Sand Production Mechanisms in Sandstone Gas Reservoirs during Cyclic Injection–Production with Variable Water Saturation Authors: Xiaoling Tang, Baomei Yang, Jinghong Hu, Yuan Zhang, Yidong Cai, Jun Lu Published: 2026 ABSTRACT Formation water is common when using depleted reservoirs to build underground gas storage, which causes serious sand production problems during the periodic injection–production operation of natural gas. Current research focuses on the sand production of the reservoir in the single-phase case. Little attention is paid to the sand production caused by injection and production in water-bearing reservoirs. To fill this gap, this study reveals the sand production mechanism of reservoir rocks under two-phase flow conditions using alternating displacement. The critical sand production pressure difference of the core under different water saturations was first experimentally measured and compared with the results under single-phase flow conditions. Results show that the critical sanding pressure difference of water-bearing cores is generally higher than that of dry cores. Then, the impacts of water saturation on sand production during cyclic injection–production in sandstone gas storage were experimentally evaluated as well. The sand production shows a nonmonotonic trend of increasing first and then decreasing when the water saturation is 0%–85%, with a peak at 30% water saturation. Experiments also show that the change of sand production is controlled by the number of alternating displacement cycles. Within the range of 15–30 alternations, the growth rate of core permeability and sand production decelerates. Beyond 30 alternations, a notable increment is observed with further increases in alternations. This research finding has significant value for the construction and operation of gas storage facilities in water-bearing sandstone reservoirs. Introduction Underground gas storage is an underground and ground integrated system for natural gas injection, storage, and production. It is an indispensable key link in the overall industrial chain of natural gas production, supply, storage, and sales. The depleted oil and gas reservoir gas storage is the main form of gas storage in the world and in China, and the working gas volume accounts for more than 80%. − In the actual reservoir gas injection development process, it is often accompanied by a gas and water two-phase seepage phenomenon. In depleted underground gas storage, the widespread existence of formation water will aggravate the sand production during the process of cyclic injection and production. The emergence of this situation will significantly shorten the service life of gas storage. − In the process of alternating injection and production of water-bearing underground gas storage, sand production generally occurs in the reservoir. This reflects that the rock stratum has been damaged or weakened to a certain extent under the action of periodic injection and mining. − A review of domestic and international literature reveals that numerous studies focus on experimental research concerning factors such as reservoir physical properties, gas injection rate, cycle times, and the burial depth of gas storage facilities. − The main research methods are physical simulation experiments, such as visualization experiments and core displacement experiments. Tang et al. demonstrated that gas-phase permeability in the gas-drive-water pure zone is progressively enhanced over successive injection–production cycles, stabilizing after approximately four to five cycles. Lu et al. identified a key mechanism whereby the gas injection–production rate governs the evaporation rate of gas into the water phase. This mechanism ultimately controls the evolution of matrix water saturation and gas-phase permeability throughout the cycling process. Li et al. − conducted long-core gas–water flooding experiments and revealed a direct impact of repeated injection cycles. They observed that increasing cycles led to more complex gas–water distributions, higher residual gas saturation, a narrower two-phase flow range, and an overall adverse impact on the displacement efficiency. Using visual microfluidic experiments, Li et al. , revealed that the gas–water flooding process is subject to phenomena such as two-phase entrapment, interlocking, and bypass flow. Once trapped, these phases are difficult to remobilize in subsequent injection–production cycles. Jiang and Qi constructed a microscopic visualization model to simulate the gas–water migration law of water-invaded gas storage in the initial stage of gas injection and the cyclic injection–production process. After multiple rounds of gas–water displacement, the effective gas–water seepage interval was constricted, consequently reducing the movable pore space to a range of 42% to 54%. Although considerable research exists on the mechanisms of gas storage construction, studies specifically focused on the retrofitting of high-water-cut reservoirs for this purpose are relatively scarce. Constrained by existing experimental methods for determining gas–water relative permeability, current physical models are unable to evaluate the effect of gas injection rate on relative permeability under unsteady displacement conditions, nor can they assess the impact of variable injection–production rates on