# Substituting Eq into Eq rearranging and put in a

During the experiment, the fluid volume flow was measured, together with the inlet and outlet temperatures. No real insulation was made to the upper part of the BHE. The experiment was conducted in a cooling mode, i.e. injection of heat into the ground. Fig. 3 shows the measured inlet and outlet temperatures versus time.

Fig. 3. Measured inlet and outlet temperatures during TRT.Figure optionsDownload full-size imageDownload as PowerPoint slide

4.2. Computational procedure

In this AVL-292 work, no attempt was made to conduct inverse calculations by a minimization of an objective function. (This will be carried out in a forthcoming work.) Rather, several spectral analyses were conducted by keeping all parameters fixed, but varying the soil thermal conductivity, until there is a match between the experimental results and the computed ones.

Time discretization of Tin and Tair signals was conducted using the forward FFT algorithm. The number of samples was 16,384 (214) and the sample length was 30 s, giving a time window of approximately 136 h. Spatial discretization of the soil mass was conducted using 100 Bessel function roots, and the far field boundary of the region-of-interest R was calculated asequation(44)R=6αtwhere α is the thermal diffusivity of the soil and t is the time when the temperature at point R reaches its maximum [1]. In this work t was set equal to 100 days, giving R approximately 12 m. A discussion on this choice is given in Ref. [16].