Interlayer Interference Mechanisms and Key Controlling Factors in Low-Permeability Porous Carbonate Gas Reservoirs

Abstract

To address the pronounced interlayer productivity disparity and uneven reserve utilization during the development of multilayer low-permeability porous carbonate gas reservoirs, the G gas field on the right bank of the Amu Darya River was selected as the study area. Core-parallel physical simulation experiments, orthogonal numerical simulations, and production logging test (PLT) data were integrated to investigate the mechanisms of interlayer interference and its key controlling factors under multilayer commingled production. The results show that interlayer interference is primarily controlled by the permeability contrast and production differential. With increasing permeability contrast, high-permeability layers contribute a larger proportion of total production, whereas the utilization of medium- and low-permeability layers declines, thereby intensifying interlayer interference. Under the same permeability configuration, the interference coefficient increases with increasing production differential. Moreover, compared with the two-layer commingled-production cases, the three-layer system showed a stronger response to pressure-differential variation. When the production differential increased from 1 MPa to 5 MPa, the interference coefficient in the three-layer system increased from 9.84% to 27.83%, indicating more pronounced productivity loss in the medium- and low-permeability layers. Orthogonal numerical simulation indicates that the sensitivity of the main controlling factors follows the order of production differential ≥ permeability ratio > thickness ratio > gas viscosity. PLT data further validate the reliability of the experimental and numerical simulation results. During the development of Well G-22, the XVac layer consistently dominated gas production, whereas the XVm and XVp layers acted as supplementary contributors, indicating a dynamic production pattern in which high-permeability layers are preferentially activated and medium- and low-permeability layers contribute progressively at later stages. These findings demonstrate that permeability heterogeneity is the fundamental cause of interlayer interference, while the production differential serves as an important amplifying factor. This study provides a theoretical basis for zonal production allocation, optimization of the production differential, and stable production management in multilayer low-permeability porous carbonate gas reservoirs.

1. Introduction

Multilayer commingled production is widely used in the development of multilayer gas reservoirs because it can increase single-well productivity, lower development costs, and enhance overall field development efficiency. Nevertheless, when multiple layers are produced under the same bottom-hole flowing pressure, significant differences in original formation pressure, permeability, thickness, and water saturation may lead to pronounced interlayer interference [1,2,3]. This interference often results in an uneven distribution of production among layers and poor utilization of low-permeability reserves. In multilayer porous carbonate gas reservoirs, these issues are particularly prominent because reservoir heterogeneity is usually strong. As a result, development commonly exhibits preferential production from high-permeability layers, delayed contribution from medium- and low-permeability layers, and unbalanced reserve recovery. Such characteristics seriously hinder stable production and the efficient development of gas reservoirs.

Considerable research has been conducted on interlayer interference in multilayer commingled production. Existing studies indicate that differences in formation pressure and permeability are the primary factors controlling the intensity of interlayer interference and the distribution of production capacity among layers [4,5,6,7,8,9,10,11,12]. Physical simulation studies have shown that, during multilayer commingled production, high-pressure and high-permeability layers generally make greater production contributions, and backflow from high-pressure layers to low-pressure layers may occur during the early production stage [13,14,15,16,17,18]. With the deepening of research, the understanding of interlayer interference has gradually expanded from simple multilayer gas production to coupled seepage behavior under multilayer, multiphase, and complex development conditions [18,19,20,21,22,23]. Previous studies have also demonstrated that stronger gas–water interaction, an increasing number of producing layers, and changes in development strategy all intensify interlayer flow competition and affect reserve utilization characteristics [24,25,26]. These studies have provided an important foundation for understanding the mechanisms and governing patterns of interlayer interference under multilayer commingled production conditions.

Existing studies have mainly focused on multilayer sandstone gas reservoirs, whereas research on interlayer interference in multilayer porous carbonate gas reservoirs remains relatively limited. Compared with sandstone reservoirs, carbonate reservoirs differ significantly in pore structure, heterogeneity characteristics, and layer utilization patterns; therefore, conclusions drawn from sandstone reservoirs cannot be directly applied to carbonate systems. In addition, previous studies on this type of gas reservoir have primarily concentrated on reservoir evaluation and production performance analysis, while systematic understanding of the mechanisms of interlayer interference, the main controlling factors, and their relative importance under multilayer commingled production conditions is still lacking. Therefore, it is necessary to integrate experimental simulation, numerical simulation, and field dynamic data to conduct targeted investigations of interlayer interference in this type of gas reservoir [27,28,29,30,31,32,33].

Carbonate gas reservoirs may exhibit markedly different flow behaviors depending on whether the storage and flow spaces are dominated by matrix pores, fractures, or fracture–vug networks. In fractured or fractured–vuggy carbonate reservoirs, gas production is commonly controlled by fracture connectivity, fracture–matrix exchange, and preferential high-conductivity flow paths [34]. In contrast, porous carbonate reservoirs are mainly governed by matrix pore–throat systems, pore-scale heterogeneity, and layer-scale permeability differences. Therefore, interlayer interference in porous carbonate gas reservoirs is more closely related to permeability contrast, effective flow capacity, and production-pressure differential between layers, rather than fracture-controlled channeling alone. The G gas field investigated in this study is characterized by low porosity, low permeability, and relatively continuous porous carbonate intervals, and no field-scale connected fracture–vug system is considered as the dominant flow pathway. Accordingly, this study focuses specifically on interlayer interference in low-permeability porous carbonate reservoirs under multilayer commingled production [35,36,37,38,39].

To address the pronounced interlayer productivity differences and uneven reserve utilization in multilayer porous carbonate gas reservoirs, the G gas field on the right bank of the Amu Darya River was selected as a case study. Core-parallel physical simulation experiments, orthogonal numerical simulations, and production logging test (PLT) data were integrated to systematically investigate the interlayer interference behavior and its main controlling factors under multilayer commingled production conditions. The study focused on evaluating the effects of the permeability contrast, production differential, thickness ratio, and gas viscosity on interlayer interference, and on exploring strategies for production allocation and production differential optimization. The findings provide theoretical and practical support for the efficient development of similar low-permeability carbonate gas reservoirs.

2. Overview of the Research Area

The G gas field is located on the right bank of the Amu Darya River in northeastern Turkmenistan. It is bordered by the Amu Darya River to the southwest and the Turkmenistan–Uzbekistan border to the northeast, with the Falabu gas-bearing structure to the east and the Iligik gas-bearing structure to the west. Tectonically, the field belongs to the Charkuz uplift. Its main gas-bearing intervals occur in the Upper Jurassic carbonate reservoirs of Callovian–Oxfordian age and are subdivided, from bottom to top, into the XVac, XVm, XVp, XVa1, and XVa2 layers. Logging data from seven wells show that the reservoir burial depth ranges from 2147.5 to 2232.5 m, porosity varies from 7.84% to 13.22%, and permeability ranges from 0.14 to 2.34 mD. The reservoir is therefore characterized as a low-porosity, low-permeability carbonate system, although locally developed high-porosity and high-permeability zones are present. Formation pressure is relatively consistent among layers, ranging from 22.10 to 22.89 MPa, with a maximum interlayer difference of 0.79 MPa. The pressure coefficient ranges from 1.04 to 1.08, while the formation temperature varies from 94.60 to 103.63 °C, with an average geothermal gradient of 3.64 °C/100 m.

3. Methodology

The flow behavior of multilayer gas reservoirs is generally regarded as a complex coupled flow problem under strong reservoir heterogeneity. Existing theoretical and numerical studies have demonstrated that interlayer property differences and the production differential are the main factors inducing interlayer interference and affecting reservoir development performance. These factors can lead to uneven production allocation among layers, suppressed production from low-pressure layers, and even backflow, thereby further complicating gas–water interactions. However, most previous studies have been conducted under idealized mathematical or numerical conditions, and their results are highly sensitive to key parameters that are often difficult to constrain accurately in actual reservoirs. As a result, discrepancies may arise between theoretical predictions and field dynamic behavior. Therefore, an integrated approach combining physical simulation experiments and numerical simulation is required to investigate the dynamic evolution of interlayer interference during multilayer gas reservoir development. Physical simulation can compensate for the limitations of purely theoretical and numerical analyses, provide a more direct characterization of interlayer interference behavior, and supply essential data for model calibration and optimization of layer combinations and production strategies. The reliability of the results can be further verified using PLT data.

3.1. Interlayer Interference Physical Simulation Experiment

3.1.1. Experimental Sample

Core samples: The experimental samples were selected on the basis of core data from the G gas field. Most samples were taken from actual reservoir cores of Well G-22. To extend the permeability range represented in the experiments, additional core samples from the adjacent Well Kish-21 were used as supplements.

3.1.2. Experimental Setup and Conditions

To satisfy the requirements of the interlayer interference experiments, a three-core parallel displacement apparatus was established. The experimental system comprised three main parts: a gas supply unit, a dynamic simulation unit, and a data measurement and recording unit. The apparatus mainly consisted of gas sources, displacement pumps, confining-pressure pumps, flow meters, pressure sensors, core holders, six-way valves, and connecting pipelines.

3.1.3. Experimental Plan Design

• Experimental plan;

The core-parallel method was used to investigate interlayer interference under different permeability combinations. The experimental program included a control group and an experimental group. In the control group, two cores with similar permeability were connected in parallel as a reference case, and the results were compared with those of single-core experiments to reduce the influence of heterogeneity. In the experimental group, both two-core and three-core parallel configurations were employed. Three sets of experiments were designed to assess the sensitivity of interlayer interference to the permeability contrast and production differential. Flow data were obtained for different permeability combinations, including high + low, medium + low, and high + medium + low, to evaluate the effect of permeability contrast on interlayer interference. At each production differential, the system was allowed to reach a stable flow state before data acquisition. Flow was considered stable when the outlet flow rate and pressure showed no obvious fluctuation during consecutive measurements, after which the instantaneous gas production and cumulative gas production were recorded.

2.

Experimental parameters;

Prior to the experiments, the core samples were pretreated through cleaning and drilling for sampling. Subsequently, porosity and permeability were measured, and the basic sample properties are summarized in

Table 1. Controlled experimental group rock sample basic porosity permeability table.

Serial Number	Length (cm)	Diameter (cm)	Gas Measurement Permeability (mD)	Porosity (%)	Note
8	3.285	2.489	4.1686	17.62	Control group
18	4.509	2.472	0.0122	1.44
38	2.996	2.488	5.9895	18.57	(Control group)Khigh
50	4.209	2.465	0.0135	2.57
53	4.245	2.496	0.0153	4.31	Klow
151	3.741	2.496	0.0898	8.03	Kmiddle

.

Fluids: A CaCl₂ aqueous solution with a salinity of 97.32 × 10³ mg/L was used as the liquid phase. The solution was prepared as a standard brine according to the compositional analysis of formation water from the G gas field. High-purity nitrogen (99.99%) was used as the gas phase to simulate reservoir natural gas.

Temperature and pressure conditions: Based on the reservoir conditions of the study area, the initial pressure, temperature, and confining pressure were set to 22 MPa, 98 °C, and 26 MPa, respectively.

Production differential design: Different inlet pressures were imposed while the outlet pressure was kept constant, and the displacement pressure differential (ΔP) was increased stepwise to 1, 2, and 5 MPa to evaluate the flow behavior under different production differentials.

3.

Experimental procedures.

Single-core baseline measurements: The permeability of each sample was first determined, and representative high- and low-permeability cores were selected for the experiments. Separate displacement tests were performed on homogeneous cores with different permeability levels to obtain their intrinsic flow-rate curves (Figure 1).

The standard core samples were then assembled in high-sealing-performance core holders to form long-core models representing high-, medium-, and low-permeability layers. Multiple holders were connected in parallel to establish the experimental system. Confining pressure was subsequently applied using a confining-pressure pump, and nitrogen saturation was performed under different pressure conditions to construct the physical model.

According to the experimental scheme, long-core models with different permeability combinations were connected in parallel for the interlayer interference experiments. During the tests, the outlet was opened, and the outlet flow rate was controlled using a back-pressure valve and a flow-control valve to simulate depletion production. Pressure in each layer, production time, instantaneous gas production, and cumulative gas production were recorded at regular intervals throughout the experiment.

3.2. Numerical Simulation and Orthogonal Scheme Design

Physical simulation experiments are time-consuming, costly, and limited in scale, which makes systematic investigation difficult. Therefore, numerical simulation was incorporated for further analysis. The orthogonal design method is widely used in scientific research because it is suitable for multifactor, multilevel studies and can efficiently reveal the effects of different factors on simulation results. Based on the parameter analysis of the physical simulation experiments and the geological characteristics of the G gas field, a reservoir numerical model was established. An orthogonal design scheme was then employed to conduct numerical simulations of multilayer commingled production, and sensitivity analysis was performed to identify the main factors controlling interlayer interference under different reservoir conditions.

3.2.1. Model Establishment and Historical Fitting

To evaluate the reliability of the numerical model, a grid-sensitivity analysis was performed before conducting the orthogonal simulations. Three grid schemes were compared, including a coarse grid, the base grid, and a refined grid. The simulated cumulative gas production, layer contribution ratio, and interference coefficient showed only minor differences between the base and refined grids, whereas the computational cost increased significantly in the refined model. Therefore, the base grid with a planar size of 100 m × 100 m and an average vertical thickness of approximately 0.8 m was selected as a balance between numerical accuracy and computational efficiency. A single-porosity gas–water two-phase model was established for the numerical simulation. The planar grid size was set to 100 m × 100 m, and the vertical grid thickness was approximately 0.8 m. The total number of grid cells was 3,803,184, as shown in Figure 2 and

Table 2. Basic parameters of the gas reservoir numerical simulation model.

Parameters		Value
Model		Plane: Corner point grid; Vertical: 29 layers
Physical properties	Reservoir thickness(m)	115
	porosity(%)	10.43
	Permeability(mD)	3.94
Water body multiplicity		2.0

. The single-porosity formulation was adopted because the target reservoir is interpreted as a matrix-dominated porous carbonate gas reservoir rather than a fracture- or vug-dominated carbonate system. Although dual-porosity or dual-permeability models are commonly required for fractured carbonate reservoirs, their use requires reliable constraints on fracture porosity, fracture permeability, fracture spacing, shape factor, and matrix–fracture transfer parameters. Such parameters are not sufficiently supported by the available core, logging, production, and PLT data for the G gas field. In addition, the production behavior considered in this study is mainly controlled by interlayer permeability contrast and pressure-differential-driven flow competition among the XVac, XVp, and XVm layers. Therefore, a single-porosity gas–water two-phase model is appropriate for describing the layer-scale interference behavior of this low-permeability porous carbonate reservoir. This modeling assumption limits the applicability of the present model to matrix-dominated porous carbonate systems and does not aim to represent strongly fractured or fractured–vuggy reservoirs.

In the G gas field, based on the geological and production characteristics of the study area, the activities of marginal water and bottom water are currently weak and have not had a significant impact on the current development performance of the gas field. Therefore, in this study, the water body diversity value of 2.0 is only introduced as a preliminary boundary condition for model construction and possible future development considerations. The water body quantity is set at 2.0, which indicates that it has a moderate water-bearing layer supporting effect. This parameter reflects the relative strength of the connected water bodies and is calibrated through pressure and production history matching. A smaller value would underestimate the pressure support effect, while a larger value would overestimate the water-bearing layer energy and weaken the simulated pressure consumption response. The value of 2.0 is quite consistent with the observed performance of the gas field and is therefore used in the basic numerical model.

Based on the production data from the G gas field, history matching was conducted, as shown in Figure 3. The geological model was thereby further refined, and the key petrophysical parameters of the numerical model were systematically calibrated.

3.2.2. Orthogonal Experimental Design Scheme

Based on the established model, the experimental understanding of multilayer gas reservoir production, and the production characteristics of the G gas field, numerical simulations were conducted to investigate production behavior under varying permeability, thickness, production differential, and gas viscosity conditions. Three levels were assigned to each of these four factors. An L9(3⁴) orthogonal design was adopted to cover the reasonable variation range of each parameter, resulting in nine simulation schemes. The production contribution rate and interference coefficient of each layer were then statistically analyzed for all schemes, providing the basis for evaluating interlayer interference patterns, as shown in

Table 3. Interlayer interference L9 orthogonal design table.

