THEORETICAL ANALYSIS AND EXPERIMENTAL VERIFICATION OF PARABOLIC TROUGH SOLAR COLLECTOR WITH HOT WATER GENERATION SYSTEM

THEORETICAL ANALYSIS AND EXPERIMENTAL VERIFICATION OF PARABOLIC TROUGH SOLAR COLLECTOR WITH HOT WATER GENERATION SYSTEM by Amirtham VALAN ARASU and Samuel Thambu SORNAKUMAR Original scientific paper UDC: 662.997:519.876.5:697.132 BIBLID: 0354-9836, 11 (2007), 1, 119-126 The mod el ing of a par a bolic trough col lec tor with hot wa ter gen er a tion sys - tem with a well-mixed type stor age tank us ing a com puter sim u la tion pro - gram is pre sented in this pa per. This is fol lowed by an ex per i men tal ver i fi - ca tion of the model and an analysis of the experimental results. The max i mum dif fer ence be tween the pre dicted and the ac tual stor age tank wa - ter tem per a ture val ues is found as 9.59% only. This vari a tion is due to the dif fer ence be tween the ac tual weather dur ing the test pe riod com pared to hourly val ues and the con vec tion losses from the col lec tor re ceiver, which were not con stant as ac counted by the com puter sim u la tion pro gram. Key words: par a bolic trough collector, storage tank, hot water generation, sys tem performance, simulation program Introduction In dia is gen er ally en dowed with re new able en ergy re sources: so lar, wind, and bio mass. In dia re ceives so lar en ergy equiv a lent to over 5000 tril lion kwh/year, which is far more than the to tal en ergy con sump tion of the coun try. The daily av er age global ra di - a tion is around 5 kwh per sq. m. per day with the sun shine hours rang ing be tween 2300 and 3200 per year [1]. In dia to day has among the world s larg est pro grams in so lar en - ergy. The In dian Min is try of non-con ven tional en ergy sources is im ple ment ing coun try - wide pro grams on: (1) So lar Ther mal Pro gram and (2) So lar Photovoltaics Pro gram. So - lar ther mal tech nol o gies uti lize the heat en ergy from the sun for var i ous pur poses. De pend ing on the tech nol ogy, the tem per a ture of the out put ther mal en ergy can vary from as low as am bi ent tem per a ture to as high as 3000 C. This opens up a vast area of ap - pli ca tions in clud ing power gen er a tion and re frig er a tion. So lar wa ter heat ing is an im por - tant ap pli ca tion of so lar en ergy. From many types of so lar col lec tors de vel oped, three types merit fur ther con sid er ation for hot wa ter gen er a tion: the flat plate col lec tor (FPC) [2], the com pound par a bolic col lec tor (CPC) [3], and the par a bolic trough col lec tor (PTC) [4]. The first two are sta tion ary col lec tors, whereas the last one is a tack ing col lec - tor. PTCs are gen er ally of me dium con cen tra tion ra tio (15-40) and hence they re quire some level of so lar track ing. The ad van tage of con cen trat ing col lec tors is that the heat DOI:10.2298/TSCI0701119V 119

THERMAL SCIENCE: Vol. 11 (2007), No. 1, pp. 119-126 losses are in versely pro por tional to the con cen tra tion ra tio. The most im por tant ad van - tage of PTC as com pared with the other two types of col lec tors is its abil ity to func tion at high tem per a tures with high ef fi ciency. For ex am ple, at a tem per a ture of 100 C, PTCs works at an ef fi ciency of about 62%, CPCs at about 32% and the FPC at about 10% [5]. In re cent days, PTC has been used many ap pli ca tions such as steam gen er a tion for in dus trial ap pli ca tions [6] and power gen er a tion [7, 8], hot wa ter gen er a tion, sea wa ter de sa li na tion [9, 