Dynamic simulations

By means of the calibrated model the system operation was simulated from June to August using climatic data for a typical year in Milano. The aim was to verify the importance of the progressive temperature increase in the perturbed ground on the system performance.

An on/off control acting on the pumps was introduced in the simulation model, with an air temperature set point T set = 24.5 °C. The mass flow rates had the same values as in the experimental campaign. Simulation results, in terms of discomfort and COP, are reported in Table 1 for each month and for the whole summer. The minimum discomfort is obtained in June but even in July and August it assumes very low values: discomfort values in the cooled room are about ten times lower than in the reference room.

discomfort|free floating

(h)

discomfort|sys

(h)

relative discomfort index

COP

June

230

11

0.05

3.1

July

797

93

0.12

2.6

August

584

68

0.12

2.1

Whole

summer

1611

172

0.11

2.4

Table 1: dynamic simulation results for the system in its current configuration

Assuming again typical weather conditions for Milano, a parametric study was carried out investigating the influence of the following parameters:

— the ground loop mass flow rate mg and the room loop mass flow rate mr

— the area of the radiant ceiling surface Ac

— the number of pipes N, the length of each pipe L and the distance between the axes of adjacent pipes d

Variation of any of the above-mentioned parameters affects the pressure drop in the loops. Then for each configuration the power demand of the pumps was calculated by considering the characteristic curve (pressure drop vs mass flow rate) of the system and average values of mechanical efficiencies for commercial pumps.

Mass flow rate. The influence of the mass flow rates was studied by giving the other parameters the current values in the experimental set up i. e. Ac= 10.8 m2, N = 10, L = 6 m and d = 1.5 m. For simplicity we chose mg — mr. The relative discomfort and the COP

dependence on the mass flow rates are plotted in Figure 4, while Qroom and Epjmps are reported in Figure 5. In the explored mass flow rate range, the flow regime remains laminar and so the convective coefficient in the pipes is the same for all mass flow rates. The pumps energy consumption Epumps rapidly grows with the mass flow rate and drives the COP trend. An optimal operating condition can be found minimising the relative discomfort: choosing a mass flow rate value around 100 kg/h gives the lowest discomfort with a still high energy efficiency.

Geometric parameters. The influence of the four geometric parameters was investigated by setting mg — mr = 100 kg/h and by varying Ac, N, L and d simultaneously, with the values given below:

— Ac: 3.6, 7.2, 10.8 and 14.4 m2 corresponding to respectively 25, 50, 75 and 100 % of the floor surface of the cooled test room

— N: 5, 10, 15, 20 and 30

— L: 3, 6, 10, 15 and 20 m

— d: 0.75, 1.5 and 3 m

From the simulation outputs it results that the dependence of the relative discomfort on the geometric parameters of the ground heat exchanger can be conveniently represented in terms of an effective heat exchange surface Seff. This quantity is defined as Seff = S • d/D,

where S = N.2nRpL is the total heat exchange surface in the ground and d/D is the ratio between the pipe spacing and a suitable linear dimension of the earth surface occupied by the pipes. The factor d/D weights the actual heat exchange surface taking into account the pipes density in the ground. At high density, i. e. low values of d/D and Seff, the tubes will be in contact with a portion of ground where temperatures might be significantly higher than in the undisturbed soil, hence reducing the effectiveness of the heat exchange. As it is shown in Figure 6, for every value of the radiant ceiling surface AC the relative discomfort is an exponentially decreasing function of Seff. The fitting parameters are reported in Table 2.

fitting function: y(Seff) = 1 + A(1 — e Seff/S*c)

Ac (m2)

A

Seff, c (m )

Z2

3.6

0.768

1.561

0.002

7.2

0.959

1.478

0.006

10.8

0.999

1.357

0.001

14.4

1.007

1.251

0.001

Table 2: Exponential fit parameters

The larger AC the lower the horizontal asymptote, i. e. the minimum discomfort achievable. In order to supply the comfort demand the radiant ceiling surface must be larger than 50 % of the ambient surface, but increasing from 75 to 100 % gives marginal advantages.

Design guidelines for the ground heat exchanger can be derived from the decay coefficient Seff. c of the exponential that decreases with AC. Once the effective surface is 4 times Seff, c, any additional increase produces marginal effects on the system performance.

The COP is a decreasing function of the relative discomfort, as shown in Figure 7. In fact the considered variation of the geometric parameters has a significant influence on the delivered thermal energy Qroom but little influence on the system pressure drops and on the electricity consumption Epjmps. Hence the best geometrical configuration from the point of view of comfort is also optimal from the point of view of efficiency.

Conclusions

In this paper the study of a ground cooling system carried out by means of a measurement campaign and dynamic simulations is reported. The experimental results highlight the interesting potentials of the proposed system, looking at the comfort conditions obtained and at the low electricity consumption. A TRNSYS simulation model of the system was developed and calibrated against measured data. A discomfort index and an energy efficiency index were introduced in order to evaluate the system performance. Dynamic simulations on the current system configuration for typical weather conditions in Milano showed that the progressive increase of the ground temperature during summer due to the system operation does not impair the system performance. A parametric study was carried out by changing the water mass flow rates and the main geometric parameters of the radiant ceiling and of the earth-to-water heat exchanger. Increasing the mass flow rate over a certain limit has little effect on the delivered thermal energy but compromises the energy efficiency. An optimal value of about 100 kg/h, giving the lowest discomfort and a rather high COP (around 10), was found. From the point of view of their influence on the relative discomfort index, the parameters of the ground heat exchanger can be grouped together into an effective heat exchange surface in the ground. For every value of the radiant ceiling surface the relative discomfort is an exponentially decreasing function of the effective surface. Design guidelines were extracted from these plots. In order to supply the comfort demand the radiant ceiling surface must be greater than 50 % of the ambient surface and the effective surface must be 3-4 times Seff, c. The COP is a decreasing function of the relative discomfort, so that a geometric configuration which minimises the discomfort maximises the COP at the same time.

The research will prosecute by studying the system behaviour under different climatic conditions and in connection with buildings of different characteristics.

[1]

Figure 2: Operative temperature in test room 1 (cooled) and in test room 2 (free floating),
radiant ceiling surface temperature in test room 1 and outdoor air temperature

SHAPE * MERGEFORMAT

mass flow rate (kg/h)