Petroleum Exploration & Development, 2020, 47(6): 1307-1315 doi: 10.1016/S1876-3804(20)60138-6

Prediction of oil recovery in naturally fractured reservoirs subjected to reinfiltration during gravity drainage using a new scaling equation

AGHABARARI Amirhossein1,2, GHAEDI Mojtaba,1,2,*, RIAZI Masoud1

1. Department of Petroleum Engineering, School of Chemical and Petroleum Engineering, P.O. Box 7193616511, Shiraz University, Shiraz, Iran

2. Reservoir Modeling and Simulation Center (RMSC), P.O. Box 7193616511, Shiraz University, Shiraz, Iran

Corresponding authors: * E-mail: m.ghaedi@shirazu.ac.ir

Received: 2020-04-6   Online: 2020-12-20

Abstract

By comparing numerical simulation results of single-porosity and dual-porosity models, the significant effect of reinfiltration to naturally fractured reservoirs was confirmed. A new governing equation was proposed for oil drainage in a matrix block under the reinfiltration process. Utilizing inspectional analysis, a dimensionless equation suitable for scaling of recovery curves for matrix blocks under reinfiltration has been obtained. By the design of experiments, test cases with different rock and fluid properties were defined to confirm the scope of the presented equation. The defined cases were simulated using a realistic numerical simulation approach. This method can estimate the oil amount getting into the matrix block through reinfiltration, help simulate the oil drainage process in naturally fractured reservoirs accurately, and predict the recovery rate of matrix block in the early to middle periods of production. Using the defined scaling equation in the dual-porosity model can improve the accuracy of the predicted recovery rate.

Keywords: naturally fractured reservoir ; gravity drainage ; reinfiltration ; scaling equation ; dual-porosity simulation ; inspectional analysis

PDF (1348KB) Metadata Metrics Related articles Export EndNote| Ris| Bibtex  Favorite

Cite this article

AGHABARARI Amirhossein, GHAEDI Mojtaba, RIAZI Masoud. Prediction of oil recovery in naturally fractured reservoirs subjected to reinfiltration during gravity drainage using a new scaling equation. [J], 2020, 47(6): 1307-1315 doi:10.1016/S1876-3804(20)60138-6

Introduction

The naturally fractured reservoir can be simplified as a system combined by matrix blocks and fractures[1]. For simplicity, the matrix blocks are considered as cuboids of different dimensions. The shapes and dimensions of the matrix blocks depend on the density and direction of fractures[2,3].

As the production of a fractured reservoir goes on, the gas-oil contact in it starts dropping. The gas with high mobility overtakes the oil in the matrix blocks through the fractures and traps the oil in the matrix blocks. Due to gravity, oil would drain out from a matrix block. But under the joint effect of gravity and capillary force, the oil would re-infiltrate into the matrix block below the former matrix block[4,5,6,7,8].

Interaction between matrix blocks can give rise to reinfiltration and capillary continuity[9]. The reinfiltration is the reimbibition of oil from the fracture into the matrix block. The capillary continuity occurs when oil moves from one matrix block to another through a liquid bridge between fractures[10].

Aspenes et al. conducted a series of experiments on the capillary continuity mechanism. They found that in the case of noticeable capillary continuity, oil movement towards producing well took place in the matrix block network instead of the high permeable fractures[11]. Labastie claimed that the oil recovery of naturally fractured reservoirs strongly depended on capillary continuity[12]. Horie et al. evaluated the effect of capillary continuity on the recovery of a stack of matrix blocks through some experiments[13]. Sajadian et al. introduced the critical fracture aperture to characterize the effective capillary continuity[14]. Miri et al. developed a mathematical model to predict the critical fracture aperture[15]. Harimi et al. studied the effects of fracture roughness and frequency on the capillary continuity theoretically[16]. Mashayekhizadeh et al. investigated the stability of liquid bridges formed between fractures at different fracture apertures and dip angles by some experiments using glass micromodels[17,18]. Stones et al. found through experiments that a reduction of the liquid surface tension would reduce the effect of fracture on the ultimate recovery[19]. Dejam et al. developed a mechanistic model for formation, growth, and detachment of horizontal liquid bridges between matrix blocks[20,21] and proposed two dimensionless parameters to describe the shape of the liquid bridge surface[21].

The reinfiltration would slow down the oil movement in the fractured reservoirs and cause a delay in production[22]. Festoy and Van Golf-Racht carried out a fine grid simulation and proved the presence of reinfiltration between two stacked matrix blocks[23]. In a theoretical study, Firoozabadi and Ishimoto found that the matrix block reinfiltration rate was always more than or equal to the drainage rate[24] and the reinfiltration rate across vertical faces of the matrix block varied significantly[25]. Moreover, a grid simulation was performed to find out the slope effect on the reinfiltration[26]. Firoozabadi et al. studied the impact of fracture aperture on the rate of oil drainage across stacked matrix blocks by experiments[27]. Sajjadian et al. sorted out the main parameters affecting the reinfiltration process[28]. Aghabarari et al. numerically simulated the reinfiltration process in 3 stacked matrix blocks to find the effect of different reservoir properties on reinfiltration[29]. After some experiments and numerical simulations, Mollaei et al. related the reinfiltration rate to the horizontal fracture width and the size of the matrix pore-throat[30]. More studies about the reinfiltration and capillary continuity can be found in the literature[31,32,33,34].