relative permeability and residual water saturation across different cycle periods. − Currently, how this retention subsequently affects injection–production cycles and triggers sand production has yet to be fully understood. Therefore, mitigating sand production caused by alternating displacement under complex gas–water flow conditions has become an important objective for optimizing the operation of aquifer gas storage reservoirs. Physical simulation experiments offer an effective approach to address these questions. − Since the overall peak-shaving capacity of existing gas storage in China is limited and the cost of new storage remains high, clarifying sand production during cyclic injection–production in two-phase reservoirs is urgent. Based on the geological parameters of sandstone reservoirs, this study systematically evaluates sand production behavior under high water saturation. Crucially, we recognize that during actual operation, the saturation of water will change due to the invasion of formation water. Therefore, we abandoned the traditional dry/wet core comparison test method and adopted the “variable water saturation” experimental design instead. This method can more accurately simulate the actual operation of underground natural gas storage. At the same time, the periodic stress change of the underground gas storage makes the reservoir rock to bear the cyclic stress, which leads to the accumulation of fatigue damage. Therefore, it is necessary to study the sand production mechanism under cyclic injection–production conditions. In this study, the effects of the injection–production cycle, alternating displacement, and core water saturation on the sand production mechanism were systematically analyzed. This study provides an important theoretical basis for optimizing the injection–production scheme and designing natural gas storage facilities. It is beneficial for reducing damage during the cyclic injection–production process and is of great significance in guiding the design and construction of facilities. Materials and Methods Materials The experimental core was obtained from a sandstone gas storage facility in Northern China. The target layer is a sandstone reservoir with medium porosity and permeability, developed in braided river delta front deposition. The buried depth of the reservoir is −2310 to −2620 m. The ranges of core permeability and porosity are shown in Table . 1Core Basic DataCore numberCore depth (m)Number of coresCore porosityPermeability range before experiment (10–3 μm2)1–112824.87519.84%–23.88%102.54–197.79 Core Basic Data Core analysis indicates that the experimental rock cores exhibit an average quartz content of 43.93%, a feldspar content of 11.71%, a lithic fragment content of 25.85%, and an interstitial material content of 7.47% (the interstitial material content is illustrated in Figure ). The detrital grains are loosely cemented, with clay minerals being the main cementing material, averaging 8.08% in content. The cementing type is mostly pore-filling cementation. Relative content in interstitial material. Note: These values represent the relative mass fractions of key interstitial components, and thus the total does not sum to 100% of the whole-rock mineralogy. The wettability of a reservoir refers to the tendency of a fluid to spontaneously replace another fluid on the rock surface due to intermolecular interactions in an oil–water–rock system. As a key reservoir rock property, it governs the distribution of oil, gas, and water in pores, directly impacting displacement efficiency and ultimate hydrocarbon recovery. The wettability test of the experimental core was carried out by water samples. The measurement results are shown in Figure . It was found that the contact angle of the core is 45.69°on both sides, which is less than 90°. Experimental results indicate that the core sample is strongly hydrophilic, which means that water within the pore space is largely retained and resists displacement during production. Contact angle of experimental core samples. The water samples used in the water injection experiment were taken from the reservoir formation water of the blocks. The water characteristics are listed in Table . 2Summary Table of Block Formation Water PropertiesContent of formation water ion (mg/L) CationAnion Na+ + K+ Ca2+ Mg2+ Cl– HCO3– CO3 2– Total mineralizationWater type1379–22400–500–361176–24341252–26690–6154252–6692NaHCO3 1735241215461822835300 Summary Table of Block Formation Water Properties Experimental Apparatus Figure shows a schematic diagram of the experimental setup used to investigate rock damage in gas storage. This setup can simulate the cyclic injection and production processes under multiphase conditions. The experimental setup is capable of simulating single-phase, two-phase, and three-phase flow conditions. It allows for alternating gas displacement to be performed directly on cores containing an oil–water mixture without the need to replace any displacement equipment. The experimental device is an integrated component that integrates functional modules such as gas–liquid switching, pressurization, formation simulation, sensor data acquisition, and sampling components. The device is connected to the computer system through the sensor, which can record the key parameters such as gas flow rate, pressure