Research on the Optimization of the Volume Fracturing Shut-in and Drainage System of Unconventional Reservoirs in the Erlian Block

Abstract

Aiming at unclear imbibition replacement mechanisms and flowback/production strategies in unconventional reservoirs of the Erlian Block, this study proposes a systematic approach integrating “imbibition-flowback-productivity synergy” to optimize post-fracturing shut-in and production regimes. By developing numerical models incorporating geological and engineering factors, we analyzed fluid dynamics during both the shut-in and production phases. Concurrently, crude oil displacement-fracturing fluid imbibition replacement experiments were conducted to guide parameter optimization. The results indicate that optimized shut-in time and production rates substantially increase recovery efficiency while mitigating reservoir damage and proppant flowback. The well shut-in time of the Erlian Block can achieve the optimal shut-in replacement effect in about 20–25 days. The optimized flowback rate of the unconventional reservoir in the Erlian Block is 25–30 m³/d. The findings offer theoretical insights and practical recommendations for the efficient development of unconventional resources.

1. Introduction

The target unconventional reservoir in the Wuliyasitai Sag is located in the northeastern part of the Manite Depression of the Erlian Basin. Formed on a Hercynian folded basement, it is an Early Cretaceous faulted sag with an overall NNE-trending elongated shape, divided from south to north into three secondary sub-sags: southern, central, and northern [1]. The Arxan Formation in the Wuliyasitai Sag primarily consists of lithic arkose and feldspathic litharenite with low compositional maturity, along with minor amounts of arkose and litharenite [2]. Currently, post-fracturing management in the block relies mainly on field experience, with parameters such as shut-in time, choke size, and the transition from flowback to artificial lift lacking a scientific basis. The impact of these management strategies on post-fracturing productivity has also not been systematically understood. Therefore, establishing a numerical model and simulation method for shut-in and flowback in this block, integrating imbibition experiments with field practice to reveal the patterns of imbibition replacement during shut-in and flowback, and developing optimized methods and technologies for post-fracturing shut-in and flowback protocols suitable for shale oil wells are of great significance for improving the development efficiency of unconventional reservoirs.

Although previous studies have explored post-fracturing flowback regimes to some extent, their methodologies exhibit notable limitations. Liao et al. [3] used fluid flowback dynamics to invert fracture parameters but failed to adequately account for the coupling between geological conditions and engineering operations. Yu Yangkang [4] proposed a choke control method that, while practical, relies heavily on high-quality data, limiting its broad applicability. Miao Guojing [5] developed a production scheduling system based on field statistics, yet it is only applicable during the self-flowing stage of horizontal wells and lacks lifecycle versatility. Weng Dingwei et al. [6] identified factors influencing shut-in time but did not propose a systematic optimization approach. Eltahan et al. [7] analyzed imbibition potential from a geomechanical perspective but did not integrate actual geological and fracturing parameters into a unified model. Overall, most existing studies on imbibition displacement remain confined to mechanistic investigations or single-stage simulations, failing to effectively incorporate real-world geological and operational constraints. This leads to a significant gap between research findings and field applications. Moreover, while many studies regard imbibition equilibrium time as the key factor in shut-in design, most overlook the ongoing imbibition effects during the flowback process, thereby compromising the accuracy of production forecasts.

To address these research gaps, this study integrates actual geological parameters of the Arxan Formation in the Wuliyasitai Sag with fracturing operation data to establish a geo-engineering integrated numerical model for shut-in and flowback. By integrating the actual geological model with the inverted fracture geometry and conductivity, a flow model for oil–water two-phase flow in distinct domains was constructed. This model couples capillary pressure equations and saturation equations, comprehensively achieving effective integration of geological and engineering models. Through a systematic combination of experimental studies, numerical simulations, and field data analysis, this research aims to elucidate the coupling mechanism and spatiotemporal evolution of fracturing fluid imbibition and oil displacement, clarify the impact of key parameters such as shut-in time and flowback rate on stimulation effectiveness, and develop an engineering-applicable optimization method for shut-in and production scheduling. The core innovation of this work lies in the deep integration of NMR experiments with geo-engineering numerical simulation, enabling, for the first time, quantitative characterization of imbibition displacement efficiency in the Erlian Block and providing a tailored production system optimized for this specific region.

