Final Report Summary - EGS PILOT PLANT (European geothermal project for the construction of a scientific pilot plant based on an Enhanced Geothermal System)
The EGS PILOT PLANT project started in 2004 and aimed to design, build and test a geothermal power plant based on multi-well enhanced geothermal systems (EGS).
The general objective was to produce electricity by designing technically and economically large industrial units up to >25 MWe. Thus, in order to be able to exploit unconventional reservoirs developed in fractured granite within the upper Rhine Valley, the following scientific objectives had to be tackled:
- improve the hydraulic performances of the three ultra deep wells and investigate both the inter-wells and the well to the geothermal reservoir connections;
- investigate the short-, medium- and long-term hydraulic, chemical, thermal and mechanical behaviour of the three-well EGS reservoir system;
- investigate, and analyse the risks for installations, environment, operating in order to define diagnostic, prevention and risk management policies;
- elaborate guidelines for future EGS plants.
For reaching those objectives, several tasks organised by work packages (WPs) were defined:
- WP1: Short term tests of the existing three-wells heat exchanger / reservoir system;
- WP2: Demonstration / long-term testing by building a power plant;
- WP3: Technical and economical design of larger industrial EGS units;
- WP4: Testing and improving 'strategic' equipment;
- WP5: Methodology of stimulations for EGS;
- WP6: General management and communication activities.
The Soultz experimental geothermal site consists of three deviated, 5 km deep wells drilled from the same platform within fractured granite. Geothermal water is pumped from the two peripheral production wells and re-injected at lower temperatures into the central re-injection well. The three geothermal wells are fully cased from the surface down to 4,5 km depth. A fourth geothermal well corresponding to a former exploration well drilled to 3.6 km depth could be used as an additional re-injection well if needed. The geothermal fluid which is natural brine (100 g/l) is pumped from the two deep production wells equipped with down-hole pumps. On surface, as a binary power plant has been designed, the geothermal water is delivering its geothermal energy to an organic working fluid (isobutane) coupled to a radial turbine which is itself connected to a generator. After giving its energy to the isobutane, the cooled brine is fully re-injected with a pump into the deep granite reservoir. The geothermal wells were drilled into a granite basement rocks during the previous EU phases and subsequently hydraulically and chemically stimulated between 2000 and 2007 in order to enhance the low initial permeability of the 5 km deep reservoir.
The three deep wells were first subjected to hydraulic stimulations, which resulted in an improvement of the productivity index of GPK2 and GPK4 approximately 20 times more than initially and of GPK3 approximately 1.5 times. Natural productivities or injectivities cover a wide range of values between 0.02 and 0.20 l/s/bar. Lowest values characterise production wells, GPK2 and GPK4, whereas the highest natural value was measured in the injection well GPK3. The impact of the hydraulic stimulation in GPK2 is significant although its initial natural productivity is low. The impact of the hydraulic stimulation in GPK4 is moderate although its initial productivity was very low. A variety of acid stimulations were conducted on all deep wells after the hydraulic stimulations. The improvement in three-day injectivity / productivity index by the acid treatments was a factor of about 1.25 for GPK2, about 1.15 for GPK3, and about 2.5 for GPK4.Hydraulic stimulations generated micro-seismic activity which was interpreted in terms of major structures in order to try to link event location with fault and fracture organisation. The stimulation of GPK2 in 2000 induced more than 700 seismic events with a magnitude greater than 1.0 which. The seismicity describes a dense, homogeneous cloud, without any apparent large-scale fault. The injectivity has been increased by a factor 20. The stimulation of GPK3 induced only about 250 events with a magnitude greater than 1.0 but with a greater proportion of large events, up to 2.9. The high clustering of the events suggests that the seismicity occurred principally on a single fault zone striking approximately NS and highly dipping to the west.
Several limited-duration circulations have been performed in the lower reservoir to date. The first circulation test of the triplet of wells penetrating the lower reservoir (4.5-5.0 km) took place for 5 months between July and December 2005. Tracer tests conducted during the circulation showed that approximately 25 % of the injected tracer was recovered from GPK2, but only 2% from GPK4. This asymmetrical response reflects the complex organisation of fractures describing different fluid circulation loops, the hydraulic connections between GPK3 and GPK2 being much more direct and faster than between GPK3 and GPK4. Results point to the conclusion that the exchanger is connected to a deep natural reservoir.
