Application of Dual Horizontal Well Systems in the Shenhu Area of the South China Sea: Analysis of Productivity Improvement

Abstract

The horizontal well technology was successfully applied in the Chinese second natural gas hydrate (NGH) field test in the Shenhu area of the South China Sea in 2020. However, the results show that the threshold for commercial exploitation has not been broken, judging from daily gas production and cumulative gas production. Consequently, the paper presents the effects of dual horizontal well systems for exploitation in this area. The NGH reservoir model in the Shenhu area was established with CMG software. The influence of various layout options and various spacing of dual horizontal well systems on the production capacity was investigated. Further, we simulated the production effect of dual horizontal well systems joint auxiliary measures, such as well wall heating, heat injection, etc. The results show that the production capacity of dual horizontal well systems increased by about 1.27~2.67 times compared with that of a single horizontal well. The daily gas production will drop significantly, no matter which method was used, when exploitation lasts for about 200 d. Meanwhile, well wall heating and heat injection have limited effects on promoting production capacity. In conclusion, attention was drawn to the fact that the synergistic effect could be fully exerted to accelerate NGH dissociation when dual horizontal well systems are applied. The NGH reservoirs in the Shenhu area may be more suitable for short-term exploitation. The research results of this paper can provide a reference for the exploitation of the Shenhu area.

1. Introduction

Natural gas hydrate (NGH), a clathrate formed under low-temperature and high-pressure environments [1,2], usually contains various gases, among which the methane gas content generally accounts for more than 99% of the total gas. According to statistics, the amount of total carbon content contained in NGH is about twice that of other fossil energy sources in the world [3,4]. Due to the advantages of being clean and having a high energy density, a wide distribution range, and huge resource reserves [5], NGH is considered to be the unconventional geological energy with the most development potential in the 21st century [6]. Meanwhile, about 97% of NGH occurs in seabed sediments, and only about 3% occurs in terrestrial permafrost [7]. Consequently, it has become an international consensus to actively explore the means of commercial exploitation of marine NGH in the context of global energy transition, despite the uninterrupted discussions and controversies about the impact of the development of marine NGH on environmental climate and seabed disasters [8].

Up to now, scholars have carried out a series of studies on the exploitation methods of NGH through laboratory experiments, numerical simulation, theoretical analysis, and other means. Common methods to exploit NGH include the depressurization method, inhibitor injection method, thermal stimulation method, CO₂ replacement method, solid-state fluidization, etc. [9,10]. The principles of all these methods are changing the temperature and pressure conditions of NGH and then promoting NGH dissociation for exploitation. Among them, the depressurization method is considered to be the NGH exploitation method with the most potential due to its advantages of low cost and high production efficiency [11]. As shown in

Table 1. Summary of the field production tests by Japan and China [12,13,14,15,16].

	Field Location	Year	Production Method	Average Rate (m3/d)	Cumulative Gas Volume (m3)
Japan	Eastern Nankai Trough	2013	Depressurization	119,500	19,917
		2013	Depressurization	41,000	3417
				222,500	9271
China	Shenhu area	2017	Solid-state fluidized	/	81
	Liwan area	2017	Depressurization	30,900	5150
	Shenhu area	2017	Depressurization	861,400	28,713

, Japan [12,13] and China [14,15,16] have carried out five field trial productions for marine NGH from 2013 to 2020. Four of them used the depressurization method.

The vertical well, for some time, will still be the primary wellbore structure for field tests of NGH due to its lower technical threshold and operating costs [8]. Consequently, many scholars have carried out relevant research on production behaviors and stimulation measures when using vertical wells to exploit NGH. Yin et al. studied the effect of various wellbore designs on gas production behaviors when using vertical wells to exploit NGH and found that water production could be reduced to a certain extent by adjusting the location of perforations [17]. Zhao et al. explored the production behaviors of vertical wells combined with a stepwise depressurization strategy and found this method can help reduce the occurrence of NGH regeneration, but the total gas production will decrease [18]. Zhang et al. studied the effect of an enlarged wellbore radius on total methane production and water production [19]. The results show that when the well wall radius increased from 0 to 5 m, the gas and water production increased by four times and three times, respectively, but the exploitation cost may be much higher. Wang et al. investigated the gas production behavior of NGH in porous media by indoor experiments using dual vertical wells combined with thermal stimulation and found that the higher the injection rate, the lower the energy efficiency [20]. Jin et al. conducted research on the multiple vertical well system and found that the gas production decreased due to the exhaustion of NGH between wells when using a three-well system for exploitation [21]. All of these studies have shown that the NGH production capacity could be improved to a certain extent by vertical wells combined with various optimization strategies, but it is still not enough to break through the production threshold.