2. Fracturing Fluid Imbibition–Crude Oil Displacement Experiment

Current NMR-based imbibition studies enable the calculation of oil exchange percentage during imbibition by analyzing the amplitude variation of T₂ spectrum signals. Furthermore, after determining the surface relaxivity, quantitative analysis of crude oil redistribution across pores of different sizes under high-pressure imbibition conditions can be conducted, characterizing the contribution of varying pore throat sizes to oil recovery through imbibition [8,9,10].

Based on NMR scanning results combined with core weighing, the imbibition displacement ratio of fracturing fluid to crude oil at different stages and the ultimate imbibition displacement volume were quantitatively analyzed. The imbibition recovery extent of the matrix core continuously increases over time. The imbibition recovery curve, serving as a key fitting parameter for numerical simulation, digitally characterizes the dynamic behavior of imbibition. This approach enables the modification of reservoir capillary pressure when establishing numerical models representative of actual reservoirs, thereby enhancing the predictive accuracy of numerical simulations.

The volume of vacuum-saturated crude oil was calculated based on the core volume and laboratory-measured porosity of the experimental samples. Combined with NMR scanning and high-precision weighing data, the volume of oil displaced by imbibition was quantitatively characterized. The calculation of the imbibition-displaced oil volume is shown in Equation (1).

$$V_d={\displaystyle \frac{\left(m_i-m_o\right)}{\left({\rho }_f-{\rho }_o\right)}}$$

(1)

where V_d represents the oil–water displacement volume, cm³; m_i denotes the core mass at different imbibition times, g; m_o indicates the core mass after saturated with crude oil, g; ρ_f refers to the density of fracturing fluid, g/cm³; and ρ_o signifies the density of crude oil, g/cm³.

Nuclear magnetic resonance (NMR) measurements were conducted using a SPEC-RC2 NMR core analyzer to determine the longitudinal (T₁), transverse (T₂), and 2D T₁-T₂ relaxation times of pore fluids in vacuum-pressurized oil-saturated core samples. A pressure-resistant vessel (50 MPa) was employed as the experimental chamber. The experimental samples were tight sandstone cores from the Aershan Formation oil reservoir in the Erlian Basin. These cores exhibit strong heterogeneity and weak cementation strength, making them susceptible to damage during spontaneous imbibition experiments. Effective samples and corresponding experimental results were obtained from a depth interval of 1770–1797 m. The cores measured 4.09–5.18 cm in length and 2.46–2.48 cm in diameter, with porosity ranging from 7.23% to 8.72% and gas-measured permeability between 0.016 × 10⁻³ μm² and 0.065 × 10⁻³ μm². The oil used in the experiments was dehydrogenated simulated oil, with a viscosity of 4.69 mPa·s and a density of 1.875 g/cm³ at 50 °C. The fracturing fluid consisted of filtered, broken gel fluid of clean fracturing fluid. Detailed experimental protocols are summarized in

Table 1. Pressurized imbibition experimental core.

Core Number	Diameter/mm	Length/mm	Sampling Depth/m	Lithology	Fracturing Fluid Type
T21-1	24.80	49.00	1796.46	Gravel	Active water
T21-2	24.60	51.20	1789.88	Gravel	Glue-breaking fracturing fluid
T21-3	24.80	51.80	1770.62	Gravel	Glue-breaking fracturing fluid

, with results presented in

Table 2. Results of core pressurized imbibition experiments.

Core Number	Diameter /mm	Length /mm	Fluid Viscosity /mPa s	T/°C	Confining Pressure /MPa	Saturated Crude Oil Volume /mL	Imbibition Volume /mL	Imbibition Rate/%
T21-1	24.80	49.00	1	70	20	1.494	0.296	19.8
T21-2	24.60	51.20	5	70	20	1.101	0.159	14.4
T21-3	24.80	51.80	5	70	20	2.076	0.244	11.8

and Figure 1, Figure 2 and Figure 3.

The reservoir in the Erlian Block exhibits strong heterogeneity, with high porosity and permeability. The imbibition rate ranges between 10% and 20%. Active water-based fracturing fluid demonstrates better imbibition performance than online fracturing fluid, indicating that a higher oil–water viscosity ratio leads to a lower imbibition rate, while a lower oil–water viscosity ratio results in a higher imbibition rate.

3. Numerical Simulation of a Shut-in Well and Drainage System

3.1. Model Assumption Conditions

According to the reservoir geology of the Erlian Block and the actual fracturing construction on site, the following assumptions are made on the model:

(1) The reservoir in the study block is composed of an unconventional reservoir matrix, artificial fracturing fractures, and secondary natural fractures, in which fluids can flow.

(2) The main fracture is a discrete vertical fracture, its wings are symmetrically distributed at both ends of the horizontal wellbore, and the fracture height is higher than the reservoir thickness [11].

(3) Secondary natural fractures are orthogonal to artificial fracturing fractures.

(4) The fluid in the reservoir is slightly compressible, and the flow state is oil–water two-phase flow.

(5) Ignore the heat exchange of the fluid seepage process, and assume that its flow is isothermal seepage [12].

The oil–water two-phase shale oil fracturing horizontal well kill well backflow model is illustrated in Figure 4.

3.2. Mathematical Model Establishment

3.2.1. Mathematical Model of Reservoir Dual Media

The internal matrix system of the reservoir has not been fractured, because it is a shale reservoir. Considering that its early seepage is a low-permeability non-Darcy seepage, the starting pressure gradient cannot be ignored [13,14,15,16], and the seepage equations are shown as (2) and (3):

Oil phase:

$$\nabla \left[{\rho }_o{\displaystyle \frac{K_m{K}_{ro}}{{\mu }_o}}(\nabla P_m-G_o)\right]+{\displaystyle \frac{\alpha {\rho }_o{K}_m{K}_o}{{\mu }_o}}\left(P_{sf}-P_m\right)={\displaystyle \frac{\partial ({\rho }_o{s}_o\varphi )}{\partial t}}$$

(2)

Water phase:

$$\nabla \left[{\rho }_w{\displaystyle \frac{K_m{K}_{rw}}{{\mu }_w}}\left(\nabla P_m-G_w\right)\right]+{\displaystyle \frac{\alpha {\rho }_w{K}_m{K}_{rw}}{{\mu }_w}}\left(P_{sf}-P_m\right)={\displaystyle \frac{\partial ({\rho }_w{s}_w\varphi )}{\partial t}}$$

(3)

where

G_o is oil phase gravity, MPa; G_w is water phase gravity, MPa; $\varphi$ is porosity, dimensionless; t is time, s; s_o is oil saturation, %; s_w is water saturation, %; K_rw is water relative permeability; K_ro is oil relative permeability; μ_o is oil viscosity, mPa·s; μ_w is water viscosity, mPa·s; ρ_o is oil density, kg/m³; ρ_w is water density, kg/m³; K_m is matrix permeability, mD; P_sf is threshold pressure gradient, MPa/m; P_m is matrix pressure, MPa; ∇ is the gradient operator; α is non-Darcy flow coefficient, dimensionless.

Differential equation of natural fracture seepage: The seepage of the microfracture system conforms to Darcy’s law, and the seepage equations are shown in Equations (4) and (5).

Oil phase:

$$\nabla [{\rho }_o{\displaystyle \frac{K_m{K}_{ro}}{{\mu }_o}}\nabla P_{sf}]+{\displaystyle \frac{\alpha {\rho }_o{K}_m{K}_o}{{\mu }_o}}(P_m-P_{sf})={\displaystyle \frac{\partial ({\rho }_o{s}_o\varphi )}{\partial t}}$$

(4)

Water phase:

$$\nabla \left[{\rho }_w{\displaystyle \frac{K_m{K}_{rw}}{{\mu }_w}}\nabla P_{sf}\right]+{\displaystyle \frac{\alpha {\rho }_w{K}_m{K}_{rw}}{{\mu }_w}}\left(P_m-P_{sf}\right)={\displaystyle \frac{\partial ({\rho }_w{s}_w\varphi )}{\partial t}}$$

(5)

Differential equation of seepage in hydraulic fractures:

$$-{\displaystyle \frac{\partial }{\partial \chi }}({\rho }_o{V}_o)-2{\displaystyle \frac{{\rho }_o{V}_{y,o}}{W}}={\displaystyle \frac{\partial }{\partial t}}({\rho }_o{S}_0\varphi )$$

(6)

In Equation (6), x is the displacement, m; V_o is the velocity of the oil phase, m³/s; V_y,o is the seepage velocity of the oil phase in the fracture, m/s; W is the width of hydraulic fractures, m; and S_o is the saturation of the oil phase in the fracture, which is not dependent.

3.2.2. Auxiliary Equations

Capillary force equation:

$$p_{\mathfrak{c}}^{\mathrm{F}}(S_{\mathfrak{w}}^{\mathrm{F}})\hspace{1em}=p_{\mathfrak{o}}^{\mathrm{F}}-p_{\mathfrak{w}}^{\mathrm{F}}$$

(7)

$$p_{\mathfrak{c}}^{\mathfrak{m}}(S_{\mathfrak{w}}^{\mathfrak{m}})\hspace{1em}=p_{\mathfrak{o}}^{\mathfrak{m}}-p_{\mathfrak{w}}^{\mathfrak{m}}$$

(8)

where $p_c^F$ is the capillary pressure of the fracture system, MPa; $p_c^m$ is the matrix system capillary pressure, MPa; $S_{ro}^F$ is the residual oil saturation of the fracture system, without cause; $S_{ro}^m$ is the residual oil saturation of the matrix system, without factor; $p_o^F$ is the fracture system oil phase pressure, MPa; $p_o^m$ is the oil phase pressure of the matrix system, MPa; $p_{ro}^F$ is the capillary pressure in the residual oil state of the fracture system, MPa; and $p_{ro}^m$ is the capillary pressure in the residual oil state of the matrix system, MPa.

Saturation equation:

$$S_{\mathrm{o}}^{\mathrm{F}}+S_{\mathrm{w}}^{\mathrm{F}}=1$$

(9)

$$S_{\mathrm{o}}^f+S_{\mathrm{w}}^{\mathrm{f}}=1$$

(10)

$$S_{\mathrm{o}}^m+S_{\mathrm{w}}^{\mathrm{m}}=1$$

(11)

where $S_o^F$ is the oil saturation of the hydraulic fracture system, without factor; $S_o^f$ is the oil saturation of the natural fracture system, without factor; $S_o^m$ is the oil saturation of the matrix system, without factor; $S_w^F$ is the aqueous saturation of the hydraulic fracture system, without factor; $S_w^f$ is the aqueous saturation of the natural fracture system, without factor; and $S_w^m$ is the aqueous saturation of the matrix system, without factor.

Stress sensitivity equation:

$${\phi }^{\mathrm{F}}\hspace{1em}={\phi }_0^{\mathrm{F}}{e}^{-C_{\mathrm{r}}^{\mathrm{F}}(p^{\mathrm{F}}-p_0^{\mathrm{F}})}$$

(12)

$${\phi }^{\mathrm{f}}\hspace{1em}={\phi }_0^{\mathrm{f}}{e}^{-C_{\mathrm{r}}^f(p^f-p_0^f)}$$

(13)

$${\phi }^{\mathrm{m}}\hspace{1em}=\left({\phi }_0^{\mathrm{m}}-{\phi }_{\mathrm{sw}}\right)e^{-C_{\mathrm{r}}^{\mathrm{m}}\left(p^{\mathrm{m}}-p_0^{\mathrm{m}}\right)}$$

(14)

where φ^F is the porosity of hydraulic fractures, without factor; φ^f is the porosity of natural fractures, without factor; φ^m is the porosity of the matrix, without factor; the superscript “F, f, m” represents hydraulic fractures, natural fractures, and matrix systems, respectively; ${\phi }_0^F$ is the initial porosity of hydraulic fractures, without factor; ${\phi }_0^f$ is the initial porosity of natural fractures, without factor; ${\phi }_0^m$ is the initial porosity of the matrix, without factor; φ_sw is the porosity of matrix bound water, without factor; $C_r^F$ is the compression coefficient of hydraulic fractures, MPa⁻¹; $C_r^f$ is the compression coefficient of natural fractures, MPa⁻¹; $C_r^m$ is the compression coefficient of the matrix system, MPa⁻¹; p^F is the current effective stress of hydraulic fractures, MPa; $p_0^F$ is the initial effective stress of hydraulic fractures, MPa; p^f is the current effective stress of natural cracks, MPa; $p_0^f$ is the initial effective stress of natural cracks, MPa; $p^{\mathrm{m}}$ is the current effective stress of the matrix system, MPa; and $p_0^m$ is the initial effective stress of the matrix system, MPa.

Compression equation:

$${\rho }_{\mathrm{w}}={\rho }_{\mathrm{w}0}{e}^{C_{\mathrm{w}}(p_{\mathrm{w}}^{\mathrm{F},\mathrm{f},\mathrm{m}-p_{\mathrm{m}0}^{\mathrm{F},\mathrm{m}})}}$$

(15)

$${\rho }_{\mathrm{o}}^{-1}={\rho }_{\mathrm{o}0}^{-1}[1+Co(p_{\mathrm{o}}^{\mathrm{F},\mathrm{f},\mathrm{m}}-p_{\mathrm{o}0}^{\mathrm{F},\mathrm{f},\mathrm{m}})]$$

(16)

where ρ_w0 is the current aqueous density, kg/m³; ρ_o0 is the current oil phase density, kg/m³; C_w is the aqueous phase compression coefficient; C_o is the oil phase compression coefficient; $p_w^{F,f,m}$ are the current compressive forces of the aqueous phase in different media, MPa; and the superscripts F, f, and m represent hydraulic fractures, natural fractures, and matrix systems. $p_{w0}^{F,f,m}$ is the initial pressure of the aqueous phase in different media, MPa; $p_{\mathrm{mo}}^{F,f,}$ is the current pressure of the oil phase in different media, MPa; and $p_{mo0}^{F,f,}$ is the initial pressure of the oil phase in different media, MPa.

3.2.3. Initial Conditions and Boundary Conditions

Initial conditions:

$$p_{\mathrm{w},i}(x,y,z,t){|}_{t=0}=p_{\mathrm{wi},i}\hspace{1em}(i=\mathrm{F},\mathrm{f},\mathrm{m})$$

(17)

$$S_{\mathrm{w},i}\left(x,y,z,t\right){|}_{t=0}=S_{\mathrm{wi},i}\hspace{1em}(i=\mathrm{F},\mathrm{f},\mathrm{m})$$

(18)

where p_w,i is the pressure of class i media, MPa; p_wi,i is the initial pressure of class I medium, MPa; S_w,i is the water saturation of class I media, %; S_wi,i is the in. itial water saturation of class I media, %; and F, f, and m are hydraulic fractures, natural fractures, and matrix systems, respectively.

Boundary conditions:

The confining pressure condition is used as the outer boundary of the reservoir:

$${\displaystyle \frac{\partial p}{\partial n}}|\mathsf{\Gamma }=0$$

(19)

During the shut-in period, no production was seen, and the definite displacement was selected as the boundary condition:

$$q_{\mathrm{w}}^{\mathrm{FW}}\left(\begin{array}{c}t\end{array}\right)\hspace{1em}={\displaystyle \frac{{\rho }_{\mathrm{w}}{k}^{\mathrm{t}}{k}_{\mathrm{rw}}h}{{\mu }_{\mathrm{w}}}}(p_{\mathrm{w}}^{\mathrm{F}}-p_{\mathrm{wf}})\hspace{1em}=0$$

(20)

$$0<t⩽t_{\mathrm{shutin}}$$

(21)

During the flowback, the flow pressure at the bottom of the fixed well is selected as the boundary condition:

$$p_{\mathrm{wf}}(t)=p_{\mathrm{wf}}$$

(22)

$$t>t_{\mathrm{shutin}}$$

(23)

where P is the pressure of the overlying rock layer of the matrix, MPa; n is the normal direction of the outer boundary of the reservoir; Γ is the outer boundary of the reservoir; $q_w^{FW}$ is the water phase flow between the wellbore and the hydraulic fracture, m³/s; k^t is the total permeability, mD; h is the thickness of the reservoir, m; p_wf is the bottom flow pressure of the well, MPa; t_shutin is the time of the stuffy well, d; and ρ_w is the density of water, kg/m³.

3.3. Model Solution

By substituting the cross-flow between different media and the governing equations of oil–water two-phase flow within each medium into the material balance equation, the continuity equations for the oil and water phases in hydraulic fractures, secondary fractures, and shale matrix are derived. The nonlinear partial differential equations of the shut-in and flowback mathematical model are discretized using the finite difference method, resulting in a system of nonlinear difference equations. Together with auxiliary equations, boundary conditions, and initial conditions, a semi-implicit approach is applied to linearize the nonlinear coefficient terms, yielding a linear equation system. The Gauss–Seidel iterative method is employed to solve the system. During the computation, when the error falls below a predefined threshold (10⁻⁹), the solution is considered converged, and the process proceeds to the next iteration.

3.4. Numerical Model Establishment and Parameter Input

Current numerical studies on post-fracturing soaking in unconventional reservoirs mainly focus on fracture characterization and productivity prediction [14,17,18]. However, most models skip the injection phase and start simulation at the soaking stage. These approaches are often mechanism-based and do not incorporate realistic geological features, leading to oversimplified results with limited practical use.

To better understand fluid flow during soaking and flowback, this work introduces an integrated workflow that begins with a hydraulic fracturing simulation. Using commercial simulator CMG, the method involves considering the imbibition effect during the injection of fracturing fluid with separate parameters for fractures and matrix; tracking fluid distribution during shut-in and flowback using dynamic geological models; integrating wellhead, stratigraphic, logging, and trajectory data; applying shale content (SH) cutoffs from logs to distinguish sand and shale layers; and generating reservoir properties through sequential Gaussian simulation, tuned with field data from the Erlian Block. This approach enables physics-based simulation of the entire fracturing-to-production process, improving model accuracy and practical relevance for reservoir management. The three-dimensional geological model established is shown in Figure 5. The geological model was imported into CMG software for simulation, and to achieve optimal matching accuracy and computational efficiency, grid sensitivity analysis was conducted, which determined the optimal grid size of 10 m × 10 m × 10 m for individual simulation cells. The numerical model of shut-in well backflow is shown in Figure 6.

Based on the on-site fluid PVT data and on-site construction data, the model is as realistic as possible, and the values of the different parameters are shown in

Table 3. Fluid input parameters for the numerical model of well-stuffed flowback.

Title 1	Parameter	Numeric Intervals
Production grids	Horizontal grid size/m	10
	Longitudinal grid size/m	10
	Vertical grid size/m	10
Fluid PVT	Formation pressure P/MPa	14.6–18.9
	Formation temperature T/℃	70
	Crude oil density ρ/g/cm3	0.8381–0.9263
	Crude oil viscosity μ/mPa·s	4.2–9.46

.

The input parameters of geological engineering are shown in

Table 4. Numerical model geoengineering input parameters.

	Parameter	Numeric Intervals
Geological parameters	Porosity φ/%	7.9–20.8
	Permeability K/mD	0.188–0.499
Engineering parameters	Number of fracturing segments Ns	8–30
	Number of clusters in a single segment Nc	4–6
	Cluster spacing Sc/m	5
Horizontal well parameters	Horizontal well segment length L/m	360
	Horizontal well spacing Sh/m	/
Crack parameters	Half-length of the crack Lf/m	140
	Crack height Hf/m	35
	Fracture conductivity FCD/D·cm	10
	Secondary crack density ρsf/strip/m2	0.46
	Secondary fracture conductivity fCD/D·cm	2

, in which the values of geological parameters are the dynamic inversion results of the flowback of shut-in wells, the engineering parameters and horizontal well parameters are the values of on-site fracturing construction parameters, and the fracture parameters are the inversion results of fracturing construction parameters.