Based on numerical tool improvements, the modelling of hydraulic stimulation tests performed in the Soultz wells, followed by a post-stimulation hydraulic injection test, has been carried out. In a first step, the results show how the irreversible permeability increase is created by shearing within the fault zone plane during the hydraulic stimulation tests and the existence of a remaining hydraulic aperture once the well has been shutdown. As a second step, the 2005 circulation test was successfully simulated by taking into account the tracer (fluorescein) response in the two production wells GPK2 and GPK4. This transport modelling procedure takes into account the complexity of the fracture network at different scales (small-scale fractures, fracture zones) as well as the stress field.
The chosen heat-power conversion scheme is the organic Rankine cycle (ORC). Thus, a first 1.5 MWe ORC unit was designed and ordered to a European consortium. The different components of the power plant have been installed and power production of the first geothermal kWh was achieved on June 2008. However, to reach a significant fluid production, it is necessary to install down-hole production pumps, because the artesian production is not sufficient. Thus, two types of production pumps were deployed and tested in the wells: a line shaft pump (LSP) and a electro- submersible pump (ESP). Several hydraulic circulation tests were done with two wells (GPK2, GPK3) in summer 2008 and March 2009, and with three wells (GPK2, GPK3, GPK4) in November-December 2008. The first kWh was produced mid-June 2008. Several significant technological improvements were done above the surface (filtering, pump, lubrication system, heat exchanger), and the power plant is close to being fully operational despite some challenges yet to be solved.
At Soultz, corrosion and scaling were investigated based on an on-site testing system using three different steels in order to determine the reaction of those materials to the geothermal brine with high salinity. An innovative corrosion pilot was set up on the geothermal loop at the surface and tested on site. Newly created deposits, mainly sulphates, formed on the surface of samples. Corrosion occurred beneath the deposits and corresponded to pitting. A series of physico-chemical parameters, such as pH, Eh, conductivity and temperature, was measured in the meantime at re-injection conditions.
Microseismic activity related to the development of a geothermal site (stimulation, circulation) is one of the main challenge to solve for minimising large-scale event (M > 2) that could be felt by the local population. A risk analysis of the microseismicity during stimulation has been conducted through a numerical model describing induced seismicity during stimulation of the three deep Soultz wells. This study indicates that production of one or several wells after stimulation decreases the seismic activity. However, large magnitude (post-injection) events may still occur if the reservoir cannot be depressurised quickly enough.
Beyond this project phase, it is expected that this plant will be operating for more than 10 years and will continuously produce the important information flow required to accompany the development of an EGS industry in Europe.
The general objective was to produce electricity by designing technically and economically large industrial units up to >25 MWe. Thus, in order to be able to exploit unconventional reservoirs developed in fractured granite within the upper Rhine Valley, the following scientific objectives had to be tackled:
- improve the hydraulic performances of the three ultra deep wells and investigate both the inter-wells and the well to the geothermal reservoir connections;
- investigate the short-, medium- and long-term hydraulic, chemical, thermal and mechanical behaviour of the three-well EGS reservoir system;
- investigate, and analyse the risks for installations, environment, operating in order to define diagnostic, prevention and risk management policies;
- elaborate guidelines for future EGS plants.
For reaching those objectives, several tasks organised by work packages (WPs) were defined:
- WP1: Short term tests of the existing three-wells heat exchanger / reservoir system;
- WP2: Demonstration / long-term testing by building a power plant;
- WP3: Technical and economical design of larger industrial EGS units;
- WP4: Testing and improving 'strategic' equipment;
- WP5: Methodology of stimulations for EGS;
- WP6: General management and communication activities.
The Soultz experimental geothermal site consists of three deviated, 5 km deep wells drilled from the same platform within fractured granite. Geothermal water is pumped from the two peripheral production wells and re-injected at lower temperatures into the central re-injection well. The three geothermal wells are fully cased from the surface down to 4,5 km depth. A fourth geothermal well corresponding to a former exploration well drilled to 3.6 km depth could be used as an additional re-injection well if needed. The geothermal fluid which is natural brine (100 g/l) is pumped from the two deep production wells equipped with down-hole pumps. On surface, as a binary power plant has been designed, the geothermal water is delivering its geothermal energy to an organic working fluid (isobutane) coupled to a radial turbine which is itself connected to a generator. After giving its energy to the isobutane, the cooled brine is fully re-injected with a pump into the deep granite reservoir. The geothermal wells were drilled into a granite basement rocks during the previous EU phases and subsequently hydraulically and chemically stimulated between 2000 and 2007 in order to enhance the low initial permeability of the 5 km deep reservoir.