Factors	Level 1	Level 2	Level 3
Permeability ratio K(ac:p:m)	2.20:0.63:1.57	2.20:1.26:1.57	2.20:0.31:0.79
Thickness ratio h(ac:p:m)	18.4:15.7:17.7	27.6:15.7:17.7	9.2:15.7:17.7
Production differential (MPa)	2	5	12
Gas viscosity (mPa·s)	0.015	0.0203	0.025

.

4. Results and Discussion

4.1. Analysis of Factors Affecting Interlayer Interference

The interference coefficient is used to quantify the effect of interlayer interference on gas productivity during dual-layer commingled production. The interlayer interference coefficient, η, was calculated by comparing the measured total flow rate under parallel production with the sum of the single-layer flow rates. The interlayer interference coefficient η is dimensionless. Permeability, production differential, and the number of commingled layers were selected as the key variables for analyzing their relationship with the interlayer interference coefficient.

$$Q_{\mathrm{single}}={\displaystyle \frac{k\cdot A\cdot \Delta P}{\mu \cdot L}}$$

(1)

$$Q_{parallel}={\displaystyle \frac{\left(k_1+k_2\right)\cdot A\cdot \Delta P}{\mu \cdot L+\left(k_1+k_2\right)\cdot A\cdot R}}$$

(2)

$$\eta =1-{\displaystyle \frac{Q_{parallel}}{Q_{sum}}}$$

(3)

In the formula: $Q_{\mathrm{single}}$—Single-layer flow, m³/s;

k—Permeability of the sub-layer, m²;

A—The extent of the small-scale wave influence, m²;

$\Delta P$—Interlayer production differential, Pa;

$\mu$—Non-Darcy number, Pa·s;

$L$—Interlayer vertical distance, m;

$Q_{parallel}$—Total flow rate, m³/s;

$R$—The additional resistance term in the system;

R in the interference-coefficient formulation reflects the extra pressure loss caused by interlayer flow competition and nonlinear gas flow effects. When the production differential increases, gas flow in the dominant high-permeability pathway becomes more intense, and the pressure loss is no longer strictly proportional to the Darcy-flow term. This non-Darcy contribution strengthens the preferential flow through high-permeability layers and further suppresses the effective contribution of low-permeability layers. Therefore, increasing the production differential not only raises the total driving force but also amplifies interlayer productivity imbalance, which explains why the interference coefficient increases with the production differential under the same permeability configuration.

Samples 38, 151, and 53 were selected as representative high-, medium-, and low-permeability cores, respectively. Parallel experiments were carried out according to the experimental scheme. The permeability ratio between cores 53 and 151 was 5.87. As shown in Figure 4 and Figure 5, the interference coefficient increased progressively with increasing production differential, reaching 12.44% at 1 MPa and 22.52% at 5 MPa. The ratio of interlayer production contribution under different production differentials ranged from approximately 8.4 to 9.8, which was markedly higher than the permeability ratio. With increasing production differential, this contribution ratio decreased slightly, suggesting progressive activation of the low-permeability layer.

Core 53 and core 38 had permeabilities of 0.0153 and 5.9895 mD, respectively, giving a permeability ratio of 391.47. As shown in Figure 6 and Figure 7, the interference coefficient increased with the production differential, rising from 25.71% at 1 MPa to 34.99% at 5 MPa. Under all tested production differentials, the high-permeability layer contributed more than 99.999% of total production, indicating that an excessively large permeability contrast can cause severe interlayer interference and make the low-permeability layer almost nonproductive. Moreover, the production contribution of the high-permeability layer increased slightly with increasing production differential.

Three-core parallel displacement experiments were performed using cores with permeabilities of 0.0153, 0.0898, and 5.9895 mD, respectively, with a maximum permeability ratio of 391.47. As shown in Figure 8 and Figure 9, the interference coefficient increased with the production differential, from 9.84% at 1 MPa to 27.83% at 5 MPa. Compared with the two-core parallel case, the interference coefficient in the three-core system exhibited significantly greater sensitivity to the production differential. The high-permeability layer contributed approximately 98.92–99.21% of total production under all tested production differentials, demonstrating its absolute dominance. This indicates that interlayer interference in three-layer commingled production is substantially more intense and complex than that in two-layer commingled production, resulting in a rapid increase in productivity loss from the medium- and low-permeability layers.