The model was fitted to the production history to correct the phase permeability curve and reservoir parameters to accurately reflect the actual production in the oilfield, as shown in Figure 7, Figure 8 and Figure 9.

4. Numerical Simulation Results of Flowback in Shut-in Wells

4.1. The Law of Imbibition Displacement of Oil Wells in Unconventional Reservoirs

4.1.1. The Law of Substrate Pressurization in Shut-in Wells

During the shut-in period, the leak-off of fracturing fluid into the matrix significantly enhances formation energy, demonstrating remarkable energy-boosting effects [3,19,20]. Analysis of the dynamic pressure distribution following well shut-in indicates that as the shut-in duration increases, the fracturing fluid continuously filtrates into the reservoir matrix under the combined drive of capillary and pressure differential forces, leading to a gradual rise in matrix pore pressure. This process not only alters the fluid distribution in the near-wellbore region but also provides the driving force for crude oil imbibition displacement. After 60 days of shut-in, the reservoir matrix in the Erlian Block was pressurized by an average of 1.3 MPa. The matrix pressure change process of the block reservoir is shown in Figure 10.

4.1.2. The Law of Oil Change by Fracturing Fluid in Shut-in Wells

After the fracturing fluid retained in the post-fracturing well is in contact with the formation, the crude oil in the matrix is replaced by capillary force. After the completion of reservoir fracturing, the fracturing fluid in the hydraulic fracture begins to be filtered into the matrix, and with the increase in the shut-in time, the water saturation in the main fracture decreases, the oil saturation increases, and the water saturation of the matrix system increases. The oil saturation of the fracture after 40 days of well shut-in can reach up to 86%, and the saturation change process of the fracture system is shown in Figure 11.

4.1.3. The Leading Edge of Fracturing Fluid Infiltration in Shut-in Wells

After fracturing, the fracturing fluid will be fractured and replaced, and the fracturing fluid will enter the matrix from the fracture system, which can reflect the strength of the reservoir capillary force and the wettability of the reservoir. The fracturing fluid imbibition front refers to the farthest position of the reservoir caused by the spontaneous imbibition of the fracturing fluid under the action of capillary force, which reflects the fracturing effect of the fracturing fluid. The distribution of water saturation in the reservoirs of the Erlian Block at different boring times is shown in Figure 12 and Figure 13.

4.1.4. The Law of Fracturing Fluid Imbibition Equilibrium in Shut-in Wells

Through the numerical simulation of different shut-in times in the reservoir, the equilibrium time of fracturing fluid imbibition replacement is studied, and the amount of fracturing fluid imbibition and oil exchange is shown in Figure 14.

When the imbibition equilibrium time of the reservoir in the Erlian Block is about 20 days, the cumulative oil intake of the main fracture is about 794.53 m³.

4.1.5. The Law of Matrix Oil Leakage in the Flowback Process

After the fracturing construction is completed, the oil well flows back, and in the process of flowback, with the rapid discharge of fracturing fluid, the osmotic pressure of the formation and fracture increases, and the imbibition displacement of unconventional reservoir oil continues to occur [19]. The oil saturation distribution of unconventional reservoirs under different flowback velocities of the block reservoir is simulated, and the reservoir flowback discharge distance is shown in Figure 15.

When the flowback flow rate is 20 m³/d, the position of the imbibition front extends to the deep part of the formation, with a 30-day imbibition front distance of 2.55 m. In contrast, when the flowback flow rate is 80 m³/d, the 30-day imbibition front reaches 1.35 m, although an imbibition-driven oil change effect still occurs in the matrix. The 20 m³/d flowback system has the largest oil discharge range, and the use of a small flowback liquid is conducive to the oil–water adsorption process and increases the production degree of crude oil in the oil discharge area.

4.2. Optimization of Shut-in Time

Shut-in operations significantly enhance short-term production; however, the imbibition equilibrium time does not necessarily correspond to the optimal productivity period. Imbibition displacement during flowback also affects well performance. Furthermore, optimizing the shut-in duration can mitigate the water blockage induced by hydraulic fracturing [20,21]. According to the on-site demand and actual situation of the oilfield, the same bottomhole flow pressure is set in the flowback stage, and the shut-in time is set to 0 days, 10 days, 30 days, 40 days, 50 days, and 60 days, respectively. The total production is shown in Figure 16.

