HW
CW Space Heating
Collector Circuit
S
S
S
Domestic Hot Water
M
Boiler M
Tank
Fig. 4–1 Simplified hydraulic scheme of the simulation model of the solar combisystem.
Fig. 4–2 Layout of the TRNSYS model how it looks like in TRNSYS Studio.
The hydraulic concept of the simulation model is as close as possible as described in chapter 3 (page 33). Fig. 4–1 again shows the hydraulic concept in a TRNSYS adapted way. In Fig. 4–2 the layout of the simulation model is shown as a screen-shot of TRNSYS Studio, which is the user interface program of TRNSYS.
In contradiction to most published simulation models in this model also the hydraulic circuits between solar flat plate heat exchanger and solar tank and between boiler and tank are modeled with pipes leading to heat losses. Further on the pipes from the tank to the pump and the mixing valve are modeled as well as the return pipe to the tank.
Table 4–1 General informations and parameter settings of the simulation model.
Solar collector:
Collector area: 6 / 10 / 20 m2
Start efficiency: 80 %
Linear heat loss coefficient: 3.611 W/m2K Quadratic heat loss coefficient: 0.014 W/m2K2
Incident angle modifier: kΘ= −1 tana( / 2)Θ ; a = 3.3; for Θ = 50°: kΘ=0.92
Collector slope: 50°
Collector azimuth: South
Solar circuit:
Total pipe length primary circuit: 20 m Total pipe length secondary circuit: 11 m Inner pipe diameter: 20 mm Pipe insulation thickness: 20 mm
Pipe heat loss coefficient: 3.823 W/m2K; based on inner pipe diameter Solar collector fluid specific heat: 3.68 kJ/kgK
Solar collector fluid density: 1020 kg/m3
Solar collector flow rate: 15 kg/m2h (0.25 kg/ m2 min) Solar tank:
Total volume: 300 / 500 / 1000 Liter Overall heat loss coefficient: 2.70 / 2.73 / 3.79 W/K Auxiliary volume: 30 % of the total volume
Height: 1.6 m
Lower solar inlet height: 0.56 m (35 %)
Upper solar inlet height: 1.12 m (70 %) (or stratifier until the top) Boiler:
Maximum power: infinite
Efficiency: 100 %
Total boiler circuit pipe length: 13 m Inner pipe diameter: 20 mm Pipe insulation thickness: 20 mm
Pipe heat loss coefficient: 3.823 W/m2K; based on inner pipe diameter Load circuit:
Total load circuit pipe length: 13 m Inner pipe diameter: 20 mm Pipe insulation thickness: 20 mm
Pipe heat loss coefficient: 3.823 W/m2K; based on inner pipe diameter
4.1.1 Internal Pipes in the Tank
All the pipes inside the tank are influencing the thermal stratification in the tank due to heat exchange. If the boiler is charging the tank the forward pipe is hot and therefore also heating the lower part of the tank like an immersed heat exchanger.
Other way round, if hot water is taken from the top of the tank, e.g. for hot water preparation, some heat is lost on the way to the bottom of the tank. In order to take this effect into account, the pipe connections are modeled in the following way.
In the tank type 340 all pipe connections at the tank are done via so-called double-ports, if the goal is to exchange water. To transfer heat from or into the tank is also possible via an immersed heat exchanger. To model such an internal pipe therefore it is possible to use both elements in series. For example the boiler forward pipe (hot water into the tank) first is connected to an immersed heat exchanger with flow direction from bottom to top and then to the double-port which is placed at the correct height. The tank type 340 in principle gives the possibility to model four immersed heat exchangers, but only two are allowed to overlap. Therefore the following assumptions are done:
• Boiler circuit: If the boiler is in operation, always both the forward and the return pipe are in use. Therefore the effect of internal heat exchange is modeled only for the boiler forward pipe, which is also the long pipe. The heat transfer coefficient for this immersed heat exchanger is set to 8 W/K (Thür et.al. 2005). This value was found by a fit of two different simulation models: As a first step both the boiler return and the boiler forward pipe inside the tank were modelled in combination with an immersed heat exchanger. In the second step the heat transfer coefficient for the boiler return pipe was set to zero and the heat transfer coefficient for the boiler forward pipe was increased until the annual auxiliary demand was equal to the result of the first step.
