly average difference betweenthe short- and the long-term CDFs for index  j  ,  j   ¼  1 ; 2 ; 3 ;  . . .  ;  no. of indices considered  , Copyright # 2005 John Wiley & Sons, Ltd.  Int. J. Energy Res . 2005;  29 :723–737STEP-BY-STEP APPLICATION OF SANDIA METHOD  727
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks