IEA Solar Heating and Cooling Programm - NachhaltigWirtschaften.at

IEA Solar Heating and Cooling Programm - NachhaltigWirtschaften.at IEA Solar Heating and Cooling Programm - NachhaltigWirtschaften.at

nachhaltigwirtschaften.at
from nachhaltigwirtschaften.at More from this publisher
04.11.2013 Views

IEA SHC Task 38 Solar Air Conditioning and Refrigeration Subtask C2-A, November 9, 2009 By comparing the calculated and the measured outlet temperatures for the atypical conditions day, when the overall solar radiation fluctuated constantly, it can be seen that the model very accurately predicts the outlet temperature for the whole day, whatever the amplitude of the solar radiation fluctuations and the maximum error is always below 2°C but the model does not show any time delay with respect to the measurments. In these first two cases, the model correctly predicts the collector outlet temperature under different radiation conditions but for storage load. The next step in the validation procedure is to study the model’s response to a desiccant cooling load, i.e. storage in the morning, with regeneration in the afternoon. Day 3: 1000 900 Solar global radiation -jour 3- 90 Outlet temperature -Day 3- Experimental Model 800 Radiation [W.m -2 ] 700 600 500 400 300 Temperature [ C] 80 70 200 100 0 0 100 200 300 400 500 Time [min] 60 0 100 200 300 400 500 Time [min] Figure 12: comparison of the predicted outlet temperature of the collectors with the measured one for good radiation conditions and a desiccant load (storage in the morning and regeneration starts in the middle of the day) Under typical desiccant cooling load there is remarkable agreement between the predicted and the measured outlet temperatures, for both storage and regeneration periods even for a sever transition period at t=300 min. These three typical days show the capacity of the model to predict the outlet temperature of the collectors with a negligible error. page 40

IEA SHC Task 38 Solar Air Conditioning and Refrigeration Subtask C2-A, November 9, 2009 Storage Tank Two different scenarios are considered for the validation of the storage tank model. In the first only storage is considered (m 1 >0 and m 2 =0 in reference to Figure 5) while for the second it concerns desiccant cooling application with storage in the morning and regeneration in the afternoon (m 1 >0 and m 2 >0). The flow rates m 1 and m 2 are measured as well as the inlet temperature of the storage tank. The predicted and measured temperatures at the top and the bottom of the buffer are compared in the Figure 13 below. Storage load Typical desiccant load 90 90 80 80 70 70 Temperature [ C] 60 50 40 30 Temperature [ C] 60 50 40 30 20 10 0 Top experimental Top model Bottom experimental Bottom model 0 100 200 300 400 Time [min] 20 10 0 Top experimental Top model Bottom experimental Bottom model 0 100 200 300 400 Time [min] Figure 13: Comparison of the predicted temperature with measured one for the top and the bottom of the storage tank under a storage load (left) and a desiccant load (right) In both scenarios the model can accurately predict the temperature at the top and the bottom of the buffer. In the first case the stratification remains fairly constant but increases when the storage and the regeneration are combined. This may at first, appear incoherent as more mixing occurs in the latter case. The fact that stratification is greater with the regeneration is that the regeneration load was applied by water loss and the return water was cold and significantly below the tank temperature which amplified the stratification. Even with this extreme case the model was capable to predict accurately the temperature at the top and the bottom of the buffer. Now if the model has shown an acceptable accuracy when considered solely, we have to be sure that this accuracy will not be lost when coupling the components making the numerical simulation useless. In the next section the error propagation will be analysed and the impact on the supply temperature is examined. page 41

<strong>IEA</strong> SHC Task 38 <strong>Solar</strong> Air Conditioning <strong>and</strong> Refriger<strong>at</strong>ion Subtask C2-A, November 9, 2009<br />

Storage Tank<br />

Two different scenarios are considered for the valid<strong>at</strong>ion of the storage tank model. In the<br />

first only storage is considered (m 1 >0 <strong>and</strong> m 2 =0 in reference to Figure 5) while for the second<br />

it concerns desiccant cooling applic<strong>at</strong>ion with storage in the morning <strong>and</strong> regener<strong>at</strong>ion in the<br />

afternoon (m 1 >0 <strong>and</strong> m 2 >0). The flow r<strong>at</strong>es m 1 <strong>and</strong> m 2 are measured as well as the inlet<br />

temper<strong>at</strong>ure of the storage tank. The predicted <strong>and</strong> measured temper<strong>at</strong>ures <strong>at</strong> the top <strong>and</strong><br />

the bottom of the buffer are compared in the Figure 13 below.<br />

Storage load<br />

Typical desiccant load<br />

90<br />

90<br />

80<br />

80<br />

70<br />

70<br />

Temper<strong>at</strong>ure [ C]<br />

60<br />

50<br />

40<br />

30<br />

Temper<strong>at</strong>ure [ C]<br />

60<br />

50<br />

40<br />

30<br />

20<br />

10<br />

0<br />

Top experimental<br />

Top model<br />

Bottom experimental<br />

Bottom model<br />

0 100 200 300 400<br />

Time [min]<br />

20<br />

10<br />

0<br />

Top experimental<br />

Top model<br />

Bottom experimental<br />

Bottom model<br />

0 100 200 300 400<br />

Time [min]<br />

Figure 13: Comparison of the predicted temper<strong>at</strong>ure with measured one for the top <strong>and</strong> the<br />

bottom of the storage tank under a storage load (left) <strong>and</strong> a desiccant load (right)<br />

In both scenarios the model can accur<strong>at</strong>ely predict the temper<strong>at</strong>ure <strong>at</strong> the top <strong>and</strong> the bottom<br />

of the buffer. In the first case the str<strong>at</strong>ific<strong>at</strong>ion remains fairly constant but increases when the<br />

storage <strong>and</strong> the regener<strong>at</strong>ion are combined. This may <strong>at</strong> first, appear incoherent as more<br />

mixing occurs in the l<strong>at</strong>ter case. The fact th<strong>at</strong> str<strong>at</strong>ific<strong>at</strong>ion is gre<strong>at</strong>er with the regener<strong>at</strong>ion is<br />

th<strong>at</strong> the regener<strong>at</strong>ion load was applied by w<strong>at</strong>er loss <strong>and</strong> the return w<strong>at</strong>er was cold <strong>and</strong><br />

significantly below the tank temper<strong>at</strong>ure which amplified the str<strong>at</strong>ific<strong>at</strong>ion. Even with this<br />

extreme case the model was capable to predict accur<strong>at</strong>ely the temper<strong>at</strong>ure <strong>at</strong> the top <strong>and</strong> the<br />

bottom of the buffer.<br />

Now if the model has shown an acceptable accuracy when considered solely, we have to be<br />

sure th<strong>at</strong> this accuracy will not be lost when coupling the components making the numerical<br />

simul<strong>at</strong>ion useless. In the next section the error propag<strong>at</strong>ion will be analysed <strong>and</strong> the impact<br />

on the supply temper<strong>at</strong>ure is examined.<br />

page 41

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!