10], so lar photocatalysis [11], so lar de tox i fi ca tion of or ganic pol lut ants [12], etc. The ob jec tive of the mod el ing pre sented in this pa per is to in ves ti gate the per for mance of a PTC hot wa ter gen er a tion sys tem and com pare the re sults with the ac tual sys tem per for - mance. Modeling of the PTC hot water system In the pres ent work, a new par a bolic trough col lec tor sys tem, which has been de - vel oped for hot wa ter gen er a tion, is pre sented in fig. 1. The PTC sys tem for hot wa ter gen er a tion in cludes a PTC, a hot wa ter stor age tank (HWST) of well-mixed type and a cir cu lat ing pump. The pa rab ola of the pres ent col lec tor with a rim an gle of 90 is very ac - cu rately con structed of fiber glass. A flex i ble so lar re flec tor ma te rial (SOLARFLEX foil) from Clear Dome So lar, San Diego, Cal., USA [13] with a reflectance of 0.974 is used in the pres ent work. The so lar re ceiver con sists of a cop per tube, a glass en ve lope, and rub - ber cork seals at both ends of the glass en ve lope. The cop per tube is coated with a heat re - sis tant black paint and is sur rounded by a con cen tric glass cover with an an nu lar gap of 0.5 cm. The rub ber corks are in cor po rated to achieve an air-tight en clo sure. Wa ter from Figure 1. Parabolic trough collector with storage tank 120

Valan Arasu, A., Sornakumar, S. T.: Theoretical Analysis and Experimental Verification of... the stor age tank is pumped through cop per tube, where it is heated and then flows back into the stor age tank. The PTC ro tates around the hor i zon tal north/south axis to track the sun as it moves through the sky dur ing the day. The axis of ro ta tion is lo cated at the fo cal axis. The track ing mech a nism con sists of a low speed 12 V D. C. mo tor and an em bed ded elec tronic con trol sys tem. The in put sig nals to the con trol sys tem are ob tained from light de pend ent re sis tors. The pump for main tain ing the forced cir cu la tion is op er ated by an on-off con trol ler (dif fer en tial ther mo stat), which senses the dif fer ence be tween the tem - per a ture of the wa ter at the out let of the col lec tor (T fo ) and the stor age tank wa ter (T l ). The pump is switched on when ever this dif fer ence ex ceeds a cer tain value and off when it falls be low a cer tain value. In the pres ent work, the dif fer en tial tem per a ture con trol ler value (T fo T l ) is set as +2 C. It is as sumed that the wa ter in the stor age tank is al ways well mixed and con se quently is at a uni form tem per a ture, T l, which var ies only with time. The spec i fi ca tions of the PTC sys tem are de tailed in tab. 1. Table 1. Parabolic trough collector system specifications Items Value Collector aperture 0.8 m Collector length 1.25 m Rim angle 90 o Focal distance 0.2 m Receiver diameter 12.8 mm Glass envelope diameter 22.6 mm Concentration ratio 19.89 Water flow rate 1.0 l/min. Storage tank capacity 35 li ters Tank material Stain less steel Tank insulation material Glass wool Insulation thickness 5 cm Water pump 367.65 W Simulation program In the pres ent work, a sim u la tion model based on MATLAB pro gram is de vel - oped to de ter mine the op ti cal and ther mal char ac ter is tics of a par a bolic trough col lec tor and the hourly stor age tank wa ter tem per a ture. De tailed anal y sis of the col lec tor per for - 121

THERMAL SCIENCE: Vol. 11 (2007), No. 1, pp. 119-126 mance based on the col lec tor per for mance sim u la tion pro gram is pre sented in the pa per [14]. The pro gram gets the re quired in put such as sys tem spec i fi ca tions given in tab. 1, hourly so lar beam ra di a tion, am bi ent tem per a ture, and ini tial stor age tank tem per a ture and de ter mines the ther mal and op ti cal char ac ter is tics of the PTC. Fi nally, the last out put from the pro gram is that the new stor age tank wa ter tem per a ture af ter ev ery one hour time in ter val is de ter mined. This new stor age tank tem per a ture is taken as stor age tank liq uid ini tial tem per a ture, T li at the next hour. Thus the pro gram com putes the hourly stor age tank wa ter tem per a ture. It is as sumed that no hot wa ter is used. The pro gram flow chart is shown in fig. 2. Figure 2. Simulation program flow chart 122

Valan Arasu, A., Sornakumar, S. T.: Theoretical Analysis and Experimental Verification of... Experimentation The ex per i men tal set up used in the pres ent work for as sess ing the ac cu racy with which the sim u la tion pro gram pre dicts the PTC hot wa ter gen er a tion sys tem per for mance is as shown in fig. 1. The ther mal per for mance of the par a bolic trough so lar col lec tor ac - cord ing to ASHRAE stan dard 93, 1986 [15] and the hourly stor age tank wa ter tem per a - ture were de ter mined. The PTC hot wa ter sys tem per for mance is de ter mined for dif fer ent com bi na tions of in ci dent ra di a tion, am bi ent tem per a ture and in let wa ter tem per a ture. The so lar beam ra di a tion in ten sity (I) was mea sured by a pyrheliometer and the mass flow rate of wa ter by a rotameter. The wind speed was mea sured by a vane type an e mom - e ter. The col lec tor wa ter out let tem per a ture (T fo ), am bi ent tem per a ture (T a ), and stor age tank wa ter tem per a ture (T l ) were re corded with the help of PT 100 re sis tance tem per a - ture de vice (RTD) sen sors. All pa ram e ters were mea sured un der steady-state or quasi- -steady-state con di tions as a func tion of time i. e. for ev ery 15 min utes pe riod from 10.00 am to 16.00 pm over six months from March to Au gust. The ther mal ef fi ciency of a con cen trat ing col lec tor op er at ing un der steady state con di tions can be de scribed by ASHRAE 1986 [15]: h F h R o F U C T T I R L fi a (1) The ther mal ef fi ciency from eq. (1) is plot ted against (T fi T a )/I, a straight line will re sult pro vided U L is con stant. The in ter cept is F R h o and the slope is F R U L /C. The o ret i cal cal cu la tion of the in ter cept and slope us ing the sim u la tion pro gram [14] to gether with the ex per i men tal val ues are pre sented in tab. 2. It can be ob served from tab. 2 that there is a mi nor dif fer ence be tween the the o ret i cal and the ex per i men tal re sults with re spect to the test slope (5.92%) and a mod er ate dif fer ence is shown for the test in - ter cept (9.37%) Table 2. Collector performance No. Item Test intercept Test slope 1 Theoretical performance 0.7552 0.3636 2 Experimental performance (present work) 0.6905 0.3865 3 Difference [%] 9.37 5.92 For the es ti ma tion of hourly stor age tank wa ter tem per a ture, the fol low ing equa - tion is used [16]: T l new T l old Q u Ql ( UA) t ( Tl old Ta ) ( rvc ) ( rvc ) p l p t (2) 123

THERMAL SCIENCE: Vol. 11 (2007), No. 1, pp. 119-126 where (UA) t the product of the overall heat transfer coefficient between the tank water and the ambient air and surface area of the tank [W] and (rvc p ) l represents the heat capacity of the water in the tank [J], the heat capacity of the tank material [J]. In the present work, Q l is taken as zero and Dt =1 h. The re sults from the ex per i ments car ried out with the newly de vel oped PTC sys - tem are pre sented in tab. 3. As can be seen from this ta ble, the dif fer ence be tween the pre - dicted (T l sim ) and the ac tual stor age tank tem per a ture (T l exp ) for all the six months is rea - son ably small. Table 3. Comparison of predicted and actual storage tank temperature Day March 10, April 17, May 15, June 13, July 10, August 16, Tank water Tem per a ture, [ C] 11.00 h 12.00 h 13.00 h 14.00 h 15.00 h 16.00 h T l sim 45.32 53.55 61.25 67.51 72.34 77.48 T l exp 43.65 53.28 60.41 65.83 70.34 74.66 Difference [%] 3.83 0.51 1.39 2.55 3.22 3.78 T l sim 50.41 58.50 65.93 72.52 77.90 83.52 T l exp 49.77 57.94 65.53 71.19 74.81 77.67 Difference [%] 1.28 0.97 0.61 1.87 4.13 7.53 T l sim 49.83 57.43 64.42 70.60 75.65 80.89 T l exp 47.65 55.84 62.24 67.61 72.35 76.31 Difference [%] 4.58 2.85 3.50 4.42 4.56 6.00 T l sim 46.26 54.51 62.33 68.69 74.44 79.43 T l exp 42.21 51.32 59.29 66.42 72.83 78.45 Difference [%] 9.59 6.21 5.13 3.42 2.21 1.25 T l sim 43.43 50.85 57.69 63.76 68.72 73.87 T l exp 41.06 49.87 57.01 62.51 67.02 70.68 Difference [%] 5.77 1.97 1.19 1.99 2.54 4.51 T l sim 47.01 56.34 64.81 69.72 74.01 79.53 T l exp 45.22 55.83 62.88 68.13 71.97 75.15 Difference [%] 3.96 0.91 3.07 2.33 2.83 5.83 The dif fer ence be tween the the o ret i cal and the ex per i men tal stor age tank tem - per a ture val ues may be ac counted for by the fol low ing rea sons: the magnitude of solar beam radiation varies significantly and that of the ambient air temperature varies slightly with in one hour period. But those values are taken as constant during one hour in the simulation program, and 124

Valan Arasu, A., Sornakumar, S. T.: Theoretical Analysis and Experimental Verification of... the wind velocity, one of the important parameters, determines the convective losses is considered as constant (0.1 m/s) in the simulation program. But there was light wind velocity variation in all test conditions. Conclusions The mod el ing of a par a bolic trough col lec tor hot wa ter gen er a tion sys tem with a well-mixed type stor age tank us ing a com puter sim u la tion pro gram is pre sented in this pa per. The sim u la tion pro gram pre dicts the newly de vel oped PTC hot wa ter sys tem per - for mance to an ac cu racy of 9.59% and there fore the model can be used for long term pre - dic tion of par a bolic trough col lec tor with hot wa ter gen er a tion sys tem per for mance. Nomenclature A collector aperture area, [m 2 ] C concentration ratio, [ ] c p specific heat capacity of water, [Jkg 1 K 1 ] F R heat removal factor, [ ] I beam or direct radiation, [Wm 2 ] Q l rate at which energy is being discharged to the load, [W] Q u rate of useful heat received from the collector, [W] T a ambient temperature, [K] T e glass envelope surface temperature, [K] T fi collector water inlet temperature, [K] T fo collector water outlet temperature, [K] T l storage tank water temperature, [K] T l i initial storage tank water temperature, [K] T l new new storage tank water temperature, [K] T l old old storage tank water temperature, [K] T r receiver tube surface temperature, [K] Dt time interval, [h] U L overall heat loss co-efficient, [Wm 2 K 1 ] V volume, [m 3 ] Greec letters h thermal efficiency of the collector, [%] h o optical efficiency, [%] r density, [kgm 3 ] References [1] ***, Sur vey of Re new able En ergy in In dia, TERI Pro ject Re port No. 2000RT45, Tata En ergy Re search In sti tute, New Delhi, 2001, pp. 1-50 [2] Mor ri son, G. L., Braun, J. E., Sys tem Mod el ing and Op er a tion Char ac ter is tics of the Thermo - siphon So lar Wa ter Heater, Jour nal of So lar En ergy, 34 (1985), 4-5, pp. 389-399 125