Control group: For core samples with similar physical properties (permeabilities of 5.9895 and 4.1686 mD), and under a displacement pressure differential of 1–5 MPa, the interference coefficient was limited to 2.44–3.67% (

Table 4. Parameters of interference factors for each sample under different physical properties and production differentials.

Production Differential (MPa)	Core Number	Length (cm)	Diameter (cm)	Volume (mL)	Time (s)	Interference Factor (%)
1	8	2.996	2.488	2524.55	60.00	2.44
	38	3.285	2.489	3486.77	60.00
2	8	2.996	2.488	6851.17	60.00	3.67
	38	3.285	2.489	8842.54	60.00
5	8	2.996	2.488	35,780.04	60.00	3.51
	38	3.285	2.489	45,215.21	60.00

), indicating that permeability contrast is the dominant factor governing interlayer interference. As shown in Figure 10, interlayer interference loss increased with increasing permeability contrast. For a given permeability combination, the interference coefficient also increased with the production differential. These results suggest that multilayer commingled production is highly sensitive to the production differential, and therefore, dynamic optimization of the production differential is required during field development.

Mechanistically, the observed interlayer interference can be attributed to the competition between flow-capacity contrast and pressure-differential-driven production allocation. Under the same bottom-hole flowing pressure, the high-permeability layer has a much lower flow resistance and therefore responds earlier to pressure drawdown. As a result, a large proportion of the total production is rapidly supplied by the high-permeability layer, while the medium- and low-permeability layers remain partially suppressed because of their higher viscous resistance and weaker pressure-transmission capacity. With continued production, the pressure energy of the high-permeability layer is preferentially depleted, and its relative productivity advantage gradually weakens. This process allows the medium- and low-permeability layers to become progressively activated, which explains the delayed but increasing contribution of these layers observed in both the physical experiments and the PLT data.

Overall, interlayer interference differs markedly among different permeability combinations and becomes more severe as the permeability contrast increases. Under the same permeability configuration, the interference coefficient increases with the production differential. In three-layer commingled production, interlayer interference is more sensitive to the production differential, leading to greater productivity loss in the medium- and low-permeability layers. These findings demonstrate that permeability heterogeneity is the fundamental cause of interlayer interference, while the production differential serves as an important amplifying factor.

4.2. Sensitivity Analysis of the Main Control Factors

As shown in

Table 5. Simulation Results of Interlayer Interference for Different Schemes.

Scheme Number	Parameter Level				Output Contribution Rate (%)			Interference Coefficient
	K (ac:p:m)	h (ac:p:m)	Production Differential (MPa)	Gas Viscosity (mPa·s)	XVac	XVp	XVm
1	2.20:0.63:1.57	18.4:15.7:17.7	2	0.015	55.2	18.5	26.3	0.08
2	2.20:0.63:1.57	18.4:22.1:26.6	5	0.0203	72.3	10.5	17.2	0.18
3	2.20:0.63:1.57	18.4:11.0:14.2	12	0.025	68.9	9.8	21.3	0.33
4	2.20:1.10:2.20	18.4:15.7:17.7	5	0.025	55.7	19.7	24.6	0.12
5	2.20:1.10:2.20	18.4:22.1:26.6	12	0.015	61.2	20.6	18.2	0.15
6	2.20:1.10:2.20	18.4:11.0:14.2	2	0.0203	56.1	20.1	23.8	0.06
7	2.20:0.33:0.66	18.4:15.7:17.7	12	0.0203	79.4	8.7	11.9	0.37
8	2.20:0.33:0.66	18.4:22.1:26.6	2	0.025	76.3	11.4	12.3	0.21
9	2.20:0.33:0.66	18.4:11.0:14.2	5	0.015	69.4	13.3	17.3	0.24

, the orthogonal numerical simulation results indicate that interlayer interference varies significantly under different parameter combinations, with the interference coefficient ranging from 0.06 to 0.37. In all cases, the XVac layer remains the dominant contributor to total production, demonstrating the strong control of the high-permeability layer on overall productivity. With increasing production differential and permeability contrast, interlayer interference becomes more severe, and the production contributions of the medium- and low-permeability layers decrease markedly. In contrast, the thickness ratio has a moderate effect, whereas gas viscosity shows a relatively minor influence. These results confirm that the production differential and permeability ratio are the primary factors controlling the intensity of interlayer interference.

