The velocity sensitive test fluid was 8% standard brine (NaCl: CaCl₂: MgCl₂-6H₂O = 7: 0.6: 0.4). From the test results (Figure 6), it can be seen that the ratio of K_n/K_i increases significantly with increasing flow rate for all three samples (K_n is the original permeability and K_i is the tested permeability), and the critical flow rate ≤ 1 mL/min. According to the industry standard, velocity sensitivity damage rates are greater than 70%, and the degree of damage is strong. Sediment particles in NGH reservoirs are small, and because they are weakly consolidated, the particles are easily detached and transported with fluid flow. An acid flow rate that is too low in acidizing modification will lead to uneven pore development. Therefore, to ensure the successful implementation of acidizing modification, the sediment skeleton must be solidified; otherwise, the risk of sand production is exceptionally high.

The NGH sediments in the Shenhu Sea have a high clay mineral content. Clay minerals such as Montmorillonite will expand when encountering low-salinity injection water, and Kaolinite will migrate when encountering low-salinity water, reducing reservoir permeability. The water sensitivity test was conducted using 8% standard brine. The intermediate test fluid was 4% standard saline. The change in permeability ratio during injection is shown in Figure 7. The permeability ratios of the three samples increased slightly with decreasing fluid mineralization. According to the industry standard, water sensitivity damage rates are 5%–30%, which is weak water sensitivity. Montmorillonite is the main water-sensitive mineral, followed by illite and kaolinite. Although the NGH sediments are high in clay minerals, montmorillonite is the least abundant, so the samples are weakly water-sensitive. The damage rate of all three samples in distilled water is less than 20%, and there is no need to conduct salinity reduction sensitivity experiments.

The acid sensitivity experiment reflects the acid’s effect on the reservoir’s permeability. Hydroxide precipitates, fluoride precipitates, fluorosilicates, and gels produced by the acid and reservoir minerals may block the pores. Acidizing disrupts the reservoir’s original structure, aggravating velocity sensitivity, and the released clay particles swell and transport, leading to decreased permeability.

Permeability was measured before and after acidizing using an 8% KCl solution. The acid solution for the sensitivity test was 12% PORE-A + 3% Perm-A. The permeability ratio change curve is shown in Figure 8. Samples 1 and 3 showed increased permeability after acidizing, indicating an improvement in sample porosity. Sample 2 showed no significant change in permeability after acidizing, probably because dislodged particles blocked some of the pores. According to industry standards, acid sensitivity damage rates are less than 5%, and NGH reservoirs can be acid stimulated.

The alkali sensitivity test fluid was 8% KCl solution, and the KOH solution changed its pH. The permeability ratio is shown in Figure 9. Sample 2 had a slight decrease in permeability during the pH increase, sample 1 had a slight increase in permeability, and sample 3 had no significant change in permeability. According to industry standards, Sample 2 has an alkali sensitivity damage rate of 5%–30%, which is a weak alkali sensitivity. Sample 1 and 3 have an alkali sensitivity damage rate of less than 5%, which is no alkali sensitivity damage.

With the internal fluid output, reservoir pore pressure reduction will change the reservoir’s original force balance, leading to changes in the pore structure and affecting the fluid flow in the reservoir. Stress sensitivity factors include reservoir lithology, cementation strength, pore structure, particle sorting, and pore pressure change rules.

The use of back-pressure variations brings it closer to the reality of the NGH reservoir. Refer to the GMGS3-W19 station model Wan (Wan et al., 2018) et al. developed during hydrate depression mining. The water depth of 1,273.9 m is converted to a top pressure of 12.86 MPa, and the in-situ stress gradient is 0.02597 MPa/m, which calculates the overburden pressure to be approximately 16.3 MPa. The initial pressure at the bottom boundary of the hydrate layer was 14.3 MPa, and the net overburden pressure was calculated to be 2 MPa, which was used as the initial net stress value of the sample. Considering the low bottomhole pressure in the late production period, the minimum bottomhole pressure was set to 2.3 MPa. The confining pressure was used to replace the overburden pressure. The initial backpressure value was 14.3 MPa, which was varied sequentially to 11.3, 8.3, 5.3, and 2.3 MPa, and then slowly increased to 14.3 MPa, all with a net stress interval of 3 MPa.

The stress sensitivity curve is shown in Figure 10. The critical stress of all three samples was 8 MPa, the stress-sensitive damage rate was around 30%–40%, and the irreversible permeability damage rate was around 30%, with moderate stress sensitivity.

Inject the consolidation agent into the sample at a displacement of 0.5 mL/min and evaluate the changes in permeability, porosity, compressive strength, and pore throat structure after 6 h of consolidation. The experimental results are shown in Table 3.