Shut-in duration significantly enhances short-term production. This is because the imbibition of fracturing fluid significantly increases water saturation in the matrix near fractures, elevating capillary pressure to levels that may even exceed the pressure differential achievable during production, consequently causing water blocking. As production proceeds, reservoir pressure declines and fractures close, necessitating the dynamic adjustment of the optimal production strategy. Therefore, the shut-in time is optimized with the target of maximizing the first-year production of individual wells [22]. The cumulative oil production of the reservoir in the Erlian Block peaked significantly in about 20 to 30 days, and then tended to be stable. Combined with the analysis of the reservoir imbibition displacement law during the well shut-in period and the annual production of a single well with different well shut-in times, it is considered that the optimal well shut-in time in the Erlian Block is about 20–25 days.

4.3. Optimization of the System of Backflow of Shut-in Wells

Numerical simulation software(CMG2021.10) was used to simulate fracture closure, and the flowback velocity of the two blocks was simulated by 10 m³/d, 20 m³/d, 30 m³/d, 40 m³/d, and 50 m³/d to optimize the optimal flowback velocity for the purpose of maintaining the maximum effective retention volume of the artificial fracturing fracture [23,24,25]. The fracture closure dynamics and pressure drop gradient during the flowback process are shown in Figure 17 and Figure 18.

Under the same flowback time, the larger the flowback velocity, the greater the decrease in crack volume and the greater the pressure drop gradient. The crack closes quickly at the beginning of the flowback and gradually slows down later. When the flowback speed is 10–20 m³/d, the crack closure degree is 1.2–1.9%, and the pressure drop rate is 0.098–0.15 MPa/d. At 40–50 m³/d, the crack closure degree is 2.5–2.6%, and the pressure drop rate is 0.26–0.31 MPa/d.

5. Conclusions

(1) The research findings provide insights into addressing the lack of a scientific basis for post-fracturing management practices in the Erlian Basin, such as determining the optimal timing for well opening after operations, selecting appropriate choke sizes for fluid unloading, and setting parameters for switching from natural flow to artificial lift. The impact of management strategies on post-fracturing productivity has also not been systematically understood. This study establishes a tailored methodology for optimizing post-fracturing shut-in and flowback regimes, offering support for resolving practical engineering challenges in the field. For tighter reservoirs, longer shut-in periods may be required; for stress-sensitive formations, consideration should be given to flowing pressure control to prevent fracture damage.

(2) The drainage rate is the key factor affecting the backflow of proppant and the diversion ability of fractures. Optimizing the rate of drainage reduces proppant backflow and maintains a high fracture diversion capacity. The optimized flowback rate of the unconventional reservoir in the Erlian Block is 25–30 m³/d, and the flowback rate can be appropriately adjusted according to the on-site flowback effect, but it is not appropriate to close the artificial fractures due to excessive deviation.

(4) The research methodology (experimental + numerical simulation) serves as a universal framework applicable to other regions. The key lies in obtaining core samples from the target region to conduct similar imbibition experiments for calibrating the capillary pressure model. Parameters such as matrix permeability, fracture geometry, flow capacity, rock wettability, and formation pressure possess “region-specific characteristics” and require recalibration in the model. For the Erlian region, which is dominated by glutenite, matrix permeability is the key parameter controlling imbibition efficiency. In contrast, for shale reservoirs, the organic matter content and clay mineral composition of the rock are likely to be the primary influencing factors.

(5) Implementing an appropriately timed shut-in period can enhance ultimate recovery; however, well closure during the initial phase may impact the economic returns of individual wells. Optimized flowback rates can reduce incidents of proppant flowback and associated treatment costs, thereby generating direct economic benefits. During field implementation, it is necessary to balance theoretically optimal values with practical equipment capabilities (such as choke size control accuracy), wellbore safety, and production scheduling. Overall, reasonable shut-in durations and flowback regimes can effectively improve reservoir drainage and increase overall economic performance.