The three deep wells were first subjected to hydraulic stimulations, which resulted in an improvement of the productivity index of GPK2 and GPK4 approximately 20 times more than initially and of GPK3 approximately 1.5 times. Natural productivities or injectivities cover a wide range of values between 0.02 and 0.20 l/s/bar. Lowest values characterise production wells, GPK2 and GPK4, whereas the highest natural value was measured in the injection well GPK3. The impact of the hydraulic stimulation in GPK2 is significant although its initial natural productivity is low. The impact of the hydraulic stimulation in GPK4 is moderate although its initial productivity was very low. A variety of acid stimulations were conducted on all deep wells after the hydraulic stimulations. The improvement in three-day injectivity / productivity index by the acid treatments was a factor of about 1.25 for GPK2, about 1.15 for GPK3, and about 2.5 for GPK4.Hydraulic stimulations generated micro-seismic activity which was interpreted in terms of major structures in order to try to link event location with fault and fracture organisation. The stimulation of GPK2 in 2000 induced more than 700 seismic events with a magnitude greater than 1.0 which. The seismicity describes a dense, homogeneous cloud, without any apparent large-scale fault. The injectivity has been increased by a factor 20. The stimulation of GPK3 induced only about 250 events with a magnitude greater than 1.0 but with a greater proportion of large events, up to 2.9. The high clustering of the events suggests that the seismicity occurred principally on a single fault zone striking approximately NS and highly dipping to the west.
Several limited-duration circulations have been performed in the lower reservoir to date. The first circulation test of the triplet of wells penetrating the lower reservoir (4.5-5.0 km) took place for 5 months between July and December 2005. Tracer tests conducted during the circulation showed that approximately 25 % of the injected tracer was recovered from GPK2, but only 2% from GPK4. This asymmetrical response reflects the complex organisation of fractures describing different fluid circulation loops, the hydraulic connections between GPK3 and GPK2 being much more direct and faster than between GPK3 and GPK4. Results point to the conclusion that the exchanger is connected to a deep natural reservoir.
Based on numerical tool improvements, the modelling of hydraulic stimulation tests performed in the Soultz wells, followed by a post-stimulation hydraulic injection test, has been carried out. In a first step, the results show how the irreversible permeability increase is created by shearing within the fault zone plane during the hydraulic stimulation tests and the existence of a remaining hydraulic aperture once the well has been shutdown. As a second step, the 2005 circulation test was successfully simulated by taking into account the tracer (fluorescein) response in the two production wells GPK2 and GPK4. This transport modelling procedure takes into account the complexity of the fracture network at different scales (small-scale fractures, fracture zones) as well as the stress field.
The chosen heat-power conversion scheme is the organic Rankine cycle (ORC). Thus, a first 1.5 MWe ORC unit was designed and ordered to a European consortium. The different components of the power plant have been installed and power production of the first geothermal kWh was achieved on June 2008. However, to reach a significant fluid production, it is necessary to install down-hole production pumps, because the artesian production is not sufficient. Thus, two types of production pumps were deployed and tested in the wells: a line shaft pump (LSP) and a electro- submersible pump (ESP). Several hydraulic circulation tests were done with two wells (GPK2, GPK3) in summer 2008 and March 2009, and with three wells (GPK2, GPK3, GPK4) in November-December 2008. The first kWh was produced mid-June 2008. Several significant technological improvements were done above the surface (filtering, pump, lubrication system, heat exchanger), and the power plant is close to being fully operational despite some challenges yet to be solved.
At Soultz, corrosion and scaling were investigated based on an on-site testing system using three different steels in order to determine the reaction of those materials to the geothermal brine with high salinity. An innovative corrosion pilot was set up on the geothermal loop at the surface and tested on site. Newly created deposits, mainly sulphates, formed on the surface of samples. Corrosion occurred beneath the deposits and corresponded to pitting. A series of physico-chemical parameters, such as pH, Eh, conductivity and temperature, was measured in the meantime at re-injection conditions.
Microseismic activity related to the development of a geothermal site (stimulation, circulation) is one of the main challenge to solve for minimising large-scale event (M > 2) that could be felt by the local population. A risk analysis of the microseismicity during stimulation has been conducted through a numerical model describing induced seismicity during stimulation of the three deep Soultz wells. This study indicates that production of one or several wells after stimulation decreases the seismic activity. However, large magnitude (post-injection) events may still occur if the reservoir cannot be depressurised quickly enough.
Beyond this project phase, it is expected that this plant will be operating for more than 10 years and will continuously produce the important information flow required to accompany the development of an EGS industry in Europe.