The increase in fluid viscosity during consolidation agent injection leads to a rise in drive pressure. During the infusion of the consolidation agent, the permeability decrease was significant, and the driving pressure increased up to 7.8 MPa (Figure 13: K_n is the original permeability, and K_i is the tested permeability). The consolidation agent occupies some of the pore space, resulting in a decrease in the porosity of the samples (Figure 14). The retention of permeability of the consolidated samples was around 56%, and the retention of porosity was 38%–41%. From the CT scanning results (Figure 15B), it can be clearly seen that the pore diameter of the sample after consolidation is obviously reduced, the pore distribution is more uniform, and the pore distribution in the exit section of the sample is denser compared with that in the middle section and the entrance section, which is similar to that before consolidation, indicating that the system of the consolidation agent has been uniformly infused into the sample. It can be seen from the slice map that the “small black dots” of the sample before consolidation are denser, and the “big black dots” are occasionally seen. After solidification, the sample is more dense, the distribution of “small black dots” is more sparse, and the pore size of “large black dots” is obviously smaller. As seen from the 3D and slice map, the pores of the consolidated samples were not completely blocked but retained some porosity and connectivity.

As shown in Figure 16, the mechanical properties of the consolidated sample are improved. The average compressive strength of the unconsolidated samples was 1.57 MPa, the average Young’s modulus was 144.05 MPa, and the average Poisson’s ratio was 0.27. The average compressive strength of the consolidated samples was 2.43 MPa, the average Young’s modulus was 228.76 MPa, and the average Poisson’s ratio was 0.24. After consolidation, the compressive strength and Young’s modulus of the sample increase, and Poisson’s ratio decreases, indicating that the resistance to deformation of the sample increases.

The samples were subjected to acidizing experiments using three acid solutions: 8% PORE-A + 1% Perm-A, 5% PORE-A + 5% PORE-B, and 10% solid acid α. The results of the experiment are shown in Table 4.

The results of the 8% PORE-A + 1% Perm-A acid solution experiments showed that: The skeleton of the unconsolidated sample was severely deformed, and serious sand production occurred during the experiment. From the CT scan (Figure 17A), it can be seen that the skeleton at the injection end of the unconsolidated sample is difficult to maintain and undergoes severe deformation, and there is a clear interface in the middle part of the sample, suggesting that the silt transported to this place blocked the pore throat channel, resulting in the pore space not being developed at the outflow end of the core. It indicates that there is a problem with sand and gravel transport in the unconsolidated sample. In contrast, the skeleton of the consolidated sample (Figure 17B) was only slightly deformed, the acid reacted uniformly with the minerals in the flow direction, and no sand production occurred during the experiment, indicating that the consolidation agent can effectively prevent sand production and improve the stability of the sediment skeleton. The acidizing of the consolidated samples increased the permeability by 120.97% and porosity by 82.94%.

Similarly, the 5% PORE-A + 5% PORE-B and 10% solid acid α experiments showed that the solidified samples had better stability and were less prone to sand production. Figure 18 shows similar results to Figure 17. In Figure 18A, it can be clearly seen that the pore space at the injection end of the unconsolidated sample is very well developed. Still, the pore space at the outflow end is not clearly developed. There is a clear interface in the middle part of the sample, which suggests that the skeleton produces a serious amount of sand, and that the sand and gravel transportation blocks the seepage passage. Figure 18B shows the CT scan of the consolidated sample after acidizing, the result shows that the pore development of the consolidated sample is uniform, and the sample does not show sanding and deformation problems. Figure 19 shows the results of the 10% solid acid α flow experiment. Because of the slower reaction rate of solid acid α, the unconsolidated samples (Figure 19A) did not show significant pore throat plugging, but less pore development was observed in the unconsolidated samples compared to the consolidated samples (Figure 19B), which suggests that minor sanding may have occurred in the unconsolidated samples. The compressive strength test results also show that the consolidated samples have better compressive strength.

Unlike conventional reservoirs, NGH reservoirs are highly heterogeneous in terms of sediment distribution and hydrate saturation (Ren et al., 2022a; Ren et al., 2022b), which complicates consolidation-acidizing modification of NGH reservoirs. In the South China Sea, NGH sediments are predominantly clayey-silty sediments, and the presence of natural gas in the reservoir is in the form of cements at contracts, grain coating and pore-filling. The hydrate pore-filling type has the strongest cementing effect on the sand and gravel. With the further dissociation of gas hydrate, the sand and gravel cannot be effectively supported, which causes changes in the reservoir structure, and may cause sand production or collapse. Preventing reservoir sand production and maintaining reservoir stability are important aspects of developing NGH reservoirs. In this paper, we take the mud silt type weakly cemented NGH reservoir as the research object, and innovatively propose the consolidation-acidizing method of NGH reservoir modification. The experimental results show that the consolidant can effectively improve the strength and stability of the sediment skeleton, maintain the skeleton morphology during the hydrate dissociation process, and inhibit the sand production of the reservoir. Compared with the unconsolidated samples, the consolidated samples were stable in acid modification, not easy to produce sand, and had uniform pore development after acidification. It shows that consolidation-acidizing modification is feasible in natural gas reservoirs.