difference in the core area, and fluid flow rate in real time. The real-time monitoring of reservoir permeability changes and sand production during the cyclic injection–production process of gas storage is realized. Sketch of the apparatus used in the displacement experiment. In this experiment, formation water samples and nitrogen gas were used as experimental materials. Nitrogen was employed to simulate actual field gas injection conditions. This selection was made primarily due to nitrogen’s high chemical stability and safety. Employing nitrogen helps to eliminate interference from confounding physical and chemical factors, thereby mitigating risks associated with reservoir damage or pore blockage. This experiment was designed based on the principle of univariate control, aiming to systematically study the influence of a single variable on the sand production rate. Specifically, the influence of two variables, core water saturation and injection–production cycle number, on core sand production was investigated by controlling them separately. This method simulates the cyclic stress experienced by reservoir rock during water injection–production without the need for complex flow direction reversal mechanisms. It is important to note that this alternating pressure cycle is achieved by adjusting the inlet pressure, and the fluid flow direction remains constant throughout the process. Experimental Parameters To characterize the compressive effect of confining pressure on core permeability, we initially established the confining pressure–permeability relationship through an experiment. Specifically, we tested the core permeability under different confining pressures with a fixed inlet pressure of 0.2 MPa. The tests aimed to clarify the impact of confining pressure variations on permeability within a normal range and to determine whether excessively high confining pressures could lead to core fracture. The results of this experiment were directly used as the basis for setting the confining pressure in subsequent experiments, thus ensuring that the mechanical environment of the core more accurately simulates the actual formation conditions. The permeability of the same core was measured three times under 1 MPa confining pressure, and the results were 145.99, 150.42, and 148.55 mD, respectively. The error in the experimental system was 3.03%. Under the condition of maintaining a constant inlet pressure of 0.2 MPa, experiments were conducted by gradually increasing the confining pressure from 1 to 8 MPa. The specific data are given in Table . It can be seen that when the confining pressure increases from 1 to 8 MPa, the change rate of core permeability is only 2.91%, which is lower than the measurement error of the experimental system (3.03%). Therefore, in the range of confining pressure studied in the experiment, the influence of confining pressure loaded on the core on permeability can be neglected. 3Variation of Core Permeability with Confining PressureConfining pressure (MPa)Duration (min)Core permeability (mD)Confining pressure (MPa)Duration (min)Core permeability (mD)15148.3255151.4725145.8665157.3735149.8875150.4145147.8585152.63 Variation of Core Permeability with Confining Pressure Experimental Procedure To investigate the critical pressure difference across the core under both single-phase and two-phase (with water content) conditions, as well as structural changes and sand production during alternating displacement, the following experiments were conducted. Cores with Different Water Saturations To investigate sand production under varying water saturations, we first saturated the core samples with formation water and then displaced them to achieve different initial water contents, as shown in Figure . First, the core was saturated to 100% water content, and the porosity of the core was calculated. The saturated core was displaced by applying a range of pressure differentials, and the resulting changes in water saturation were monitored. Upon reaching an irreducible water saturation, the core was extracted and weighed to determine the water saturation corresponding to the applied pressure differential. The core schematic diagram of different water saturations. Water saturation is the ratio of the volume of water in the pore to the pore volume of the rock. After the water sample is saturated with the core, the pore volume of the core and the water volume of the core are calculated first. By measuring the pore volume of the core and the mass change before and after water flooding, the water saturation of the core is calculated by eq . The formula for calculating the water volume of the core is as follows. 