As shown in Figure 11 and

Table 6. Sensitivity Analysis of Interference Coefficient by Different Factors.

Factors	Level	Average Interference Coefficient	Range	Order of Sensitivity
Production differential (MPa)	2	0.11	0.23	1
	5	0.18
	12	0.34
Permeability ratio	1:0.15:0.30	0.28	0.22	2
	1:0.29:0.71	0.20
	1:0.50:1.00	0.06
Thickness ratio	1:0.60:0.80	0.16	0.12	3
	1:0.85:0.96	0.23
	1:1.20:1.50	0.28
Gas viscosity (mPa·s)	0.015	0.21	0.07	4
	0.0203	0.20
	0.025	0.19

, sensitivity analysis of the numerical simulation results indicates that, under three-layer commingled production conditions, the factors affecting interlayer interference are ranked as follows: production differential ≥ permeability ratio > thickness ratio > viscosity. The effects of the production differential and permeability ratio are nearly equivalent. The interaction diagram further shows that these two factors exhibit a clear coupled effect, especially under high-production-differential conditions. Accordingly, the production differential should be dynamically optimized at different development stages to mitigate interlayer interference.

Figure 12 shows that permeability heterogeneity has a decisive effect on interlayer interference. The interference coefficient decreases significantly with a decreasing permeability ratio, whereas the median interference coefficient increases from 13.2% to 24.7% as the permeability contrast increases from 2.0 to 6.7. The production differential is also positively correlated with the interference coefficient, and its variation trend is consistent with that observed in the physical simulation experiments.

4.3. Field Validation with PLT Data

PLT data from Well G-22 in the G gas field (Figure 13 and

Table 7. Statistical Table of Geological Production Contribution of Well G-22.

Number	Formation Name	Stratigraphic Section (m)	1 September 2022		23 July 2024
			Output (103 m3/d)	Contribution Ratio (%)	Output (103 m3/d)	Contribution Ratio (%)
1	XVac	2110.7–2161.5	160.69	93.92	129.75	83.07
2	XVp	2161.5–2185.4	1.16	0.68	1.55	0.99
3	XVm	2185.4–2239.5	9.24	5.40	24.90	15.94

) show a pronounced difference in production contribution between the XVac and XVm layers. Although these two layers have similar reserve volumes, single-well PLT results indicate markedly different production performances. In 2022, the XVac and XVm layers contributed 93.92% and 5.40% of total production, respectively; by 2024, their contributions had changed to 83.07% and 15.94%, respectively. This trend suggests that the production dominance of the high-permeability layer gradually weakens over time, whereas the medium- and low-permeability layers progressively supplement production during the later development stage. Overall, however, the reservoir still exhibits a production pattern dominated by the high-permeability layer, with insufficient utilization of the medium- and low-permeability layers.

The PLT-derived change in production contribution also supports the progressive-activation mechanism. In the early stage, the XVac layer dominates the gas supply because of its higher effective flow capacity. As the XVac layer undergoes stronger pressure depletion, the pressure contrast between layers is gradually adjusted, and the XVm and XVp layers begin to provide supplementary production. However, because the intrinsic permeability contrast remains large, the late-stage activation of the medium- and low-permeability layers cannot fully offset the long-term dominance of the high-permeability layer. This explains why the XVac contribution decreases over time but still remains the dominant producing interval.

