Page 303 - Geothermal Energy Systems Exploration, Development, and Utilization

5.7 Rosemanowes (United Kingdom) 279

Intact
rock
Single
discontinuity
2–3
discontinuity s t = K r Ea∆T
Several
discontinuity Restraint Restraint
Rock
mass
0 K r 1
Coefficient of restraint
Figure 5.23 Coefficient of restraint and relationship to rock mass.

modulus may be of the order of 50 GPa, and the thermal coefficient of expansion
−5
of the order of 10 . For a fully restrained system, given a 10 K cooling, this would
lead to a stress of 5 MPa, equivalent to a fluid head change of about 500 m. It can be
seen that for relatively small changes in temperature, relatively large changes in the
thermal stress can be expected comparative to the fluid pressure and in some cases
the tectonic stress. Additionally understanding the importance of the relationship
between the degree of fracturing and the coefficient of restraint becomes clear.

5.7
Rosemanowes (United Kingdom)

The Rosemanowes HDR site is located in the Carmenellis granite structure at Corn-
wall, southwest England (Figure 5.1). Stimulation and circulation experiments from
three wells (RH11, RH12, RH15) extended over a period of more than 10 years
(Richards et al., 1994). The long-term circulation test was carried out within the
RH12/RH15 reservoir (Figure 5.24a). The size and the shape of the accessible reser-
voirs were estimated from monitoring the microseismicity during the stimulation
of the wells. From statistical analysis of the data, two subvertical sets of fractures
were identified, which scatter around the two major strike directions of N165 and
N250. Therefore, a deterministic fracture network was established according to
these average orientations. Further, the number of fractures is restricted to those
that actually absorb water as detected by the well logs (Figure 5.24b).
The observed anisotropy in the tectonic stress field implies anisotropy in the
hydraulic behavior of the fracture system as well. For each case of hydraulic
anisotropy, fracture apertures were identified, which correspond to the measured
hydraulic reservoir impedance of 0.6 MPa l −1 −1
s . The thermal model was calibrated
on the basis of the production (bottom hole) temperatures in the borehole RH15.
The influence of several factors on the simulation results was studied in particular,
such as the initial temperature distribution, hydraulic anisotropy, and matrix