1 Vs=ms−m0ρw 2 V1=m1−m0ρw 3 Water saturation=V1Vp×100% where V s is the volume of the core pore, V 1 is the water volume in the pore at a specific water saturation, ms is the mass in the 100% saturated water state, m 1 is the mass of the core at a specific water saturation, m 0 is the dry core mass, ρw is the density of water. The water content within the core is regulated by an applied pressure difference. The actual water saturation is then calculated from the measured porosity and the volume of water present. This result serves as the basis for determining the critical pressure difference and conducting the subsequent alternating displacement experiment. Critical Pressure Differential Critical Pressure Difference in the Single-Phase Flow The experiment takes the pressure difference acting on the core as a variable and records the permeability and quality changes of the core under different pressure differences. According to the change of permeability and quality, the critical pressure difference leading to core sand production was analyzed. The specific experimental contents are as follows: (1) the experimental assembly is checked for leaks, the core is loaded into the holder, and the confining and displacement pressures are applied. The core is then flooded with gas heated to the formation temperature of 80 °C to commence the displacement experiment. (2) Under stable confining pressure, the pressure difference at both ends of the core is gradually increased until sand is produced in the core, and the critical pressure difference of the core is obtained. (3) In the actual experiment, it is difficult to observe the sand production. Therefore, after each change of pressure difference, the system is maintained for 20 min. After the experiment, the permeability and quality of the core are measured, and the change of the permeability and quality of the core is observed to evaluate sand production. Figure shows the change of displacement pressure with displacement time during the process of measuring the critical production pressure difference. Pressure variation with time during the critical pressure differential experiment. Critical Pressure Difference in the Two-Phase Flow The purpose of this experiment was to study the effects of different water saturation levels on the fluid distribution and sand production in cores. Different saturations were established by varying the water content, allowing for the measurement of the corresponding critical sand-production pressure difference. For each core, the liquid flow rate and sand production were recorded across various saturation conditions. The specific experimental procedure is as follows: (a) first, the core is saturated to 100% water content, and the porosity of the core is calculated. (b) Different differential pressures are used to displace the saturated core. After each differential pressure is applied to the core until no more water is produced, the core is taken out and weighed, and the water saturation under the corresponding differential pressure is calculated. (c) According to the needs of the experiment, different pressure differences are used to displace the saturated water core. After the displacement, the core is taken out, and the water saturation of the core is calculated according to eq . (d) Displacement experiments are conducted on cores with different water saturations, continuously increasing the displacement pressure difference, observing the sand production of the core, and determining the critical pressure difference of the core. Alternate Displacement Experiments The alternating experiment was carried out under the conditions of the critical pressure difference measured in Critical Pressure Differential. On the basis of the critical pressure difference, the alternating interval for the core was determined, and each set of experiments was repeated for 50 cycles. The pressure variation during the alternating process is shown in Figure . Pressure variation with time under cyclic pressure differential conditions. Measurements of core mass and permeability were taken at 5-cycle intervals during the alternating displacement process for the purpose of documenting their variation. The effects of the number of alternation cycles, displacement cycle parameters, and water saturation on sand production in the core samples were investigated. Specific experimental details are shown in Table . 4Specific Experimental ContentExperiment typeVariableExperimentationSingle-phaseNumber of times of displacement0–50 stepsUnder a constant confining pressure of 5 MPa, the inlet pressure was cyclically varied. Following every 5 cycles, the core’s integrity was examined, and its mass change was recorded. The gas flow rate through the core was measured to calculate its permeability.Replacement cycle30 s1 min2 min3 min5 minTwo-phaseWater saturation20%Under a constant confining pressure of 5 MPa, cyclic variations in the inlet pressure were applied to the core. After every 5 pressure cycles, the integrity of the core was examined, and any change in its mass was recorded. The gas flow through the core was continuously measured during the experiment. Upon completion, the core was dried to determine its postexperiment permeability.30%40%50%60%100% Specific Experimental Content Because of the difficulty in directly observing sand production during the experiment, it was indirectly determined by measuring the change in core mass before and after the displacement process. The sand amount was indirectly calculated based on the accurate difference in core dry mass before and after the experiment. The advantage is that it can completely capture all solid products, including fine particles that are easily missed by traditional collection methods. In order to ensure the reliability of the data, all experiments strictly followed reproducible standard operating procedures. In summary, this study employed standardized sample pretreatment, repeated measurements using high-precision instruments, and a systematic error