In summary, the interaction between matrix blocks has two different effects: oil reinfiltration in the naturally fractured reservoir would make the oil production rate decrease, and capillary continuity would make the ultimate oil recovery increase. As there is no method to measure the quantity of reinfiltrated oil into the matrix blocks, the simulation results of naturally fractured reservoirs would have large errors. This study aims to work out and minimize this error.

Por et al. developed a new simulator to consider the interaction between blocks by establishing an extra connection between the fractured medium and matrix blocks[35]. By constructing the drainage curves of three stacked blocks, Tan and Firoozabadi worked out a reservoir simulation method considering the interaction between blocks in fractured reservoir[36]. Donato et al. examined oil recovery by gravity drainage using the one-dimensional analytical and numerical analysis[37]. They also proposed the analytical equation for oil recovery as a function of time. Fung put forward a new method to handle the interaction between blocks in the field scale simulation of the naturally fractured reservoir, which involved the calculation of pseudo capillary potential[38]. De Guevara-Torres et al. modified the dual-porosity formula to include reinfiltration in the simulation of naturally fractured reservoir[39].

The scaling equation is an effective method to predict the production of naturally fractured reservoirs. Scaling equations were built for the imbibition process before[40,41,42,43,44,45,46,47,48,49,50]. But to the best of our knowledge, the scaling equation has not been used to study the reinfiltration phenomenon yet.

In this work, by using the inspectional analysis approach, a new scaling equation has been proposed for the gravity drainage process affected by reinfiltration. Then, experiments for cases with different rocks and fluid properties were designed, a robust simulation approach was used to mimic the drainage process with reinfiltration in the defined cases. The results show the presented scaling equation can predict the oil recovery factors under the conditions mentioned above.

1. Model building

The numerical simulator (black-oil simulator ECLIPSE100[51]) was used to study the gravity drainage process affected by reinfiltration, and the effects of different parameters such as rock and fluid properties, dimensions of matrix blocks, porosity, permeability, etc. on reinfiltration.

Single-porosity approach is an effective method to evaluate the production mechanisms in naturally fractured reservoirs. In this method, fractures and matrix blocks are defined explicitly by fine grids[52,53]. It should be mentioned that although the single-porosity approach can predict and describe production mechanisms accurately, it takes long computation time, so this approach isn’t practical for large scale field simulation. The single-porosity approach was used to build a model of fractured reservoir in this work, as this approach enables us to have an accurate and detailed description of the fluid flow in the fractured reservoir (including gravity drainage and reinfiltration).

The model had three matrix blocks and fractures (Fig. 1). The three matrix blocks are stacked over each other and each of them is completely surrounded by fractures to simulate naturally fractured reservoirs. To provide a better insight into the model, the front plane of the fractured medium is removed in Fig. 1. The matrix blocks are separated by the fracture horizontal planes. Through grid sensitivity analysis, we chose a 20×20×20 grid resolution in X, Y, and Z directions for each matrix block. Therefore, the whole model consisted of 22×22×64 grids. The grid of the matrix block was 0.762 m×0.762 m×0.762 m, and the fracture grid was 97.5μm in the aperture direction and 0.762 m×0.762 m in the other two directions. Fig. 2 and Fig. 3 present relative permeability curves of the matrix blocks and fracture medium at different oil saturations respectively. In the experiments, the oil, water, and gas had a density of 777.45, 999.97, and 1.73 kg/m3 respectively under standard state, and a viscosity of 0.574, 0.310, 0.042 mPa•s at 27.56 MPa and 82 °C. The matrix block and fractures had a porosity of 10% and 100%, and permeability of 5×10-3, 7 000×10-3 μm2, respectively.

Fig. 1.

Fig. 1.   Schematic of the stacked matrix blocks.


Fig. 2.

Fig. 2.   Oil and gas relative permeabilities in the matrix blocks.


Fig. 3.

Fig. 3.   Oil and gas relative permeabilities in the fractured medium.


Initially, all the three matrix blocks were saturated with oil, while the fractures were saturated with gas. Also, as the fluid phases in the matrix and fracture media changed, the pressure gradient between them changed accordingly. It should be mentioned that despite the fact that the pore volume of fractures could be negligible in comparison with that of the matrix blocks, as these fractures were connected to the vast volume of the gas cap, the fracture network was assigned to a certain pore volume. This way, the fractures in the model were saturated with gas all the time with no need of gas injection well.

Given the fact that capillary pressure in the fractures is much smaller than that in matrix blocks, the capillary pressure of the fracture is generally regarded as zero. Moreover, in this work, it was assumed that the oil and gas relative permeabilities and the corresponding saturation had a linear relation; the residual saturations of all phases in the fractured medium were fixed at zero.