THERMAL SCIENCE: Vol. 11 (2007), No. 1, pp. 119-126 [3] Oommen, R., Jayaraman, S., De vel op ment and Per for mance Anal y sis of Com pound Par a - bolic Solar Concentrators with Reduced Gap Losses-Oversized Reflector, En ergy Con ver - sion and Man age ment, 42 (2001), 11, pp. 1379-1399 [4] Kalogirou, S., Lloyd, S., Use of So lar Par a bolic Trough Col lec tors for Hot Wa ter Pro duc tion in Cy prus A Fea si bil ity Study, Re new able En ergy, 2 (1992), 2, pp. 117-124 [5] Kalogirou, S., Use of Par a bolic Trough So lar En ergy Col lec tors for Sea-Wa ter De sa li na tion, Ap plied En ergy, 60 (1998), 2, pp. 65-88 [6] May, E. K., Murphy, L. M., Per for mance Ben e fits of the Di rect Gen er a tion of Steam in Line-Fo cus So lar Col lec tors, Jour nal of So lar En ergy En gi neer ing, 105 (1983), pp. 126-133 [7] Zarza, E., Valenzuela, L., Leon, J., Weyers, D., Eickhoff, M., Eck, M., The DISS Pro ject: Di - rect Steam Generation in Parabolic Trough Systems. Operation and Maintenance Experience and Up date on Pro ject Sta tus, Jour nal of So lar En ergy En gi neer ing, 124 (2002), 2, pp. 126-133 [8] Zarza, E., Valenzuela, L., Leon, J., Hennecke, K., Eck, M., Weyers, D., Eickhoff, M., Di rect Steam Gen er a tion in Par a bolic Troughs: Fi nal Re sults and Con clu sions of the DISS Pro ject, En ergy, 29 (2004), 5-6, pp. 635-644 [9] Garcia-Rodriguez, L., Gomez-Camacho, C., Design Parameter Selection for a Distillation Sys tem Cou pled to a So lar Par a bolic Trough Col lec tor, De sa li na tion, 122 (1999a), 2-3, pp. 195-204 [10] Garcia-Rodriguez, L., Gomez-Camacho, C., Thermoeconomic Anal y sis of a So lar Par a bolic Trough Col lec tor Dis til la tion Plant, De sa li na tion, 122 (1999b), 2-3, pp. 215-224 [11] Arancibia-Bulnes, C. A., Cuevas, S. A., Mod el ing of the Ra di a tion Field in a Par a bolic Trough So lar Photo Catalytic Re ac tor, So lar En ergy, 76 (2004), 5, pp. 615-622 [12] Vivek Babu, G., Sharat, A., Nagaraju, J., So lar Decolorization of Rhodamine B Dye Us ing Con cen trat ing Col lec tors, SESI Jour nal, 12 (2002), 2, pp. 73-80 [13] ***, ClearDome So lar Sys tems Heat ing and Cook ing Prod ucts, San Diego, Cal., USA, 2003, www.cleardomesolar.com [14] Valan Arasu, A., Sornakumar, T., De sign and Sim u la tion Anal y sis of a Par a bolic Trough So - lar Col lec tor Hot Wa ter Gen er a tion Sys tem, The International En ergy Jour nal, 6 (), 2, pp. 13-15 [15] ***, Method of Test ing to De ter mine the Ther mal Per for mance of So lar Col lec tors, ASHRAE Stan dard 93, Amer i can So ci ety of Heat ing, Re frig er at ing and Air-Con di tion ing En gi neers, At lanta, GA, USA, 1986 [16] Valan Arasu, A., Sornakumar, T., Performance Characteristics of Parabolic Trough Solar Collector System for Hot Water Generation, The In ter na tional En ergy Jour nal, 7 (2006), 2, pp. 137-145 [17] Tiwari, G. N., So lar En ergy Fun da men tals, De sign, Mod el ling and Ap pli ca tions, Narosa Pub lish ing House, New Delhi, 2002 Author's address: A. Valan Arasu, S. T. Sornakumar Faculty of Mechanical Engineering, Thiagarajar College of Engineering, Madurai, 625 015, India Corresponding author A. Valan Arasu E-mail: a_valanarasu@yahoo.com Paper submitted: December 12, Paper revised: January 18, 2006 Paper accepted: March 15, 2006 126