Page 289 - Geothermal Energy Systems Exploration, Development, and Utilization

5.5 Groß Sch¨ onebeck 265
Beside the fluid and solid properties the geometries and properties of the
hydraulic fractures and the boreholes have to be integrated into the model as well.
The structure of the hydraulic fractures can be easily represented by vertical 2D
quadrilateral fracture elements and those of the injection well E GrSk03/90 by
vertical 1D channel fracture elements. Owing to the deviation of the production
well Gt GrSk04/05, we used arbitrary 1D channel fracture elements to connect
the three involved hydraulic fractures. The properties of the hydraulic fractures
are summarized in Table 5.2. There, the hydraulic conductivity is corrected as
2
mentioned above. For both wells a cross-sectional area of 126.7 cm (5 in. diameter)
was used. The hydraulic conductivity of 1236 m s −1 for the injection well and
1648 m s −1 for the production well were estimated according to Hagen–Poiseuille
equation. For all discrete feature elements, the Equation of state (EOS) for fluid
density was applied.

5.5.4
Results

By means of the above described model, we simulated the 30-year life cycle of
geothermal power production. The first step included the calculation of the initial
pressure and temperature field by means of a stationary model after a simulation
period of 100.000 years. On the basis of the results of the stationary model, the
30-year life cycle was simulated. For this purpose, we assumed a production and
3
injection rate of 75 m h −1 and an injection temperature of 70 C. After 30 years
◦
of simulation time it becomes obvious that the injected cold water has reached the
production well, as shown in Figure 5.12. Further, the influence of the hydraulic
fractures is shown. At the injection well, the water accesses first the induced
hydraulic fracture and afterwards the connected matrix. At the production side,
the hydraulic fractures drain the matrix. If the cold water front reaches one of the
hydraulic fractures, then it will be directly forwarded to the production well.
For a detailed observation of pressure and temperature changes during the total
time of simulation, four observation points were set up along the wells. Observation
point 1 is located at the top of the hydraulic fracture at the injection well. The
observation points 2, 3, and 4 are located at the top of each of the hydraulic fractures
of the production well. The lowest observation point, 4, only shows the pressure
and temperature behind the waterfrac in the volcanic rocks, observation point 3
shows the sum of waterfrac and the first gel/proppant frac. Observation point 2
gives a cumulative value of all three fractures. The results are shown in Figure 5.13.
The hydraulic head increases approximately 400 m due to injection and decreases
approximately 500 m due to production. Taking into account, that the wells, the
hydraulic fractures, and the reservoir matrix are in full hydraulic contact and no skin
effects are present, the real hydraulic head change should be higher than simulated.
By means of the simulation, a quasistationary state was achieved after one year of
production and injection. In contrast, the temperature does not reach a stationary
state during the time of simulation. After five years, the cold water front reaches
the nearest production fracture (second gel/proppant frac). Starting from this time,