2. Importance of reinfiltration effects

The most common approach for field-scale simulation of a naturally fractured reservoir is the dual-porosity model[54]. The dual-porosity model was first introduced to the petroleum industry by Warren and Root[55]. It is lower in computational cost than the single-porosity method. In the dual-porosity approach, the oil flow between the fracture and matrix block is defined by a matrix-fracture transfer function. As the effect of oil reinfiltration from fracture to the matrix block is not completely considered in the dual-porosity approach[31], in this work, we tried to make use of the single-porosity approach to find out a way for estimating the reinfiltration rate or drainage rate of a matrix block under the effect of reinfiltration. The result of this work can improve the accuracy of the dual-porosity approach too.

A dual-porosity test plan was made exactly similar to the abovementioned model in the last section to find the differences between single-porosity and dual-porosity models. All of the needed properties for simulation such as rock and fluid properties, dimensions of matrix blocks, porosity, permeability, etc. were the same in both approaches. Fig. 4 shows the outcomes of the simulation for the test plans. The total oil drained from the stacked matrix blocks 10 years after production were simulated with the two models. The results show the single-porosity model can simulate the gravity drainage process in the naturally fractured reservoir precisely[31,52], while the dual-porosity approach overestimated the oil production rate. In the dual-porosity model not considering the reinfiltration effect, the drained oil from one matrix block entered into the fracture, and no oil entered the matrix block below. In contrast, in the single-porosity model, the oil drained from one matrix block may reinfiltrate into the other matrix blocks below. Ignoring the reinfiltration between matrix blocks is one of the main reasons for the inaccuracy in the simulation result of the dual-porosity model. This finding is in agreement with the results of other researchers[32,35].

Fig. 4.

Fig. 4.   Total oil production predicted by single-porosity and dual-porosity models.


3. New scaling equation

The inspectional analysis was used to find a new scaling equation[56,57,58,59,60]. First, one needs to determine the governing equation of flow for the reinfiltration mechanism, in which all parameters affecting the fractured reservoir such as permeability-porosity, matrix block cross-sectional area, rock type, fluid type, and matrix block height[29], as well as the effect of capillary pressure, gravity, and viscous forces must be taken into account. Thus, the one-dimensional flow equation in multiple phase flow for oil which is the fundamental equation in reservoir simulation[61] was used in this study:

$\frac{\partial }{{\partial Z}}\left[ {{K_Z}{A_Z}\frac{{{K_{ro}}}}{{{\mu _o}{B_o}}}\left( {\frac{{\partial {p_o}}}{{\partial Z}} + {\rho _o}g} \right)} \right]\Delta Z = {V_b}\frac{\partial }{{\partial t}}\left( {\frac{{\phi {S_o}}}{{{B_o}}}} \right) - {q_{osc}}$

Equation 1 is the basic equation for an element of the reservoir, where qosc represents the oil quantity flowing in or out of an element, and it is positive or negative for production or injection wells, respectively. Here we suggest using Equation 1 to characterize the matrix block. In this way, the element of the reservoir could be extended to a single matrix block, and all of the parameters related to each matrix block could be defined. The oil reinfiltration or drainage into or out of matrix blocks can happen in an injection or production well accordingly. When a matrix block is affected by drainage and reinfiltration simultaneously, qosc can be defined as the joint result of injection and production wells. With the Equation 2 presented in reference [62], Equation 3 can be derived from Equation 1:

$\frac{{\partial {p_o}}}{{\partial Z}} = - {\rho _g}g - \frac{{\partial {p_c}}}{{\partial Z}}$
$- \frac{\partial }{{\partial Z}}\left[ {{K_Z}{A_Z}\frac{{{K_{ro}}}}{{{\mu _o}}}\left( {\frac{{\partial {p_c}}}{{\partial {S_o}}}\frac{{\partial {S_o}}}{{\partial Z}} - \Delta \rho g} \right)} \right]\Delta Z = {V_b}\frac{\partial }{{\partial t}}\left( {\phi {S_o}} \right) - {q_o}$

As Equation 3 has been taken into account in the model of this work, we could rewrite the original Eclipse 100 model into Equation 4 to represent the models with different sizes, rock, and fluid types.

$- \frac{\partial }{{\partial Z'}}\left[ {{{K'}_Z}{A_Z}^\prime \frac{{{{K'}_{ro}}}}{{{{\mu'}_o}}}\left( {\frac{{\partial {{p'}_c}}}{{\partial {S_o}}}\frac{{\partial {S_o}}}{{\partial Z'}} - \Delta {{\rho '}_{}}g'} \right)} \right] \Delta Z' = {V_b}^\prime \frac{\partial }{{\partial t'}}\left( {\phi '{S_o}} \right) - {q'_o}$

By setting the ratio of the model to the original model as $Z/Z'=h, K_z/K'_z=D, A_z/A'_z=L^2, K_{ro}/K'_{ro}=M, \mu_o/ \mu'_o=F, \lgroup \frac{\partial P_c}{\partial S_o}/ \frac{\partial P'_c}{\partial S_o}=G\rgroup, \Delta \rho g/ \Delta \rho' g'=T, \Delta Z/ \Delta Z'=H, \phi/ \phi'=c, q_o/q'_o=B$, and substituting them into Equation 3, the next equation is obtained:

$- \frac{\partial }{{h\partial Z'}}\left[ {\frac{{D{L^2}M}}{F}{{K'}_Z}{A_Z}^\prime \frac{{{{K'}_{ro}}}}{{{{\mu '}_o}}}\left( {\frac{G}{h}\frac{{\partial {{p'}_{\text{c}}}}}{{\partial {S_o}}}\frac{{\partial {S_o}}}{{\partial Z'}} - T\Delta {{\rho '}_{}}g'} \right)} \right]H\Delta Z' = H{L^2}\frac{{BC}}{{hC{L^2}}}{V_b}^\prime \frac{\partial }{{\partial t'}}\left( {\phi '{S_o}} \right) - B{q'_o}$

Multiplying Equation 5 by h/HB, we have:

$- \frac{\partial }{{\partial Z'}}\left[ {\frac{{D{L^2}M}}{{FB}}{{K'}_Z}{A_Z}^\prime \frac{{{{K'}_{ro}}}}{{{{\mu '}_o}}}\left( {\frac{G}{h}\frac{{\partial {{p'}_c}}}{{\partial {S_o}}}\frac{{\partial {S_o}}}{{\partial Z'}} - T\Delta {{\rho '}_{}}g'} \right)} \right]\Delta Z' = {V_{\text{b}}}^\prime \frac{\partial }{{\partial t'}}\left( {\phi '{S_o}} \right) - \frac{h}{H}{q'_o}$

Comparing Equation 6 with Equation 4, one can deduce that:

$\frac{{D{L^2}MG}}{{hFB}} = 1$
$\frac{{D{L^2}MT}}{{FB}} = 1$
$\frac{h}{H} = 1$

Reusing the aforementioned relations results in the following dimensionless parameters:

${t_{D,c}} = \frac{{K{K_{ro}}t\frac{{\partial {p_c}}}{{\partial {S_o}}}}}{{{\mu _o}\phi \Delta {Z^2}}}$
${t_{D,gr}} = \frac{{K{K_{ro}}t\Delta \rho g}}{{{\mu _o}\phi \Delta Z}}$

Finally, the dimensionless time shown by Equation 12 was obtained, which is suitable for scaling the drainage rate of a matrix block affected by reinfiltration.

${t_D} = t{}_{D,c} + {t_{D,gr}} = \frac{{{K_Z}{K_{ro}}t\frac{{\partial {p_c}}}{{\partial {S_o}}}}}{{{\mu _o}\phi \Delta {Z^2}}} + \frac{{{K_Z}{K_{ro}}t\Delta \rho g}}{{{\mu _o}\phi \Delta Z}}$

4. Results and discussion

Through a numerical simulation, Aghabarari et al. sorted out the parameters affecting reinfiltration. Theses parameters in descending order of importance are permeability, porosity, matrix block cross-sectional area, rock type, fluid type, and matrix block height[29]. To check the applicability of the introduced dimensionless equation (Equation 12), various cases were defined, and the ranges of the parameters were defined to characterize different types of naturally fractured reservoirs. By use of Corey correlation of oil and gas[63], three types of rocks (i.e. Types I, II, and III) were defined. These rock types have different connate water saturations, capillary pressures, and relative permeability curves according to Fig. 5 and Fig. 6. Rock type I has the lowest oil wettability while rock type III is highly oil-wet. The characteristics of fluids in the three types of reservoirs are shown in Table 1. Here oils are defined as type I to III in ascending order of density.

Fig. 5.

Fig. 5.   Oil and gas capillary pressure (PCOG) curves of rock types I, II, and III.


Fig. 6.

Fig. 6.   Oil and gas relative permeability curves of rock types I, II, and III.


Table 1   Properties of different types of fluid at different pressures.

Fluid typeρo/
(kg•m-3)
ρg/
(kg•m-3)
p/
(MPa)
µo/
(mPa•s)
µg/
(mPa•s)
I777.451.621.590.390.01
20.680.530.03
34.470.610.05
II810.291.761.590.650.01
20.680.860.03
34.470.990.04
III835.921.761.590.980.01
20.681.300.03
34.471.500.05

Note: the density is surface density of oil and gas, the temperature is 82 °C

New window| CSV


These parameters were combined with different reservoir conditions to define different experiment cases. Hence the “Taguchi experiment design” was employed in planning test cases to test all combinations of the parameters with the least number of tests (Table 2). Fig. 7 shows the ratio of produced oil to the ultimate recovery of the second matrix block in 27 test cases versus time.

Table 2   Defined cases and model parameters.

CaseRock typeFluid typeMatrix block height/mMatrix block cross-sectional area/m2Matrix permeability/
10-3 μm2
Poro-
sity/%
1II9.1483.61510
2II9.1483.612020
3II9.1483.618030
4III18.29334.45510
5III18.29334.452020
6III18.29334.458030
7IIII27.43752.51510
8IIII27.43752.512020
9IIII27.43752.518030
10III18.29752.51520
11III18.29752.512030
12III18.29752.518010
13IIII27.4383.61520
14IIII27.4383.612030
15IIII27.4383.618010
16IIIII9.14334.45520
17IIIII9.14334.452030
18IIIII9.14334.458010
19IIII27.43334.45530
20IIII27.43334.452010
21IIII27.43334.458020
22IIIII9.14752.51530
23IIIII9.14752.512010
24IIIII9.14752.518020
25IIIIII18.2983.61530
26IIIIII18.2983.612010
27IIIIII18.2983.618020

New window| CSV


Fig. 7.

Fig. 7.   The ratio of produced oil to ultimate recovery for the second matrix block in test cases versus time.


According to Li and Horne, there are various ways to establish a scaling equation for recovery[64]. In one definition, recovery can be expressed in terms of recoverable reserves (Figs. 8-14, Qo/Qo,max), while in another one, it is represented in terms of initial oil in place (Figs. 15-16, Qo/N). Fig. 7 shows the recoveries of the defined cases in Table 2 based on the ultimate recovery of each test. Fig. 8 shows the variation of recovery with dimensionless time calculated by using Equation 12 for the test cases in Fig. 7. The changes of slopes in the middle sections of curves in Fig. 7 are caused by the reinfiltration effect on the second matrix block. When the oil drainage rate of the second matrix block decreases, the reinfiltration of the oil drained out from the above matrix block makes the oil drainage rate of this block rise again. The saturation related parameters in Equation 12, particularly $\partial {p_c}/ \partial {S_o}$ and Kro vary widely with time, and will lower the scaling equation performance. It is confirmed through this study that the scaling equation reaches the best effect when pc and Kro reach the maximum values. In addition, separating the test cases into 3 groups according to the rock types can eliminate the dependency of the scaling equation on the parameters mentioned above and get a better scaling effect. The curves of oil recovery with time after groups according to the rock types are shown in Figs. 9-11. Figs. 12-14 show the recovery curves calculated by using Equation 12 for the 3 groups of cases according to rock types.

Fig. 8.

Fig. 8.   The ratio of produced oil to ultimate recovery for the second matrix block in test cases versus the proposed dimensionless time.


Fig. 9.

Fig. 9.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type I versus time.


Fig. 10.

Fig. 10.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type II versus time.


Fig. 11.

Fig. 11.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type III versus time.


Fig. 12 to Fig. 14 show that the scaling equation proposed in this paper is more suitable for the prediction of oil reinfiltration in the early to middle stages of depletion development. It is worth noting that all of the simulations in this section were done for more than 100 years to show the whole production time.

Fig. 12.

Fig. 12.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type I versus the proposed dimensionless time.


Fig. 13.

Fig. 13.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type II versus the proposed dimensionless time.


Fig. 14.

Fig. 14.   The ratio of produced oil to ultimate recovery for the second matrix block in rock type III versus the proposed dimensionless time.


Fig. 15 shows the recovery curves of all the test cases according to the initial oil in place (IOIP), and Fig. 16 shows the scaled curves obtained by using Equation 12. Since different types of rocks were used in this study, the curves show different levels of oil depletion. The initial oil in place can be calculated by using gas saturation and porosity before simulation. Compared with Fig. 8, the scaling equation for recovery based on the IOIP not only improves the scaling quality of all rock types but also is more practical. The curves in Fig. 15 can be unified into one curve approximately by using Equation 12 (Fig.16). By using the scaling equation presented in this paper, the recovery of matrix blocks with different rock and fluid types, porosities, permeabilities, cross-section areas, and height can be scaled and predicted.

Fig. 15.

Fig. 15.   The ratio of produced oil to initial oil in place (IOIP) for the second matrix block in test cases versus time.


Fig. 16.

Fig. 16.   The ratio of produced oil to IOIP for the second matrix block in test cases versus the proposed dimensionless time.


5. Conclusions

Reinfiltration significantly affects the production behavior of matrix block during the gravity drainage. The oil production of stacked matrix blocks simulated by dual-porosity and single-porosity approaches had a significant difference, proving the importance of considering reinfiltration in the simulation. A new governing equation has been proposed for the oil drainage of a matrix block under the reinfiltration effect. With this equation, a new scaling equation has been obtained by inspectional analysis. Different cases with different properties were defined to verify the applicability of the new scaling equation. The study results show that the new scaling equation is more suitable for the recovery defined based on the IOIP and has the best results in simulating the early to middle stage of depletion development.

The findings of this work can improve the accuracy of modeling and simulation of the drainage process in naturally fractured reservoirs. The new scaling equation presented can be used in the numerical simulation of naturally fractured reservoir. It can be used to calculate the amount of reinfiltrated oil into the matrix block in the gas invaded zone more accurately and predict the oil recovery of matrix block under the effect of reinfiltration. In addition, this equation can be used in the dual-porosity simulation to increase the accuracy of the results.

