Page 279 - Geothermal Energy Renewable Energy and The Environment
P. 279
270 Geothermal Energy: Renewable Energy and the Environment
Table 14.1
surface area of Fractures (m ) for the Indicated
2
dimensions
distance from Injection well (m)
length (m) 50 m 100 m 1000 m 5000 m
2 200 400 4000 20,000
4 400 800 8000 40,000
6 600 1200 12,000 60,000
8 800 1600 16,000 80,000
10 1000 2000 20,000 100,000
20 2000 4000 40,000 200,000
50 5000 10,000 100,000 500,000
100 10,000 20,000 200,000 1,000,000
will usually cause rock fragments to shear off the fracture wall and remain as loose particles within
the fracture opening. Once the fluid pressure is removed when the stimulation process is completed,
these rock fragments will tend to prop open the fracture, maintaining the enhanced porosity. It is also
an option to introduce into the hydrofracturing fluid artificial proppants, such as sand or other materi-
als of selected sizes and hardness, to maintain fracture apertures once a system has been stimulated.
Commercially viable generating facilities have design lifetimes of about 20 to 30 years. That
time period places an important constraint on an EGS system. The reservoir must be managed such
that it can provide the required heat to power the facility over its design lifetime. That requirement
places constraints on the rate at which heat can be extracted from the reservoir. Namely, it is impor-
tant to maximize the exposed fracture surface area over which a given amount of heat is extracted.
This approach allows the temperature drop in the system to be minimized, per unit area of fracture
surface. As shown in Chapter 9, the rate of fluid flow that is required to maintain a 50 MW power
plant where the resource temperature is in the range of 200–250°C is on the order of 50 kg/s.
Table 14.1 lists the exposed surface areas of rectangular fractures that have propagated the indi-
cated distance from the well, and that have the stipulated length perpendicular to that direction.
2
Armstead and Tester (1987) have shown that surface areas on the order of 100,000 m per fracture
are required at these flow rates in order to keep the temperature drop along the fracture surface
small enough to assure adequate reservoir heat for the required lifetime. Clearly, this stipulation
requires that the ideal situation to maintain a long-lived thermal reservoir is that stimulation result
in a few fractures that propagate over significant distances.
Natural fractures are, however, more likely to have complex dimensional characteristics and form
cross-cutting networks, than evolve as simple planar features. A network geometry can result in a func-
tional system, provided the permeability along flow paths is adequate. Such networks must have fracture
permeabilities on the order of 10 to 50 md in order to sustain adequate flow for power generation.
In summary, the necessary features of a sustainable EGS system are that a stimulated zone
volume of about a few km must have a fracture network sufficient to sustain a flow rate of about
3
50 kg/s for about 30 years. The location and orientation of the stimulated zone must be sufficiently
well characterized to allow it to be a drilling target for production wells. Flow rates must be matched
to the actual permeability structure of the network in order to assure a sustainable resource.
eGs efforTs To daTe
Attempts to develop EGS systems began in the late 1970s and early 1980s with the Fenton Hill
“Hot Dry Rock” project that was carried out at Los Alamos National Laboratory, supported by
the Department of Energy (Smith 1983). This project was the first to undertake deep drilling,