The horizontal well technology has been applied for the first time in the second field test of China in the Shenhu area of the South China Sea in 2020 [8]. A continuous time of 30 days and an average gas production rate of 2.87 × 10⁴ m³/d were achieved in this field test, which was a breakthrough from “exploratory test exploitation” to “experimental test exploitation”. Furthermore, this result strongly proves that horizontal well technology is a reliable means to realize the commercial exploitation of NGH. However, the threshold for NGH commercial exploitation, as suggested by scholars, is still not met in terms of exploitation efficiency. To resolve the current issue, continuous optimization of horizontal well technology is necessary. Dual horizontal wells may be an effective stimulation measure. Moridis et al. investigated the NGH production potential of dual horizontal wells under various production strategies by numerical simulation [22]. The results show that dual horizontal wells can effectively improve gas production performance, which provided a basis for selecting an efficient exploitation strategy. Yu et al. analyzed the production stimulation effects of dual horizontal wells in different spatial locations and found that the productivity was much greater than that of a single horizontal well [23]. Feng et al. studied the production behavior of dual horizontal wells combined with hot water injection based on a cubic three-dimension simulator. The results show that compared with vertical wells combined with heat injection, the energy efficiency ratio was improved [24]. Shang et al. found a synergistic effect can be produced between wells to accelerate NGH dissociation and reduce water production when dual horizontal wells are used for exploitation [25].

The above research is a preliminary exploration of dual horizontal wells technology. Comparative studies on the production behavior of dual horizontal wells and their combination with different auxiliary measures have not yet been revealed regarding the single NGH reservoir geological conditions. In this paper, the production behavior of dual horizontal well systems and their combination with various auxiliary measures were studied based on the data of the second field test in the Shenhu area of the South China Sea. The research about the layout option, the spacing, different step-down ranges of dual horizontal wells, the combined heat injection of dual horizontal wells, and the combined well wall heating are expected to provide a reference for the exploitation NGH which occurs in the Shenhu area.

2. Methodology

2.1. Geological Conditions

The Chinese second field test site is located in the Shenhu area on the northern slope of the Baiyun Sag in the Pearl River Mouth Basin in the northern South China Sea [26]. The water depth is about 1000~1500 m. In this area, fluid migration has formed many channels, gas chimneys widely developed, and high-angle faults connect with deep source rocks which form fluid migration channels for NGH enrichment and accumulation. As shown in Figure 1, the well GMGS6-SH02 is the pilot hole of the second field test, which is only about 70 m from the trial exploitation target. Qin et al. interpreted and analyzed the geophysical prospecting and drilling data at the well GMGS6-SH02 by various means, and obtained data with high accuracy [27]. Consequently, in this paper, choose the formation condition of the well as the basis for the simulation. The lithology of this well section is dominated by argillaceous silt, and the minerals are mainly divided into argillaceous, sandy, and calcareous, accounting for 47.2%, 36.4%, and 10.2%, respectively. The well can be divided into three layers from top to bottom, namely the NGH-bearing layer (NBL), mixed layer (ML), and free gas layer (FGL), with thicknesses of 45.6 m, 24.6 m, and 19 m, respectively. The average effective permeability is 2.38 mD, 6.63 mD, and 6.8 mD, respectively.

Table 2. The reservoir parameters of well GMGS6-SH02.

Reservoir	Thickness/m	Porosity/%	Permeability/mD	Saturation/%
NGH-bearing layer (NBL)	45.6	37.3	2.38	Sh = 31
Mixing layer (ML)	24.9	34.6	6.63	Sh = 11.7/Sg = 13.2
Free gas layer (FGL)	19	34.7	6.8	Sg = 7.3

shows the specific geological parameters of the well. Figure 2 shows the well GMGS6-SH02 logging integrated histogram.

2.2. Numerical Simulator

CMG is a commercial simulator widely used in conventional oilfield exploitation, in which the STARS module enables the simulation of NGH exploitation in complex phase states by adding components [5]. Meanwhile, CMG STARS can be used to simulate the NGH exploitation process with various patterns such as complex structure well. Consequently, due to its many advantages above, CMG has been widely used in the field of NGH development to further predict gas production capacity and provide guidance for on-site production. In the CMG simulator, NGH can be defined as a solid-state or high-viscosity oil. The reservoir porosity and permeability caused by the formation and decomposition of NGH cannot be considered when NGH is defined as high-viscosity oil. Consequently, in this paper, NGH is defined as the solid state to better simulate the gas-water two-phase seepage process.