Nomenclature

AZ—area of the grid perpendicular to the flow direction Z, m2;

Bo—oil formation volume factor;

h, D, L, M, F, G, T, H, c, B—ratios of the model parameters to the original model parameters;

g—gravity acceleration, m/s2;

Kr—relative permeability;

Krg—gas relative permeability;

Kro—oil relative permeability;

KZ—permeability in Z direction, m2;

N—oil geologic reserves, m3;

p—pressure, MPa;

pc—capillary pressure, Pa;

po—oil pressure, Pa;

Qo—oil production of the second matrix block during reinfiltration process, m3;

Qo,max—recoverable reserves of the second matrix block, m3;

qo—oil influx of the reservoir element under reservoir conditions, m3/s;

qosc—oil influx of the reservoir element under surface conditions, m3/s;

So—oil saturation, %;

t—time, s;

tD—dimensionless time;

tD,c—dimensionless time in the case with capillary pressure taking dominance;

tD,gr—dimensionless time in the case with gravity taking dominance;

Vb—total volume of the elements, m3;

X, Y, Z—3 coordinate directions of the model, m;

ΔZ—grid height, m;

Δρ—difference of oil and gas densities, kg/m3;

μo—oil viscosity, Pa•s;

ρg—gas density, kg/m3;

ρo—oil density, kg/m3;

ϕ—porosity, %.

Superscript:

'—parameter of the original model.

Reference

SABOORIAN-JOOYBARI H, ASHOORI S, MOWAZI G.

A new transient matrix/fracture shape factor for capillary and gravity imbibition in fractured reservoirs

Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2015,37(23):2497-2506.

[Cited within: 1]

SAIDI A M.

Reservoir engineering of fractured reservoirs: Fundamental and practical aspects

Paris: Total Edition Press, 1987.

[Cited within: 1]

van GOLF-RACHT T D. Fundamentals of fractured reservoir engineering. Amsterdam, the Netherlands: ElsevierScientific Publishing Company, 1982.

[Cited within: 1]

ZOU Caineng, TAO Shizhen, YUAN Xuanjun, et al.

Global importance of “continuous” petroleum reservoirs: Accumulation, distribution and evaluation

Petroleum Exploration and Development, 2009,36(6):669-682.

[Cited within: 1]

ZHOU You, LU Teng, WU Shouya, et al.

Models of steam-assisted gravity drainage (SAGD) steam chamber expanding velocity in double horizontal wells and its application

Petroleum Exploration and Development, 2019,46(2):334-341.

[Cited within: 1]

SAIDI A M, TEHRANI D H, WIT K.

Mathematical simulation of fractured reservoir performance, based on physical model experiments

WPC 18227, 1979.

[Cited within: 1]

ZENDEHBOUDI S, MOHAMMADZADEH O, CHATZIS I.

Investigation of gravity drainage in fractured porous media using rectangular macromodels

Calgary, Alberta: Canadian International Petroleum Conference, 2008.

[Cited within: 1]

SHARIAT A, BEHBAHANINIA A R, BEIGY M R.

Block to block interaction effect in naturally fractured reservoirs

SPE 101733, 2006.

[Cited within: 1]

DEJAM M, HASSANZADEH H, CHEN Z.

Reinfiltration through liquid bridges formed between two matrix blocks in fractured rocks

Journal of Hydrology, 2014,519:3520-3530.

[Cited within: 1]

RAHMATI N, RASAEI M R.

Quantifying the reimbibition effect on the performance of gas-oil gravity drainage in fractured reservoirs: Mathematical modelling

The Canadian Journal of Chemical Engineering, 2019,97(S1):1718-1728.

[Cited within: 1]

ASPENES E, ERSLAND G, GRAUE A, et al.

Wetting phase bridges establish capillary continuity across open fractures and increase oil recovery in mixed-wet fractured chalk

Transport in Porous Media, 2008,74(1):35-47.

[Cited within: 1]

LABASTIE A.

Capillary continuity between blocks of a fractured reservoir

SPE 20515, 1990.

[Cited within: 1]

HORIE T, FIROOZABADL A, ISHIMOTO K.

Laboratory studies of capillary interaction in fracture/matrix systems

SPE 18282, 1990.

[Cited within: 1]

SAJADIAN V A, DANESH A, TEHRANI D H.

Laboratory studies of gravity drainage mechanism in fractured carbonate reservoir: Capillary continuity

SPE 49497, 1998.

[Cited within: 1]

MIRI R, SHADIZADEH S R, KHARRAT R.

Fracture capillary pressure based on the liquid bridge dynamic stability study

Energy Sources, Part A: Recovery, Utilization and Environmental Effects, 2014,36(23):2536-2545.

[Cited within: 1]

HARIMI B, GHAZANFARI M H, MASIHI M.

Modeling of capillary pressure in horizontal rough-walled fractures in the presence of liquid bridges

Journal of Petroleum Science and Engineering, 2020,185:106642.

[Cited within: 1]

MASHAYEKHIZADEH V, GHAZANFARI M H, KHARRAT R, et al.

Pore-level observation of free gravity drainage of oil in fractured porous media

Transport in Porous Media, 2011,87:561-584.

[Cited within: 1]

MASHAYEKHIZADEH V, KHARRAT R, GHAZANFARI M H, et al.

An experimental investigation of fracture tilt angle effects on frequency and stability of liquid bridges in fractured porous media

Petroleum Science and Technology, 2012,30(8):807-816.

[Cited within: 1]

STONES E J, ZIMMERMAN S A, CHIEN C V, et al.

The effect of capillary connectivity across horizontal fractures on gravity drainage from fractured porous media

SPE 24920, 1992.

[Cited within: 1]

DEJAM M, HASSANZADEH H.

Formation of liquid bridges between porous matrix blocks

AIChE Journal, 2011,57(2):286-298.

[Cited within: 1]

DEJAM M, HASSANZADEH H, CHEN Z.

Shape of liquid bridges in a horizontal fracture

Journal of Fluid Flow, Heat and Mass Transfer, 2014,1:1-8.

[Cited within: 2]

FAMIAN S R, MASIHI M.

Special features of fracture network in Iranian fractured reservoirs

Nafta, 2010,61(1):39-47.

[Cited within: 1]

FESTOY S, van GOLF-RACHT T D.

Gas gravity drainage in fractured reservoirs through new dual-continuum approach

SPE 26980, 1989.

[Cited within: 1]

FIROOZABADI A, ISHIMOTO K.

Reinfiltration in fractured porous media: Part 1-One dimensional model

SPE 21796, 1994.

[Cited within: 1]

FIROOZABADI A, ISHIMOTO K, DINDORUK B.

Reinfiltration in fractured porous media: Part 2-Two dimensional model

SPE 21798, 1994.

[Cited within: 1]

COATS K H.

Implicit compositional simulation of single-porosity and dual-porosity reservoirs

SPE 18427, 1989.

[Cited within: 1]

FIROOZABADI A, MARKESET T.

Fracture-liquid transmissibility in fractured porous media

SPE 24919, 1994.

[Cited within: 1]

SAJJADIAN V A, DANESH A, TEHRANI D H.

Laboratory study of gravity drainage mechanism in fractured carbonate reservoir- reinfiltration

SPE 54003, 1999.

[Cited within: 1]

AGHABARARI A, GHAEDI M.

A new simulation approach to investigate reinfiltration phenomenon in fractured porous media

Journal of Petroleum Science and Engineering, 2019,181:106182.

[Cited within: 3]

MOLLAEI A, HAGHIGHI M, MAINI B.

Free-fall gravity drainage in fractured matrix blocks: Experimental and network modeling simulation findings and observations

SPE 107206, 2007.

[Cited within: 1]

SU S, GOSSELIN O, PARVIZI H, et al.

Dynamic matrix-fracture transfer behavior in dual-porosity models

SPE 164855, 2013.

[Cited within: 3]

LEMONNIER P, BOURBIAUX B.

Simulation ofnaturally fractured reserviors. State of the art: Part 1-Physical mechanisms and simulator formulation

Oil and Gas Science and Technology, 2010,65(2):239-262.

[Cited within: 2]

ZENDEHBOUDI S, CHATZIS I, SHAFIEI A, et al.

Empirical modeling of gravity drainage in fractured porous media

Energy and Fuels, 2011,25(3):1229-1241.

[Cited within: 1]

ZOBEIDI K, FASSIHI M R.

Block to block interactions and their effects on miscibility gravity drainage in fractured carbonate reservoirs, experimental and analytical results

Journal of Petroleum Science and Engineering, 2018,164:696-708.

DOI:10.1016/j.petrol.2018.01.015      URL     [Cited within: 1]

POR G J, BOERRIGTER P, MAAS J G, et al.

Fractured reservoir simulator capable of modeling block-block interaction

SPE 19807, 1989.

[Cited within: 2]

TAN J C T, FIROOZABADI A.

Dual-porosity simulation incorporating reinfiltration and capillary continuity concepts part I: Single gridcell

SPE 29113, 1995.

[Cited within: 1]

di DONATO G, TAVASSOLI Z, BLUNT M J.

Analytical and numerical analysis of oil recovery by gravity drainage

Journal of Petroleum Science and Engineering, 2006,54(1/2):55-69.

DOI:10.1016/j.petrol.2006.08.002      URL     [Cited within: 1]

di DONATO G, FUNG L S K.

Simulation of block-to-block processes in naturally fractured reservoirs

SPE 20019, 1991.

[Cited within: 1]

de GUEVARA-TORRES J E L, de la GARZA R F, GALINDO-NAVA A.

Gravity-drainage and oil-reinfiltration modeling in naturally fractured reservoir simulation

SPE 108681, 2009.

[Cited within: 1]

LI K, HORNE R N.

An analytical scaling method for spontaneous imbibition in gas/water/rock systems

SPE 88996, 2004.

[Cited within: 1]

MA S, MORROW N R, ZHANG X.

Generalized scaling of spontaneous imbibition data for strongly water-wet systems

Journal of Petroleum Science and Engineering, 1997,18(3/4):165-178.

DOI:10.1016/S0920-4105(97)00020-X      URL     [Cited within: 1]

SCHMID K S, GEIGER S.

Universal scaling of spontaneous imbibition for water-wet systems

Water Resources Research, 2012,48(3):W03507.