• Load circuit: Since the load return flow anyway is relative cold and most of the time in the same magnitude as the temperature in the lower part of the tank, for this pipe no immersed heat exchanger is modeled. Only the load forward pipe coming from the very top of the tank and passing the whole tank to the bottom is modeled with an immersed heat exchanger. The heat transfer coefficient for this immersed heat exchanger is set to 4 W/K (Thür et.al. 2005).
• Solar circuit: In type 340 no immersed heat exchanger is available anymore which is allowed to overlap, since two are used now. Anyway, since the solar forward flow is controlled via a two-way valve the long pipe only is in use if the lower part of the tank is preheated. Therefore when the flow of the solar circuit is switched to the long pipe the temperature difference between the incoming flow and the bottom part of the tank is not that large any more and the effect is also not that strong anymore.
4.1.2 Piping within the Solar Combisystem
In Table 4–1 for all four hydraulic circuits (boiler-, load-, solar primary- and solar secondary circuit) the parameters for the pipes are given. The reason for this modeling is to find out how big is the influence of the heat losses of this piping which is counted as part of the heat supply system and not of the space heating distribution system. Since these pipes (except the primary solar circuit) are in the same room as
the tank and the boiler, it is just consequent that the heat losses of these pipes are also taken into account.
In practice the situation is much more complicated, because a lot of further components than pipes are part of the piping, for example pumps and valves, which are typically not insulated and therefore can cause higher heat losses than the pipe itself.
As a first step, in this study the model is based on a simplified approach. Instead of trying to calculate an average heat loss of all components in the hydraulic circuit, the length of the insulated pipes was just enlarged. Since each system depending on the boundary conditions is looking different in that point of view, anyway these assumptions would be of low accuracy. To achieve the goal of getting some first results on this topic, it is assumed that this approach is sufficient.
4.1.3 Hot Water Preparation and Space Heating
The hot water preparation is modelled in a simplified way. It is assumed that the flat plate heat exchanger is designed in a correct way which always allows to prepare hot water at the desired tap temperature.
In practice the heat transfer coefficient of a flat plate heat exchanger of course is strong depending on the operation conditions. Also the thermal mass and the dynamic behaviour of the speed controlled pump has a quite big influence. On the other side the simulation time step is six minutes, where many hot water tappings in practice are shorter or much shorter. Additionally, measurements in the demonstration house showed (see examples in chapter 6.3.3 on page 121) that it takes up to about one minute until the process of hot water preparation is fully stable.
Therefore it was decided to keep the TRNSYS model simple and to skip the heat exchanger and the speed controlled pump and to fix the temperatures as following:
• Hot water set point temperature: 50°C
• Primary forward temperature: 55°C
• Primary return temperature: 15°C
The reason for also fixed primary return temperature is unsatisfactory, but a result of practical work with TRNSYS: When the annual oscillation of the cold water temperature was introduced to the model, due to unexplainable reasons the program stopped the calculations with error messages. The problem unfortunately could not be solved in time.
The hot water circulation in this solar combisystem concept plays a major role in practice. But since the hot water distribution system is not modeled also no hot water circulation is taken into account.
Space heating is covered with the flow rate and the forward and return temperature always according to the load file. Due to the double use of the mixing valve and the space heating pump one effect has to noted: During periods of hot water preparation the space heating load is not covered since the 3-way valve also in the model is switching to the hot water heat exchanger and therefore stopping the space heating flow rate. This leads finally to a reduced load in comparison with the theoretical load according to the load file.
The return flow coming from the hot water heat exchanger or the space heating circuit is stratified into the tank via a 3-way valve with two possible heights in the tank.
Depending on the return temperature and the tank temperature (at 30 % height) the return flow is stratified into the tank via the low or the high inlet pipe.
4.1.4 Boiler Integration
The boiler in all cases has 100 % efficiency and is able to modulate from 0 to 100 %.
The boiler is controlled slightly different, depending if used in the solar combisystem or in the reference system: