Parametric study of an enhanced geothermal system based on thermo-hydro-mechanical modeling of a prospective site in Songliao Basin

Applied Thermal Engineering 105 (2016) 1–7

Contents lists available at ScienceDirect

Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Parametric study of an enhanced geothermal system based on thermo-hydro-mechanical modeling of a prospective site in Songliao Basin Xiaoxue Huang a,b,⇑, Jialing Zhu a,b, Jun Li a,b, Chengyu Lan c, Xianpeng Jin c a b c

Key Laboratory of Efficient Utilization of Low and Medium Grade Energy, Ministry of Education, Tianjin University, Tianjin 300072, China Tianjin Geothermal Research and Training Center, School of Mechanical Engineering, Tianjin University, 92 Weijin Road, Nankai District, Tianjin 300072, China Daqing Oilfield Limited Company, Longshi Road, Ranghulu District, Daqing, Heilongjiang Province 163453, China

h i g h l i g h t s
EGS reservoir simulation with thermo-hydro-mechanical coupling was carried out.
Hydraulic conductivity is enhanced due to the thermal stress during production.
Permeability and porosity play major role in terms of the productivity.
For a vertical fracture, increasing fracture thickness improves the productivity.
For vertical wells, production wells should be deeper than injection well.

a r t i c l e

i n f o

Article history: Received 28 February 2016 Revised 2 May 2016 Accepted 24 May 2016 Available online 24 May 2016 Keywords: Reservoir property Wellbore arrangement Production optimization Thermo-hydro-mechanical modeling Enhanced geothermal system

a b s t r a c t The promising technology of Enhanced Geothermal System (EGS) using the vast extent of geothermal energy resources proves to be feasible according to research conducted over the past few decades, whereas there is still a lack of thorough understanding of the underground process. Thermo-hydromechanical modeling of a conceptual reservoir based on the geological setting of a target area in Songliao Basin was carried out. The effects of the geological background and the parameters determined during the stimulation and production processes were investigated regarding the output of the reservoir. The results demonstrated that properties that are able to strength the heat convection in the reservoir is able to evaluate the power output significantly, and wellbore arrangement that takes advantage of the buoyancy flow is also a valuable way to increase the production. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction As one of the renewable energy resources, geothermal energy is abundant but underdeveloped [1]. The fact remains that geothermal fields of sufficient quality to produce economic electricity are rare, and thus the tapping of geothermal energy is limited to a handful of locations [2]. The promising untraditional means of Enhanced Geothermal System (EGS) makes these ubiquitous resources accessible. Normally, hydraulic stimulation is applied to increase the permeability and expand the heat transfer area of the fractured reservoir, and

⇑ Corresponding author at: Key Laboratory of Efficient Utilization of Low and Medium Grade Energy, Ministry of Education, Tianjin University, Tianjin 300072, China. E-mail address: [email protected] (X. Huang). http://dx.doi.org/10.1016/j.applthermaleng.2016.05.142 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.

water is injected to be heated and pumped out. In a typical EGS project, we first stimulate a large rock volume, drill into the stimulated region to establish a connected reservoir, circulate fluid with acceptable pressure losses at near commercial rates, and generate power at the ground surface using the produced thermal energy. The total costs of a EGS project was subdivided into five categories: drilling cost, production cost, costs for feed pumps and surface installations, stimulations cost, and other operational costs [3]. Hofmann, etc. estimated that for an EGS project with 3 wells of 5 km, the cost of drilling makes up 51% of the total cost, and pumping cost for one injection well and two production wells is approximated to be 42% [3]. Therefore, costs of EGS projects can vary significantly depending on the production scheme and the effectiveness of stimulation. For a specific EGS site, the stimulation scheme and drilling plan should be optimized to achieve better hydraulic and thermal performances at the production stage [4,5].

2

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7

Nomenclature cr F E g G H h k K _ m P q T Tref u uw Wh Wp

a

b /

rock specific heat, J/(kg °C) body force per area, Pa Young’s modulus, Pa gravitational acceleration, m/s2 shear modulus, Pa depth of the well, m specific enthalpy of injection, J/kg permeability, m2 bulk modulus, Pa mass flow rate, kg/s pore pressure, Pa source term, kg/m3 or J/m3 temperature,  C reference temperature,  C velocity vector, m/s internal energy, J/kg heat output, W pump consumption, W Biot’s coefficient, dimensionless linear thermal expansion coefficient,  C1 porosity, dimensionless

Over three decades of research and development of EGS reservoirs in several countries, certain experiences were gained, and a number of problems were revealed [6]. The first attempt to make a full-scale EGS reservoir in the Fenton Hill demonstrated the feasibility of hydraulic conductivity enhancement through stimulation. However, the marked water loss rate indicated that water pressure should be controlled to avoid the reservoir growing and to minimize water losses. The pilot EGS power plant in Soultz successfully created an artificially stimulated reservoir of commercial size although with production rates still below required levels for an economical power plant. One of the production wells, GPK2, was stimulated in the open-hole section from 3211 to 3876 m, forming a stimulated volume of about 0.24 km3 [7]. Additionally, the injection well GPK3 was stimulated to extend the existing reservoir of GPK2 by an overlapping volume of enhanced permeability. Another injection well GPK4, which was separated from the two production wells by about 650 m and 600 m, was stimulated [7]. In the cooper basin in Australia, stimulation in the well Habanero-I was carried out at depths between 4136 m and 3994 m, and the fractured volume was extended to cover a horizontal pancake-shaped area of 4 km2 in the year 2005 [6]. The Ogachi project was also considered as an EGS project as the temperatures were high and the productivity was low. Two fractures were stimulated while each of them was only created in 10 m of open hole in the well bottom. After stimulation in the production well, the recovery rate was improved from 3% to 25%, whereas still being small [8]. One of the lessons learned at Ogachi is that although the permeability after fracturing was found to be 104–105 cm/s, a relatively high value, the mass flow rate at the producer was still small as the total fractured reservoir volume was only about 250 m3, suggesting that efforts to connect the two wells should be enhanced in addition to enhancement of the reservoir permeability [8]. Another EGS project in Japan, the Hijiori project, has one injector and two producers. It was demonstrated that the reservoir grew and the permeability was elevated even more during circulation tests than during the stimulation process, emphasizing the mechanical effects as water flow in the reservoir [9]. Generally, in these EGS projects, two or three wells were drilled to a depth of 3000–4000 m for heat mining. Most of the

/0

gh gP

k kr

ev u

m rm q l

initial porosity, dimensionless energy efficiency, dimensionless pump efficiency, dimensionless Lame’s constant, Pa rock thermal conductivity, W=ðm KÞ volumetric strain, dimensionless displacement vector, m Poisson’s ratio of rock, dimensionless mean normal stress, Pa density, kg/m3 viscosity, Pa s

Subscripts inj injection well pro production well r rock w water m mass h heat

recovery rates remain low, which were influenced both by the connectivity of the reservoir, namely permeability, and the scale of the reservoir. Sufficient reservoir volume and heat transfer area are crucial to the hydraulic and thermal performances [4,5], but water loss and expense would be significant simultaneously. Besides, the pressure gradient should be controlled to avoid reservoir expansion and permeability enhancement that will induce water loss. Accordingly, reservoir design and well arrangement should be optimized to achieve better performances based on an understanding of the production process. The dynamic production process was numerically investigated based on thermo-hydro-mechanical coupling of the reservoir in our study. A target area in north of Songliao Basin was simulated. Then parametric study was implemented to identify the factors that influence the production process, and suggestions for optimization of the production were derived. 2. Numerical model with thermo-hydro-mechanical coupling 2.1. Model description Fluid and heat flows are coupled with geomechanical effects for field-scale reservoir modeling. Geomechanics is fully coupled and developed from the linear elastic theory, and is formulated with the mean normal stress as well as pore pressure and thermal stress. The fluid flow and heat transfer portion is based on the general-purpose numerical simulator TOUGH2 [10,11]. Fluid flow is described with a multiphase extension of Darcy’s law, and heat flow is governed by conduction and convection as shown by Eqs. (1) and (2).

d q ð/qw uÞ ¼ k w ðrP  qw gÞ þ qm dt lw

ð1Þ

d ð/qw uw þ ð1  /Þqr C r TÞ ¼ kr rT þ qw hw u þ qh dt

ð2Þ

Assuming that boundaries of each block element can move as an elastic material, the Hooke’s law for thermo-poro-elastic medium is given by

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7


 2 rm  aP  3bKðT  T ref Þ ¼ k þ G ev 3

ð3Þ

Expressing the strain in terms of a displacement vector, and take the divergence of Eq. (3), the thermo-poro-elastic Navier equation can be written as

ar2 P þ 3bK r2 T þ ðk þ 2GÞr2 ðr  uÞ þ F ¼ 0

ð4Þ

By combining Eqs. (3) and (4), the governing geomechanical equation is

3ð1  mÞ 2 2ð1  2mÞ r rm þ r  F  ðar2 P þ 3bK r2 TÞ ¼ 0 ð1 þ mÞ ð1 þ mÞ

ð5Þ

The mean normal stress is the additional primary variable in parallel to pore pressure and temperature. The volumetric strain is another geomechanical variable, which can be solved with Eq. (3). In addition, reservoir rock properties, including porosity and permeability, are subjected to change due to rock deformation. The relationship between porosity and effective stress derived by McKee et al. [12] is applied.

ecp ðr r0 Þ   0 0 1  /0 1  ecp ðr r0 Þ 0

/ ¼ /0

0

ð6Þ

The permeability was related with the porosity by the CarmanKozeny equation.

k ¼ ki


3 / / ð1  /Þ i

ð1  /i Þ2 2

ð7Þ

In Ref. [12], the laboratory core tests of granite, which is the most common rock type encountered in the formations of EGS reservoir, demonstrated an excellent match with the relationship predicted by Eqs. (6) and (7) with correlation coefficients above 0.95. The integral finite difference method is employed, and time is discretized fully implicitly as a first-order backward finite difference. The space discretization is carried out with a grid distance of 10 m in the horizontal plane, and reduced to 5 m around the wells. The grid distance is set to be 3 m in the vertical direction. 2.2. Verification of the model TOUGH2-EGS is employed to carry out the simulation. In Ref. [13], the applicability of the model was tested against analytical solutions for temperature-induced deformation and pressureinduced flow and deformation. The poroelastic was verified by comparing the numerical result against the analytical solution of 1D consolidation problem and the thermoelastic was verified against the analytical solution of 1D heat conduction. The excellent match between the numerical results and the analytical results lends creditability to the numerical model. Finally the model was applied to model the Geysers geothermal field, and the results demonstrated that the model can be used for field-scale reservoir simulation with fluid and heat flow, together with geomechanical effects. 3. Physical model for the perspective EGS site in Songliao Basin The studied area lies in Shuangcheng Fault Depression, north of Songliao Basin, China. Based on the detailed geological background given in a previous study [14], Yingcheng formation which constitutes the high temperature reservoir is modeled (Fig. 1). The horizontal dimension of the model is 3000 m  1500 m. The formation is simplified to be uniform and treated as single-porosity, isotropic

3

porous medium and fully saturated with liquid water. The injection well is located at the model center and flanked by two production wells 800 m apart from it in x direction. 3.1. Reservoir properties Several exploratory wells exist in the region, the mechanical and thermophysical properties obtained from the well tests are shown in Tables 1 and 2. The thermophysical properties of the depth interval 3879.4–3910.6 m are employed for the base case of the simulation, and then parametric study is implemented with the properties ranging within the values given in Table 1. For the mechanical properties, Average values of the mechanical properties in Table 2 are taken for the base case, which is 2400 MPa for Young’s modulus, and 0.19 for Poisson’s ratio. The heat flux from the ground surface was measured to be 50–90 mW/m2 in Songliao Basin, and determined to be 84.15 mW/m2 in the simulated area [14]. 3.2. Fracture properties Field-scale stimulation test was carried out and it showed that vertical fractures were formed. Thus, the fractured zone is modeled as a cuboid with small thickness. The fractured zone is assigned a bulk permeability of 3 orders of magnitude larger compared to the surrounding rock formation, and the other properties are kept the same as those of the surroundings. 3.3. Boundary conditions In Ref. [14], the coupled thermo-hydraulic effect was considered for a production period of 30 years, and the simulation results demonstrated that the properties at the top surface of Yingcheng formation remains approximately constant. Therefore, the top layer in this model is set to be fixed values of 34.9 MPa for hydraulic pressure and 149 °C for temperature. In addition, the normal stress is 94.3 MPa, the value of the gravity of the upper rock layers. The corresponding horizontal stresses are set with lateral pressure coefficients of 0.8 and 1.2, respectively. For the other three directions, the displacements were set as zero. All these lateral boundaries are set with no heat and flow fluxes as in Ref. [14] no significant heat and mass transfer occurred at the locations of the model boundaries. A constant heat flux was applied at the bottom surface due to the geo-temperature gradient. Water with enthalpy of 280 kJ/kg is injected at a flow rate of 80 kg/s. 4. Results and discussion 4.1. Thermo-hydro-mechanical coupled results Fully coupled velocity, temperature and stress fields were obtained for a production period of 30 years. The results of the plane where the wells are located (y = 750 m) are shown in Figs. 2 and 3. The temperature contour in Fig. 2 suggests that the denser fluid at the injection spot flows not only horizontally toward the production well, but more significantly, also downward, which is the same as previous thermo-hydraulic studies [14]. Therefore, the area beneath the injection point is cooled down initially, and an expanding cone-shaped area of the cooled region is formed. The downward flow is held back at the boundary of the fractured zone, forming the border of the cooled region. The cooling effect of injection induces a decrease of the thermal stress, causing contraction of the rock matrix, simultaneously raises the permeability. Meanwhile, due to water injection and

4

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7

Fig. 1. Schematic of the conceptual EGS model.

4.2. Parametric study of productivity

Table 1 Thermophysical properties from the well tests. Depth interval (m)

Density (kg/m3)

Porosity (%)

Permeability

3173.4–3190.8 3191.8–3211.8 3231.4–3300.0 3879.4–3910.6 4391.4–4385.8 4417.8–4454.0 4518.75–4520.80

2400 2420 2750 2440 2610 2460 2670

12.5 11.8 4.7 8.3 9.7 9.3 7.0

3.63  1015 m2 2.67  1015 m2 0.57  1015 m2 0.32  1015 m2 0.1  1015 m2 0.08  1015 m2 0.02  1015 m2

The conceptual model in Section 4.1 is based on the geological background and well test data in the targeted site, with the properties set as the averages of the available data. As reservoir properties, the stimulation and the production schemes may vary at different sites, the parametric study is carried out to identify the factors that contribute to the production. For long time water circulation, the wellbore flow can be regarded as isenthalpic [15]. The heat output is then calculated by

Z Wh ¼

Table 2 Mechanical properties from the well tests.

t end

  _ hpro  hinj dt m

ð8Þ

0

Layer number

Young’s modulus (MPa)

Poisson’s ratio

117 118I 119

26,920 17,947 26,836

0.189 0.201 0.181

production, pore pressure changes and alters the permeability. At the injection well, pore pressure increases, and thus the permeability increases, whereas the contrary effect occurs around the production well. Due to the thermal stress and pore pressure variation, at the injection spot, effective stress, which is the difference between the stress of the solid matrix and the pore pressure, is enhanced by both effects, while at the production spot, these two effects present opposing influences on it. Fig. 3 shows the porosity variation over production. As a result of the effective stress change, for the first 20 years, porosity in most area of the fractured zone is increased as a result of the injection and the temperature drop, whereas in the area around the production well, porosity decreases as compaction occurs during extraction. It is also demonstrated in Fig. 3 that, at approximately the end of production, porosity of the whole fractured zone is elevated, and the effect of thermal stress overwhelmed that of the pore pressure as the cooled region expands to the production well eventually.

The energy efficiency of the system, g, is defined as the ratio of the total produced thermal energy to the internal energy consumption. The internal energy consumption of the system is mainly comprised of the pump consumption, W p , which constitute the pressure loss in the reservoir and the pressure driving force in the wellbore [16]. W p can be defined as

Wp ¼

1

gP

Z 0

t end

_ m

P pro

qpro



Pinj

qinj

!

  dt  g Hinj  Hpro

Z

t end

!

_ mdt

ð9Þ

0

The energy efficiency gh based on heat output is then:

gh ¼

R _ pro  hinj Þdt gP 0tend mðh Wh   ¼ Rt  R P P pro inj end WP _ q  q dt  g Hinj  Hpro 0tend mdt _ m 0 pro

ð10Þ

inj

4.2.1. Effects of reservoir properties The reservoir properties vary from site to site, influencing the results of the heat output and lifespan of the geothermal reservoir. In addition, permeability of the fractured zone depends on the effectiveness of stimulation, and is a major factor determining the heat convection. Fig. 4 shows the sensitivity of W h and gh to the rock conductivity (k), porosity (/), Yong’s modulus (E) and Poisson’s ratio (m), permeability in the fractured zone (k) and heat flux (q) at the bottom surface.

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7

5

Fig. 2. Time evolution of temperature profile at the chosen section plane (y = 750). (a) t = 5 years; (b) t = 10 years; (c) t = 20 years; and (d) t = 30 years.

Fig. 3. Time evolution of porosity profile at the chosen section plane (y = 750). (a) t = 5 years; (b) t = 10 years; (c) t = 20 years; and (d) t = 30 years.

k is considered isotropic and changed by magnitudes of orders for the parametric study. It can be easily derived that the permeability is the most sensitive factor among the others shown in Fig. 4. As indicated by Eq. (1) that the flow rate is proportional to

the permeability, convection is accelerated due to the permeability increase. Therefore, higher enthalpy of production can be obtained especially during the later stages of production as premature thermal breakthrough shown in Fig. 2 can be alleviated.

6

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7

Fig. 5. Relative change in heat output with varying reservoir dimensions.

Fig. 4. Relative change in heat output with different reservoir properties.

The heat mining ability is also sensitive to the porosity. Increase of / provides a higher volume of in-situ fluid, which is the heat source for convection of the cold injected fluid. As the intensity of heat convection is higher than the conduction, the sensitivity of porosity is higher than that of the rock conductivity. The rock thermal conductivity impacts W h in a similar way as /. Higher heat flux supplied to the fluid from the rock matrix can be attained with an increase in the thermal conductivity, whereas as shown by the well tests, variation in k of different rock samples is less than other properties, therefore, influence of k is not as significant as / although the slope of k is relatively high in Fig. 4. The influence of the bottom heat flux is almost negligible, which also suggest that convection in the fractured zone dominates the heat transfer in the whole modeled domain, and recovery by q through heat conduction in the rock matrix beneath the fractured zone exhibits very little influence on the temperature field in the reservoir. Thus, q plays a minor role in reservoir selection if only the heat mining process is considered, whereas it definitely could help shorten the recovery interval for sustainable usage of the reservoir, as in a typical EGS reservoir, the only heat source during recovery period is the heat flux supplied at the bottom surface. The mechanical properties, Young’s modulus and Poisson’s ratio of the rock, are inversely proportional to the heat output. As decrease of both E and m incur more significant variation in the rock strain, the rock porosity increase is elevated. Therefore, smaller E and m provide higher W h , although the influence is negligible compared with the other properties in Fig. 4. 4.2.2. Effects of reservoir dimensions Regarding the reservoir dimensions, the economic targets for the effective heat transfer area and the fractured rock volume that will make it a viable proposition are 2  106 m2 and 2  108 m3, respectively, which are just below the values used for the base simulation [17]. Many studies assumed that the heat transfer area would be formed by the sides of a single ‘Penny-shaped’ fracture, and in this targeted area, the stimulation tests demonstrated that the hydraulic fractures are vertical. Thus, in the above base case, the dimension of the fracture is set to be 2100  200  488 m3. The reservoir volume is enlarged by extending the thickness (b) in y direction or lengthening the fracture height (h) in z direction. As shown in Fig. 5, the former presents more significant impact on

the output while the latter is negligible when the fractured volume is increased. It can be explained that the vertical flow is limited to the density variation due to the temperature difference: although the vertical dimension is prolonged, the effective heat transfer area is not extended to that extent as the fluid path remains almost unchanged. When b or h is decreased, W h decreases more when reducing b than that induced by reducing h. Another point to note is that the absolute value of output variation of decreasing the fracture rock volume through these two ways is higher than that of increasing it, suggesting that this magnitude of fractured reservoir volume is approximately the critical one as the output elevation by more investment in stimulation is comparatively small but the output reduction is noteworthy if less stimulation volume is obtained. Fig. 6 compares the energy efficiency with different parameters, including the reservoir properties and the fracture dimensions. gh of the base case is shown as the intersection of the coordinate axes in Fig. 6. Evaluation of the fracture permeability enhances both the heat transfer process and the hydraulic flow, therefore, gives the highest value of gh . The fracture dimensions also play a vital role in the determination of gh . In terms of the fracture height, increase of h results in a larger area with low flow resistance, and thus, a higher gh . Interestingly, decrease of h also brings about a higher gh compared to the base case, that is because of the weaken of vertical flow, and thus a smaller amount of pump consumption.

Fig. 6. Energy efficiency with different parameters.

X. Huang et al. / Applied Thermal Engineering 105 (2016) 1–7 Table 3 Relative change in the heat output of different cases. Case

#1

#2

#3

#4

#5

#6

#7

Relative change (%)

3.66

3.98

10.73

11.37

7.10

12.99

18.97

Increase of the fracture thickness definitely improves the heat extraction amount as shown in Fig. 4, whereas b is inversely proportional to gh as thicker fracture may incur more significant water loss and require higher pump power for the fluid circulation. 4.2.3. Effects of production scheme As the drilling cost makes up a major percentage of the total investment in an EGS project, and the well arrangement plays an important role in the production [18], the production scheme regarding the wellbore numbers, the well distances and well depths is evaluated. Results are shown in Table 3. Case #1 and #2 changes the well spacing from 800 m to 600 m and 1000 m, respectively. In Case #3–#6, the well depth is different from the base case in which the well bottoms are located at the top of the fractured zone. In Case #3 and Case #5, the injection well is penetrated into the bottom and the middle of the fracture zone, respectively. In Case #4 and Case #6, the production well is drilled into the bottom and the middle, respectively. Case #7 employed two wells with well spacing remaining 800 m. As shown by Table 3, decrease of the horizontal well spacing incurs earlier thermal breakthrough as heat exchange area is reduced. To take advantage of the vertical flow resulting from the density variation of cold water injection, the injection well should be placed higher than the production well, as both Case #3 and #5 shows a reduction of output when injection well is below the production wells. The production increases with increasing depths of the production wells, whereas Case #6 with the production wells at the middle of the fractured zone is better than Case #4 with those at the bottom. This can be explained that in Case #6, the path of fluid is longer than in Case #4 as hot water flows upward after reaching the bottom of the rock, and thus larger effective area of heat exchange is available, and more heat extraction is reached in the long run. It is also clear that two wells with the same well spacing as the base case exhibits a much lower production ability mainly because of the enthalpy decrease as thermal depletion occurs earlier, Thus, utilization of three production wells performs better if the economic factors including the drilling cost and stimulation cost are not considered. 5. Conclusion Allowing for the dynamic change of stress field, the hydraulic performance is improved due to permeability enhancement by the cooling effect of injection as the influence of thermal stress overwhelms that of the pore pressure. Heat production can be increased at the cost of creating a larger fractured volume, and the increase is more pronounced when fracture thickness is increased than that by extending the height of the fracture. The permeability and the porosity of the fractured rock enhance the thermal extraction significantly by providing more heat transfer area for the heat convection. The mechanical proper-