The equation for NGH dissociation is as follows:

$$C{H}_4\cdot N_h{H}_2{O}_{(solid)}\leftrightarrow C{H}_4{}_{(gas)}+N_h{H}_2{O}_{(liquid/ice)}$$

(1)

In the equation, N_h is the number of water molecules bound by NGH.

Based on the dissociation kinetic model proposed by Kim et al. [28], the dissociation rate equation of NGH is:

$$\frac{\mathrm{d}{c}_h}{\mathrm{dt}}=K_d{A}_d\left(P_e-P_g\right)$$

(2)

In the equation, C_h is the NGH concentration; K_d is the NGH dissociation rate constant; A_d is the NGH dissociation area; P_e is the phase equilibrium pressure; P_g is the gas phase pressure in porous media.

Based on the Arrhenius formula, the decomposition rate constant could be defined as:

$$K_d=K_d^0\exp \left(-\frac{E}{RT}\right)$$

(3)

In the equation, K is the internal dissociation rate of NGH; E is the activation energy of the reaction; R is the universal gas constant; T is the reaction temperature.

Combining Equations (2) and (3), the NGH dissociation and synthesis rate could be obtained as:

$$\frac{\mathrm{d}{c}_h}{\mathrm{dt}}=K_d^0{A}_{dec}\left(P_e-P_g\right)\exp \left(-\frac{\mathsf{\Delta }E}{RT}\right)$$

(4)

2.3. Numerical Model

The specific geological parameters of the overburden layer (OL) and underlying layer (UL) of the well GMGS6-SH02 location are not publicly reported in the literature. Therefore, when establishing the numerical model, referred to the well SHSC-4 data. As shown in Figure 1b, the well SHSC-4 location and the well GMGS6-SH02 location are closely adjacent. Meanwhile, Zhao et al. have demonstrated its reliability in a related study [29]. Consequently, the simulation is reliable. The top surface of the model corresponds to the sea floor. In both the x and y directions, the model was extended to 1220 m to avoid boundary effects. In the z direction, the total thickness is 501 m, to optimize the numerical model, the thickness of each reservoir is finely adjusted, which is divided into OL (210 m), NBL (45 m), ML (25 m), FGL (20 m), and UL (210 m) from top to bottom. The whole model is discretized into 163,724 grids (61 grids in both X and Y directions, 44 grids in Z direction). The model grid is non-uniformly distributed to better observe the physical changes of the NGH-bearing reservoir, and the grid near the NGH-bearing layer is encrypted. The initial pressure field of the model obeys the hydrostatic pressure. The initial pressure at the bottom of the NGH reservoir is 14.78 MPa. The initial temperature at the top of the NGH reservoir is 12.8 °C and the geothermal gradient is 4.02 °C/100 m.

Table 3. Numerical simulation model parameters.

Parameter	Value	Unit
NGH molar mass	0.119543	kg/gmole
NGH density	919.7	kg/m3
The heat of dissociation of NGH	51,858	J/mole
Thickness	OL = UL = 210	m
	NBL = 45
	ML = 25
	FGL = 20
Initial temperature of the top of NGH	12.7	°C
Initial pressure of the top of NGH	14,120	kPa
Formation temperature	T = 12.7 + 0.0402 × z, (z: depth, m)	°C
Formation pressure	P = 14,120 + 10 × z, (z: depth, m)	kPa
Permeability	OL = UL = 6, NBL = 2.38, ML = 6.63, FGL = 6.8	mD
Saturation	NBL: Sh = 0.31	/
	ML: Sh = 0.117; Sg = 0.132
	FGL: Sg = 0.073
Thermal conductivity of rock	1.0	W·(m·K)−1
Thermal conductivity of water	0.9	W·(m·K)−1
NGH thermal conductivity	3.1	W·(m·K)−1
Porosity change model	$\phi ={\phi }_o\left(1-S_i\right)$
Capillary force model	$P_{\mathrm{c}}=-P_{co}{\left[{\left(S^{*}\right)}^{-\frac{1}{\lambda }}-1\right]}^{1-\lambda }{S}^{*}=\frac{S_w-S_{wr}}{1-S_{wr}-S_{gr}}$

In the table, S_i is the ice saturation; ϕ and ϕ_o are the effective porosity and initial porosity; P_c is the capillary force, N; P_co is the capillary force endpoint value, N; λ is the Van Genuchten parameter; S_wr and S_gr are the residual water saturation and the residual gas saturation.

shows the specific parameters of the numerical model.