The PLT results further validate the physical and numerical simulation results from the perspective of field dynamic behavior, and the consistent overall trends demonstrate the reliability of the corresponding interpretations. In Well G-22, the XVac layer remained the dominant producing interval at different test times, whereas the XVm and XVp layers acted as supplementary contributors. This indicates that the high-permeability layer is preferentially activated and dominates production for a long period, while the medium- and low-permeability layers are activated later and progressively contribute during subsequent development. Although the XVac and XVm layers have similar reserve volumes, their production contributions differ significantly, suggesting that production performance is governed not only by reserves, but also by permeability, pore structure, interlayer pressure communication, and production strategy. Overall, interlayer interference persists throughout the development process, causing rapid production differentiation in the early stage and influencing the succession capacity of different layers in the middle and late stages. Therefore, field development should strengthen zonal production allocation, optimize the production differential, and implement dynamic regulation to limit excessive depletion of the high-permeability layer while promoting the progressive contribution of medium- and low-permeability layers, thereby improving reserve utilization and maintaining stable production. Although the XVm and XVp layers still retain development potential, the dominance of the high-permeability layer remains difficult to change when the permeability contrast is excessively large. Overall, the PLT results are consistent with both the physical experiment and numerical simulation results, confirming that the high-permeability layer remains the dominant contributor over a long production period, whereas the medium- and low-permeability layers gradually supplement production during the later development stage.

5. Suggestions for Multi-Layer Commingling Production in Porous Carbonate Rock Gas Reservoirs

Porous carbonate reservoirs are typically highly heterogeneous, and their production behavior under multilayer commingled production shows distinct characteristics. Based on the integrated results of physical and numerical simulations, four key aspects should be considered for the effective implementation of multilayer commingled production in such reservoirs.

• Optimize well placement. Owing to interlayer heterogeneity, wells should be deployed in zones where the reservoir is well developed, and multilayer commingled production should be implemented in appropriately selected intervals. This approach can improve reserve control and help maintain higher gas production during the initial development stage. In addition, a combination of reasonable layer division and progressive commingled production is essential for improving gas recovery.
• Optimize the combination of commingled layers. For wells under multilayer commingled production, greater permeability contrast between layers leads to stronger interlayer interference and poorer development performance of low-permeability layers. Therefore, an excessive interlayer permeability contrast should be avoided as much as possible. Moreover, as the number of commingled layers increases, interlayer flow competition becomes more complex, the system becomes more sensitive to the production differential, and productivity loss in medium- and low-permeability layers becomes more pronounced. Accordingly, the combination of producing layers should be carefully optimized.
• Control production differential. Based on the results of physical simulation experiments and reservoir numerical simulation, a reasonable production differential should be determined to avoid the amplification of interlayer interference caused by excessive pressure drawdown and to limit the excessive dominance of high-permeability layers in gas supply.
• Select the optimal timing for commingled production. The initial production stage is the most favorable period for implementing commingled production. A well-designed production schedule can improve early-stage performance, while supplementary development or infill measures for low-permeability layers can be considered at later stages to support stable production during the middle and late development periods.

6. Conclusions

• This study integrated physical simulation experiments, orthogonal numerical simulation, and PLT field data to investigate the interlayer interference behavior in multilayer low-permeability porous carbonate gas reservoirs. The results demonstrate that interlayer interference is fundamentally controlled by reservoir heterogeneity, while the production differential acts as a key factor that amplifies productivity imbalance during commingled production. A strong permeability contrast causes long-term dominance of high-permeability layers and suppresses the effective utilization of medium- and low-permeability intervals, especially in multilayer commingled systems with increasing production complexity.
• The combined experimental, numerical, and field-production analyses reveal that production allocation among layers evolves dynamically during reservoir depletion. High-permeability layers contribute predominantly during the early production stage, whereas medium- and low-permeability layers gradually participate in gas supply during later development stages. This dynamic succession behavior indicates that balanced reserve utilization cannot be achieved solely through commingled production and requires coordinated production management.
• Sensitivity analysis further indicates that the production differential and permeability ratio are the most critical factors affecting interlayer interference, while the thickness ratio and gas viscosity exert comparatively weaker influences. Therefore, development strategies for multilayer carbonate gas reservoirs should prioritize dynamic optimization of the production differential and rational layer combination design in order to mitigate excessive depletion of dominant layers and improve the contribution of less favorable intervals.
• The study provides important guidance for multilayer carbonate gas reservoir development, particularly in terms of zonal production allocation, commingled-layer combination optimization, production-pressure management, and stable production control. The proposed understanding of interlayer interference mechanisms and production evolution characteristics can support more efficient reserve utilization and long-term stable development of similar low-permeability carbonate gas reservoirs.