• No results found

radiation falling on the system was also used as the input for the Type 109 component, generally used for TMY data.

Figure 45: TRNSYS model of BIT-SWH system.

For the simulation the BIT-SWH was assumed suitable for use in a medium residence (4-5 occupants) in Hamilton, NZ. As such, the components used in the experimental BIT-SWH such as the collector with an area of 6m2 mounted at 5

used in conjunction with a 180 L storage tank was assumed suitable. In terms of thermal load two cases were assessed, the first being when no load was drawn from the system, this test was only used as a way of validating the accuracy of the simulation otherwise this test is impractical. The second test used a standardised load profile. This allowed for the use of the hourly load profile of the system which was simulated in the experiments by controlling the solenoid valve. The profile used is based on a peak daily thermal load of 25.6 MJ/day which is typical

4.2 No Load Simulation

To assess the accuracy of the simulation a pre-requisite test was conducted for the no load situation. This would allow a comparison to be made with the no load results obtained from the experiments. The collector parameters used in the simulation as well as the environmental data obtained from the experiments were applied so as to provide comparable results.

From an observation of Figure 46 and Figure 47, it is evident that the simulation of the temperature distribution in the tank matches closely to that of the results obtained from experiments. The temperatures within the tank for the simulation reach temperatures of over 60 °C, which is similar to the experimental results.

This result is also confirmed by the collector temperature readings and radiation plot, shown in Figure 48 for the simulation and Figure 49 for experimental.

However, there is a slight difference in the results for the collector plots. This is due to the simulation only using the radiation readings for the period starting at 8 am through to 4 pm where the radiation readings are above 0 W/m2. Aside from this, the temperatures predicted by the simulation of the collector reaching temperatures just below 70 °C have been confirmed experimentally. Therefore, it can be stated that this simulation model provides an accurate validation method for the performance of the BIT-SWH system.

Figure 46: Simulation results of tank temperature distribution for no load.

Figure 47: Experimental results of tank distribution for no load.

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

12:00 a.m. 4:48 a.m. 9:36 a.m. 2:24 p.m. 7:12 p.m. 12:00 a.m. 4:48 a.m.

Temperature (°C)

Local Time

Tt1 Tt2 Tt3 Ttank Telement

Figure 48: Simulated collector inlet and outlet temperature and radiation.

Figure 49: Measured collector temperatures and environment conditions.

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

-200.0 0.0 200.0 400.0 600.0 800.0 1000.0 1200.0

12:00 a.m. 4:48 a.m. 9:36 a.m. 2:24 p.m. 7:12 p.m. 12:00 a.m. 4:48 a.m.

Temperature C)

Radiation (W/m2)

Local Time

Radiation Ambient Collector Inlet Collector Outlet

4.3 Load Profile Simulation

Now that the simulation of a no load situation is complete, thus validating the method, a simulation can be carried out to predict the systems performance based on the standardised load profile. However, rather than use the efficiency equation of the ‘poorly’ performing BIT a hypothetical scenario was investigated.

In the case where the collector heat loss had been reduced, a hypothetical value of 5.55 for the heat loss was chosen such that it is comparable to a good performing integrated collector. These are shown by Equation 24 and Equation 25, representing the BIPVT collector investigated by Anderson (2009) and the modified BIT, respectively. This provides comparable results when running the simulation of the system operating under the standardised load. The weather data that was used was TMY data for Hamilton, New Zealand (Anderson, 2009).

η = 0.60 – 5.55 (Tci – Ta / G) (24) η = 0.75 – 5.55 (Tci – Ta / G) (25)

The measurement of the performance of both collectors in water heating applications was characterised using solar fraction, (f), shown in Equation 26 (Duffie and Beckman, 2006). The solar fraction is the amount of energy provided by the solar water heating system divided by the total energy required.

f = LS / L (26)

Based on the collector efficiencies shown in Equations 24 and 25 it can be seen in Figure 50, that during a week of operation over a summer week, in this case in the month of January, there is a noticeable difference in the solar fractions. The BIT collector which has the sole purpose of heating water almost meets the load requirement achieving solar fractions as high as 98 % for the last three days of the week. This suggests that if the heat loss issues of the BIT are addressed, its performance will increase considerably.

Figure 50: Solar fraction of BIPVT and BIT systems over summer.

If the simulation is also applied for a winter week, in this case TMY data for the month of June, it can be seen from Figure 51, that there is also a noticeable difference between the performances of the collectors compared with the summer week. The lower solar fraction values are due to the relatively low ambient conditions coupled with lower levels of solar radiation during winter (Anderson,

0.00 0.20 0.40 0.60 0.80 1.00 1.20

1 2 3 4 5 6 7

Solar Fraction



between the BIPVT-SWH system and SWH system. At its best the BIT-SWH system provided over 70 % of the required load which is shown for the final day of the week.

Figure 51: Solar fraction of BIPVT and BIT systems over winter.

In conclusion, there is strong evidence to suggest that if the heat loss of the BIT collector is reduced considerably (in this case from 13 down to 5) there will be a significant increase in its thermal performance. It is recommended that the issues relating to the high heat losses are addressed and resolved.

0.00 0.20 0.40 0.60 0.80 1.00 1.20

1 2 3 4 5 6 7

Solar Fraction



4.4 Simulation Summary

From the simulation modelling of the BIT-SWH system it is apparent that there are a number of issues to be addressed, particularly concerning the design of the collector.

It has been shown that the performance of the collector in an actual domestic water heating application, can be modelled accurately using the TRNSYS model.

Furthermore, simulation results when no load was drawn from the systems show a good correlation between the predicted and actual performance.

The modelling of the collector in a hypothetical situation when load was drawn showed that by reducing the collector heat loss, possibly by using extruded insulation, significant improvement in its performance characterised by solar fraction can be achieved. The outcome of this finding is that the BIT-SWH system becomes a viable means of achieving a large portion of the daily hot water requirements of a domestic household.

Chapter 5: Conclusions and Recommendations of Future