[Cited within: 1]

MASON G, FISCHER H, MORROW N R, et al.

Correlation for the effect of fluid viscosities on counter-current spontaneous imbibition

Journal Petroleum Science and Engineering, 2010,72(1/2):195-205.

DOI:10.1016/j.petrol.2010.03.017      URL     [Cited within: 1]

MASON G, FISCHER H, MORROW N R, et al.

Oil production by spontaneous imbibition from sandstone and chalk cylindrical cores with two ends open

Energy and Fuels, 2010,24(2):1164-1169.

DOI:10.1021/ef901118n      URL     [Cited within: 1]

ZHOU D, JIA L, KAMATH J, et al.

Scaling of counter-current imbibition processes in low-permeability porous media

Journal of Petroleum Science and Engineering, 2002,33(1/3):61-74.

DOI:10.1016/S0920-4105(01)00176-0      URL     [Cited within: 1]

XIE X, MORROW N R.

Oil recovery by spontaneous imbibition from weakly water-wet rocks

Petrophysics, 2001,42:313-322.

[Cited within: 1]

LI K, HORNE R N.

Generalized scaling approach for spontaneous imbibition: An analytical model

SPE 77544, 2006.

[Cited within: 1]

ABBASI J, GHAEDI M, RIAZI M.

Discussion on similarity of recovery curves in scaling of imbibition process in fractured porous media

Journal of Natural Gas Science and Engineering, 2016,36:617-629.

DOI:10.1016/j.jngse.2016.11.017      URL     [Cited within: 1]

ABBASI J, RIAZI M, GHAEDI M, et al.

Modified shape factor incorporating gravity effects for scaling countercurrent imbibition

Journal of Petroleum Science and Engineering, 2017,150:108-114.

DOI:10.1016/j.petrol.2016.11.037      URL     [Cited within: 1]

MIRZAEI-PAIAMAN A, KORD S, HAMIDPOUR E, et al.

Scaling one- and multi-dimensional co-current spontaneous imbibition processes in fractured reservoirs

Fuel, 2017,196:458-472.

DOI:10.1016/j.fuel.2017.01.120      URL     [Cited within: 1]

GeoQuest.

ECLIPSE 100 reference manual

Houston, TX: Schlumberger Geoquest, 2010.

[Cited within: 1]

ABUSHAIKHA A S, GOSSELIN O R.

Matrix-fracture transfer function in dual-medium flow simulation: Review, comparison, and validation

SPE 113890, 2008.

[Cited within: 2]

GURPINAR O, KOSSACK C A.

Realistic numerical models for fractured reservoirs

SPE 59041, 2000.

[Cited within: 1]

GHAEDI M, MASIHI M, HEINEMANN Z E, et al.

Application of the recovery curve method for evaluation of matrix-fracture interactions

Journal of Natural Gas Science and Engineering, 2015,22:447-458.

DOI:10.1016/j.jngse.2014.12.029      URL     [Cited within: 1]

WARREN J E, ROOT P J.

The behavior of naturally fractured reservoirs

SPE 426, 1963.

[Cited within: 1]

KHANDOOZI S, MALAYERI M R, RIAZI M, et al.

Inspectional and dimensional analyses for scaling of low salinity waterflooding (LSWF): From core to field scale

Journal of Petroleum Science and Engineering, 2020,189:106956.

DOI:10.1016/j.petrol.2020.106956      URL     [Cited within: 1]

MIRZAEI-PAIAMAN A, MASIHI M.

Scaling equations for oil/gas recovery from fractured porous media by counter-current spontaneous imbibition: From development to application

Energy and Fuels, 2013,27(8):4662-4676.

DOI:10.1021/ef400990p      URL     [Cited within: 1]

MIRZAEI-PAIAMAN A.

Analysis of counter-current spontaneous imbibition in presence of resistive gravity forces: Displacement characteristics and scaling

Journal of Unconventional Oil and Gas Resources, 2015,12:68-86.

DOI:10.1016/j.juogr.2015.09.001      URL     [Cited within: 1]

GHAEDI M, RIAZI M.

Scaling equation for counter current imbibition in the presence of gravity forces considering initial water saturation and SCAL properties

Journal of Natural Gas Science and Engineering, 2016,34:934-947.

DOI:10.1016/j.jngse.2016.07.066      URL     [Cited within: 1]

RAPOPORT L A.

Scaling laws for use in design and operation of water-oil flow models

Transactions of the AIME, 1955,204:143-150.

DOI:10.2118/415-G      URL     [Cited within: 1]

ERTEKIN T, ABOU-KASSEM J H, KING G R.

Basic applied reservoir simulation

Texas: Society of Petroleum Engineers, 2001.

[Cited within: 1]

LI K, HORNE R N.

Characterization of spontaneous water imbibition into gas-saturated rocks

SPE 74703, 2001.

COREY A T.

The interrelation between gas and oil relative permeabilites

Producers Monthly, 1954,19:38-41.

[Cited within: 1]

LI K, HORNE R N.

Scaling of spontaneous imbibition in gas- liquid- rock systems

SPE 75167, 2002.

[Cited within: 1]

/