analysis. These procedures ensured the validity and reliability of the data acquired by the quality difference method, laying a solid foundation for the subsequent quantitative analysis of the relationship between pressure differences and sand production variation. Experimental Results and Discussion Determination of Water Saturation The gas-displacement experiment was conducted on the water-saturated core. The displacement pressure was gradually increased to 1 MPa. Water saturation was monitored throughout the process until gas flooding no longer induced changes in the core mass, at which point the core was considered to have attained irreducible water saturation. Following each test, the core was weighed to determine its water saturation for the corresponding pressure differences. The relationship curve between the displacement pressure difference and water saturation is shown in Figure . Relationship curve between displacement pressure difference and water saturation. Analysis of water saturation in different cores under varying pressure differentials demonstrated that water saturation decreases with an increasing pressure differential. This dependence was utilized to determine the core’s initial water saturation before the displacement experiment. The irreducible water saturation of each core was also determined through this experiment. The specific data are shown in Table . 5Core Water SaturationCore numberCore porosity (%)Irreducible algebraic equation water saturation (%)Target water saturation (%)621.6629.5420722.2031.0030821.9535.4540919.8430.61501020.7231.99601123.8835.38100 Core Water Saturation The target and irreducible water saturations of the cores presented in the table established the foundation for the subsequent investigation of the influence of different water saturations on the critical pressure differential. Critical Drawdown Pressure In the Single-Phase Displacement Observation of the quality changes in various cores under different critical pressure differentials revealed significant alterations at specific pressure thresholds. The critical pressure difference test experiments were carried out on ten experimental cores, and the permeability and quality changes of the cores are shown in Figure . Change of core quality and permeability under different pressure differences for (a)– (j) Cores 1–10. As shown in Figure , the pressure differential that causes a sudden change in the core quality is therefore considered to be the critical sand production pressure differential. The critical point was defined as the threshold pressure differential at which the cumulative sand mass exhibited a sharp increase, overreaching the initial baseline fluctuations. A series of parallel experiments revealed that the critical sand production pressure differential ranged from 2.2 to 3.0 MPa. It should be noted that the loss during the process of removing the core was eliminated during the experiment. The specific sanding critical pressure differences are shown in Table . 6Critical Pressure Difference of Core SampleCore numberCritical pressure differential (MPa)Core numberCritical pressure differential (MPa)12.662.622.673.032.282.242.892.652.6102.8 Critical Pressure Difference of Core Sample In the Two-Phase Displacement To this end, and based on the previous results, we set the water saturations of cores 6, 7, 8, 9, 10, and 11 as 20%, 30%, 40%, 50%, 60%, and 100%, respectively. Each core then underwent a test to determine its critical sand production pressure difference. The specific data records are given in Table . 7Core Water SaturationCore numberExperiment typeInitial core quality (g)Initial water saturation of core (%)Core quality after saturation (g)6Critical pressure difference experiment52.942054.14750.163051.67854.934057.34955.535058.471057.496061.541155.3710061.53 Core Water Saturation The change in core quality with displacement pressure is shown in Figure . Change of core quality with pressure difference under different water saturations : (a) 20%, (b) 30%, (c) 40%, (d) 50%, (e) 60%, and (f) 100%. Figure demonstrates that at various water saturations (20%, 30%, 40%, 50%, 60%, and 100%), the core mass remained stable near the irreducible water saturation until a sudden shift occurred. It was concluded that the pressure differential applied when the abrupt mass change occurred represented the critical pressure differential for sand production. The critical sand production pressure differentials for Cores 6, 7, 8, 9, 10, and 11 were determined to be 5.0, 4.8, 4.8, 4.6, 4.8, and 5.0 MPa, respectively. Consequently, the subsequent alternating displacement experiment was conducted with pressure difference limits set between 0.5 MPa (lower) and 4.5 MPa (upper). Alternating Displacement Experiments Effect of Alternating Times on Sand Production In order to reduce the random error, five repeated experiments were carried out with five dry similar cores (6–10). A pressure difference of 0.5 and 2.5 MPa was applied alternately in the experiment, and each pressure difference lasted for 1 min. The cumulative changes in permeability and mass were recorded, and their average values were analyzed to observe the effect of increasing the number of alternation cycles. The changes in permeability and the changes in sand production with mass are analyzed, as shown in Figure . The experimental results demonstrate that both the core permeability and the sand production volume increase during the alternating displacement process. Change of core sand production and permeability during the alternating change of confining pressure. (a) Sand production with different alterations. (b) Permeability with different alterations. The results show that in the initial 0–15 alternate displacement process, the degree of particle binding in the core is weak. At this time, a large number of small-volume particles are located in the pores. These particles are taken away from their original positions under the action of alternating airflow shear force and pressure fluctuation. The migration of a large number of particles leads to the rapid expansion of the existing flow channels. The “dominant flow path” is formed inside the core. In turn, the permeability and sand production are significantly increased. During the alternating 15–30 cycle stages, the slowdown in growth is the result of the combined effects of multiple mechanisms. After the easily mobilized sand is largely washed away in the initial stages, the remaining particles require greater drag force or longer fatigue failure time to initiate movement due to stronger cementation or larger particle size. At the widened pore throats, larger particles tend to form temporarily stable “bridging” structures; meanwhile, some of the migrating fine particles can become lodged in smaller pore throats, causing localized blockage. These factors collectively temporarily suppress the increase in sand production and permeability. Furthermore, although alternating stress accumulates fatigue damage at the contact points of the skeletal particles, it does not reach the critical point for large-scale fracture. Therefore, this stage becomes a critical turning point characterized by energy accumulation and temporary equilibrium. After more than 30 cycles of alternating stress, the growth accelerated again, marking the entry of the rock core into an accelerated failure period characterized by structural instability. The core mechanism lies in the fact that alternating stress cycles cause fatigue damage to accumulate at the particle contact points until a critical point is reached, leading to the large-scale loosening and detachment of the previously stable framework particles, providing a new source of sand with larger particle sizes. Simultaneously, the temporary bridge-like structures formed during the transitional period are destroyed at this stage. They are not only washed away themselves but also trigger a chain reaction of destruction, rapidly expanding the pores and flow channel network. Furthermore, the advantageous flow channels formed in the early stages begin to be directly eroded by the airflow, effectively “cleaning” the channels. As the channel continues to widen, once the airflow velocity exceeds the critical flow velocity again, its ability to carry particles increases nonlinearly, thus efficiently carrying away more eroded and loosened particles. Figure shows the changes in the particle migration behavior within the rock core at different stages. Changes in particle migration within rock cores at different stages. Effect of Alternating Cycle on Sand Production Alternating displacement is essentially a loading–unloading stress cycle. The longer the single displacement time, the longer the residence time in the state of a high-pressure difference. For the airflow to stably carry particles out of the core sample, a certain critical flow velocity must be reached and maintained. Under continuous high stress, the contact points between particles have more time for fatigue microcracks to initiate, propagate, and eventually coalesce, leading to irreversible damage to more strongly bonded framework particles. The experiment sets different time intervals to change the injection–production direction and observes the sand production of the core by changing the frequency of alternating injection–production. Cores 1–5 were subjected to alternating cycles with periods of 30 s, 1 min, 2 min, 3 min, and 5 min, respectively, to observe the corresponding sand production under each condition. In the absence of observed sand production, the analysis was therefore directed toward the variation in core permeability with the number of alternation cycles. A linear fit was subsequently applied to characterize the relationship between the permeability and alternation frequency. Specific data are shown in Figure . Permeability changes of cores with different alternating cycles. A comparison of the fitting curve slopes across different periods revealed that longer alternating periods correspond to a steeper increase in permeability with cycles, demonstrating that extended injection–production cycles accelerate permeability growth and consequently elevate the sand production risk. The changes in the permeability before and after the experiment are shown in Table . 8Change of Alternating Injection–Production Core during Different PeriodsCore numberFrequencyDuration (min)Alternating upper and lower limits (MPa)Initial permeability (mD)Permeability after experiment (mD)Permeability change (%)130 s50.5/2.5141.14143.571.6921 min10156.86171.409.2632 min20151.60167.8610.7243 min30133.91149.0111.2855 min50120.70155.7829.06 Change of Alternating Injection–Production Core during Different Periods The variations in the core permeability before and after the experiment were compared across multiple alternating cycles. Following 50 alternating cycles with periods of 30 s, 1 min, 2 min, 3 min, and 5 min, the core permeability increased by 1.69%, 9.26%, 10.72%, 11.28%, and 29.06%, respectively. The experimental results show that the length of the injection–production cycle is the key control factor affecting the core permeability change and sand production. The longer the cycle, the more significant the change of permeability and the higher the sand production. In the short period, the duration of the effective fluid velocity exceeding the critical sand-carrying velocity is short, and the fluid can only displace the movable particles that are already on the migration path. The long period provides a longer supercritical flow rate action time so that the fluid has enough time to convert more into movable solid particles due to strength weakening. Effect of Water Saturation on Alternate Injection–Production Sand To investigate the effect of water saturation on sand production after alternating displacement, core samples were initially saturated with water to achieve varying water saturation levels. Due to the limited quantity of experimental materials, this study tests only one representative core sample under each experimental condition. To ensure the rigor and reliability of the research conclusions, this study employed a strict multivalidation method. All experiments were conducted under standardized conditions to eliminate the influence of environmental fluctuations on the results. Alternating gas displacement experiments were performed on rock cores with water contents ranging from 20% to 60% and also at 100% under alternating pressure amplitudes of 0.5/4.5 MPa. The mass change of the core and the gas flow rate under a fixed pressure were tracked as the number of alternations increased, with the results shown in Figure . Variation of cores with different water saturations: (a) 20%, (b) 30%, (c) 40%, (d) 50%, (e) 60%, (f) 100% with the number of alternations. This study used regression analysis to investigate the relationship between sand production and water saturation. Due to the limited availability of high-precision experimental data points, the Newton interpolation method was selected over the least-squares method (typically used for extrapolation and prediction). Tables and list the changes of core quality and permeability before and after the experiment. As shown, a polynomial model was applied, which can relate sand production to changes in core permeability and water saturation. 9Quality Change of Core before and after Alternating Displacement under Different SaturationsCore NumberInitial quality (g)Water saturation (%)Quality after experiment (g)Core quality after drying (g)Mass difference (g)652.942052.4152.260.68750.163049.7249.430.73854.934055.0754.310.62955.535055.6154.930.61057.496058.3457.100.391155.3710057.1655.110.26 Quality Change of Core before and after Alternating Displacement under Different Saturations 10Permeability Change of Core before and after Alternating Displacement under Different SaturationsCore NumberInitial permeability (mD)Water saturation (%)Permeability of core after drying (mD)Permeability variation (%)6143.5720159.6611.217176.6430193.219.388206.4340228.6510.769198.6350222.4712.0010226.3660259.5214.6511193.70100230.8619.19 Permeability Change of Core before and after Alternating Displacement under Different Saturations Its key advantage is that it forces the fitting curve through every data point, providing an accurate representation of trends and variable relationships. The resulting trends are presented in Figure . As shown in Figure (a), there is a nonmonotonic relationship between sand production and water saturation, exhibiting an overall trend of “increase–decrease–increase”. Figure (b) shows that the change in core permeability with water saturation before and after the experiment also follows a nonmonotonic pattern, specifically exhibiting a “first decrease, then increase” trend. Fitted curve of sand production and permeability. (a) The relationship between core sample sand production and water saturation. (b) The relationship between core permeability variations and water saturation. In the experiment, it was observed that when the saturation of water was in the range of 0% to 30%, the sand production gradually increased, while the core permeability first decreased and then increased. This phenomenon is believed to be that under the action of continuous cyclic stress, the skeleton of the core gradually produces fatigue damage, resulting in the shedding of weakly bonded particles. These particles migrate with the gas phase and gradually accumulate at the pore throat to form a blockage, which is macroscopically manifested as an increase in sand production. At the same time, because of the obstruction of pore channels, the permeability decreases accordingly. When the water saturation is in the range of 30% to 85%, the amount of sand produced gradually decreases, while the permeability increases accordingly. The main reason for the decrease in sand output is that the water film formed on the surface of the sand particles enhances the adhesion effect. Specifically, the residual water film covering the surface of the sand and gravel provides a significant cohesive force, and the particles can be further fixed through physicochemical adsorption. These two mechanisms together inhibit particle migration. Meanwhile, the increase in water saturation exacerbates water erosion. On the other hand, the fluid carries some migrating particles, which scour and enlarge the pore channels, potentially forming new microfractures or pore networks. This restructuring of the pore structure expands the effective fluid flow paths, thereby improving the overall permeability of the rock cores. The observation shows that when the water saturation rises to 85%–100%, the sand production increases again. When the saturation exceeds 85%, the liquid phase aggregates and causes the apparent cohesion to disappear. At the same time, under the condition of high water cut, the hydrodynamic drag effect is intensified, which causes the secondary damage of the rock skeleton, and the sand production finally shows an upward trend. These two factors together promote the migration of sand grains. However, due to the limited number of experimental samples within this saturation range, the generality of this trend and its quantitative relationship still require further verification with more data. According to the model fitting results (Figure ), the R 2 values for this interval reached 0.9788 and 0.9965, respectively, indicating that the model has excellent fitting performance within the current range of observed data. Conclusion In this study, a series of experiments were conducted to examine sand production during alternating displacement. By comparing sand production under single-phase and two-phase conditions, the key factors were determined. The principal findings of this research are summarized below:(1)The critical sanding pressure differential increases markedly from 2.2–3.0 MPa in dry cores to 4.6–5.0 MPa in water-bearing cores, which results from the water film binding effect.(2)Within the range of 15–30 alternations in the single phase, the growth rate of core permeability and sand production decelerates. Beyond 30 alternations, a notable increment is observed with further increases in alternations. Additionally, a longer injection–production cycle will increase the possibility of sand production.(3)The sand production shows a nonmonotonic trend with the increase of water saturation. When the water saturation ranges from 0% to 85%, the sand production first increases and then decreases. The maximum sand production is observed at the critical water saturation of 30%, which represents the most critical state for the stability of the sandstone skeleton in this reservoir. When the saturation is close to the saturation limit, sand production rises again.(4)The core permeability decreases first and then increases with the change in water saturation before and after the displacement experiment. When the water saturation is 20%, the change in permeability is the smallest. (1)The critical sanding pressure differential increases markedly from 2.2–3.0 MPa in dry cores to 4.6–5.0 MPa in water-bearing cores, which results from the water film binding effect. The critical sanding pressure differential increases markedly from 2.2–3.0 MPa in dry cores to 4.6–5.0 MPa in water-bearing cores, which results from the water film binding effect. (2)Within the range of 15–30 alternations in the single phase, the growth rate of core permeability and sand production decelerates. Beyond 30 alternations, a notable increment is observed with further increases in alternations. Additionally, a longer injection–production cycle will increase the possibility of sand production. Within the range of 15–30 alternations in the single phase, the growth rate of core permeability and sand production decelerates. Beyond 30 alternations, a notable increment is observed with further increases in alternations. Additionally, a longer injection–production cycle will increase the possibility of sand production. (3)The sand production shows a nonmonotonic trend with the increase of water saturation. When the water saturation ranges from 0% to 85%, the sand production first increases and then decreases. The maximum sand production is observed at the critical water saturation of 30%, which represents the most critical state for the stability of the sandstone skeleton in this reservoir. When the saturation is close to the saturation limit, sand production rises again. The sand production shows a nonmonotonic trend with the increase of water saturation. When the water saturation ranges from 0% to 85%, the sand production first increases and then decreases. The maximum sand production is observed at the critical water saturation of 30%, which represents the most critical state for the stability of the sandstone skeleton in this reservoir. When the saturation is close to the saturation limit, sand production rises again. (4)The core permeability decreases first and then increases with the change in water saturation before and after the displacement experiment. When the water saturation is 20%, the change in permeability is the smallest. The core permeability decreases first and then increases with the change in water saturation before and after the displacement experiment. When the water saturation is 20%, the change in permeability is the smallest.