7

ties and the heat flux at the bottom surface are not sensitive factors in terms of the heat output and the energy efficiency. Regarding the wellbore arrangement, the production wells should be lower than the injection well to take advantage of the buoyancy flow, but not to be at the bottom of the fractured reservoir. Acknowledgement The work was supported by the National Natural Science Foundation of China (Grant No. 41272263). The authors would like to express their thanks to Daqing Oilfield Limited Company, PetroChina for supplying geological data and Professor Yu-shu Wu at Colorado School of Mines for sharing the code TOUGH2-EGS. References [1] Zhe Zhou, Pei Liu, Zheng Li, Weidou Ni, An engineering approach to the optimal design of distributed energy systems in China, Appl. Therm. Eng. 53 (2) (2013) 387–396. [2] C.W. Chan, J. Ling-Chin, A.P. Roskilly, A review of chemical heat pumps, thermodynamic cycles and thermal energy storage technologies for low grade heat utilization, Appl. Therm. Eng. 50 (1) (2013) 1257–1273. [3] Hannes Hofmann, Tayfun Babadagli, Günter Zimmermann, Hot water generation for oil sands processing from enhanced geothermal systems: process simulation for different hydraulic fracturing scenarios, Appl. Energy 113 (2014) 524–527. [4] Abdul Ravoof Shaik, Sheik S. Rahman, Nam H. Tran, Thanh Tran, Numerical simulation of fluid-rock coupling heat transfer in naturally fractured geothermal system, Appl. Therm. Eng. 31 (10) (2011) 1600–1606. [5] Fu-Zhen Zhang, Rui-Na Xu, Pei-Xue Jiang, Thermodynamic analysis of enhanced geothermal systems using impure CO2 as the geofluid, Appl. Therm. Eng. 99 (2016) 1277–1285. [6] Katrin Breede, Khatia Dzebisashvili, Xiaolei Liu, Gioia Falcone, A systematic review of enhanced (or engineered) geothermal systems: past, present and future, Geothermal Energy 1 (1) (2013) 1–27. [7] Albert Genter, Xavier Goerke, Jean-Jacques Graff, Nicolas Cuenot, Gérard Krall, Marion Schindler, Guillaume Ravier, Current Status of the EGS Soultz Geothermal Project (France), World Geothermal Congress, Bali, Indonesia, 2010. [8] Pascal Audigane, Jean-Jacques Royer, Hideshi Kaieda, Permeability characterization of the Soultz and Ogachi large-scale reservoir using induced microseismicity, Geophysics 67 (1) (2002) 204–211. [9] Pierre Durst, François-D Vuataz, Fluid-rock interactions in hot dry rock reservoirs. A review of the HDR sites and detailed investigations of the Soultz-sous-Forets system, in: Proceedings World Geothermal Congress, Kyushu-Tohoku, Japan, 2000. [10] Karsten Pruess, Curt Oldenburg, George Moridis, TOUGH2 User’s Guide, Version 2, California, USA, 2011. [11] Chaoshui Xu, Peter Alan Dowd, Zhao Feng Tian, A simplified coupled hydrothermal model for enhanced geothermal systems, Appl. Energy 140 (2015) 135–145. [12] Chester R. Mckee, Amar C. Bumb, Robert A. Koenig, Stress-dependent permeability and other geologic formations, SPE Formation Eval. 3 (1) (1988) 81–91. [13] Litang Hu, Philip H. Winterfeld, Perapon Fakcharoenphol, Yu-Shu Wu, A novel fully-coupled flow and geomechanics model in enhanced geothermal reservoirs, J. Petrol. Sci. Eng. 107 (3-4) (2013) 1–11. [14] Xiaoxue Huang, Jialing Zhu, Chengke Niu, Jun Li, Hu Xia, Xianpeng Jin, Heat extraction and power production forecast of a prospective Enhanced Geothermal System site in Songliao Basin, China, Energy 75 (2014) 360–370. [15] Karsten Pruess, On production behavior of enhanced geothermal systems with CO2 as working fluid, Energy Convers. Manage. 49 (6) (2008) 1446–1454. [16] C. Coskun, Z. Oktay, I. Dincer, Performance evaluations of a geothermal power plant, Appl. Therm. Eng. 31 (17–18) (2011) 4074–4082. [17] Roy Baria, Jörg Baumgärtner, Fritz Rummel, Robert J. Pine, Yoshiteru Sato, HDR/HWR reservoirs: concepts, understanding and creation, Geothermics 28 (4-5) (1999) 533–552. [18] Qing Gao, Xue-Zhi Zhou, Yan Jiang, Xiang-Liang Chen, Yu-Ying Yan, Numerical simulation of the thermal interaction between pumping and injecting well groups, Appl. Therm. Eng. 51 (1–2) (2013) 10–19.