3. The Dual Horizontal Well System Case Design and Simulation Results

The length of the horizontal wells should be 250–300 m, in relatively “high saturation” NGH reservoirs according to the simulated production result in the Chinese second field test. Meanwhile, the NGH occurrence characteristics of the reservoir can be better utilized to maximize the recovery capacity when the horizontal well is located on the top of the ML. Consequently, in this paper, the length of the horizontal well section is 300 m, and the diameter (d) of the horizontal well is 0.25 m. In addition, the production pressure (P) is set to 4.5 Mpa, and the production time is set to 500 d to better reveal the NGH production behavior.

3.1. Different Layout Options of Dual Horizontal Well Systems

Table 4. Production parameters for the cases of different well types and layout option.

Well Type	No.	Layout Option	Pw/MPa	Time/d
Single horizontal well	Case 1-1	/	4.5	500
Dual horizontal well	Case 1-2	Parallel (horizontal)	4.5	500
	Case 1-3	Parallel (up and down)	4.5	500
	Case 1-4	Cross (up and down)	4.5	500

shows production parameters for the cases of different well systems and layout options. Figure 3 shows the spatial location distribution of a single horizontal well system and the different layout options of dual horizontal well systems. The single horizontal well (Figure 3a, case 1-1) is located in the center of the model, and the two wells of the dual horizontal well system are, respectively, set up to be horizontally parallel (Figure 3b, case 1-2), up and down parallel (Figure 3c, case 1-3), and up and down cross (Figure 3d, case 1-4). The synergistic effect when the two wells of the dual horizontal well system are located in different NGH layers for exploitation was better revealed by the above case designs. For case 1-2, the horizontal spacing is set to 100 m. For case 1-3, the upper well is located in the center of the NBL, and the lower well is located in the ML and is 30 m from the top surface of the ML. For case 1-4, both two wells are located in the ML, with the upper one 10 m from the top surface of the ML and the lower one 35 m.

The daily gas production (Q_d) and cumulative gas production (V) over time for the four cases are shown in Figure 4. The results show that the V in the first 30 days of the single horizontal well system is 79 × 10⁴ m³, which is comparable to the gas production (86 × 10⁴ m³) in the Chinese second field test, indicating the validity of the model. Meanwhile, regardless of the layout options of the dual horizontal well systems, its Q_d and V curves are always higher than that of the single horizontal well. For cases 1-1,2, and cases 1-3,4, the changing trends of the Q_d curves are substantially similar. At the beginning of exploitation, the NGH around the well dissociates rapidly, and the generated gas quickly enters the wellbore, so Q_d reaches a high level instantly. The Q_d drops rapidly and enters the stable gas production period when the gas around the well is collected. In this stage, the Q_d curve first declines slowly and then rises slightly, and when Q_d rises to the maximum value, it continues to decline until the end of exploitation. It is worth noting that this phenomenon appeared first (140 d) in case 2-2, and the rising rate was also the largest. In the other three cases, this phenomenon did not appear until 160 d after exploitation. The reason may be that the influence range of pressure drop gradually expanded outwards in the early period of exploitation. During this process, the NGH dissociated and absorbed heat, which led to a drop in the surrounding temperature of the exploitation area. Consequently, Q_d also began to decrease slowly. The NGH dissociation speed is accelerated again when the heat from the surrounding formations is gradually transferred to the exploitation area. Meanwhile, the exploitation area affected is larger due to the layout of the two wells in horizontal parallel in case 2-2, so this phenomenon appears the earliest and is the most obvious.

Similarly, from cases 1-2~4, in the first 250 d of exploitation, the Q_d of case 1-2 is the highest at 9.2 × 10⁴ m³, case 1-4 is next at 5.3 × 10⁴ m³, and case 1-3 is the lowest at only 3.9 × 10⁴ m³. The above-mentioned difference in Q_d occurs because when the two wells of the dual horizontal well system are horizontally parallel, the influence range of the pressure drop is the largest, while the two wells that are “up and down cross” are the smallest. After 250 d of exploitation, similarly to case 1-1, Q_d all decreased to about 2.2 × 10⁴ m³, which indicates that dual horizontal well systems are more effective in short-term exploitation. In case 1-2, the two Q_d curves of the two wells are similar. In case 1-3, the Q_d curve of the well (well 1) located in the NBL, after reaching the maximum value at the beginning, is always in a declining state which has dropped to 7.78 × 10³ m³ at the end of exploitation. In another well (well 2) located in the ML, its Q_d was always rising for the first 300 days of exploitation. The natural gas production in well 1 mainly originates from the dissociated gas in the ML, while in well 2, it mainly originates from the free gas. The reservoir porosity and permeability around well 1 are small; therefore, after reaching the maximum value at the beginning of exploitation, the dissociated gas in the distance takes a longer time for the generated gas to flow into. The Q_d curves of the two wells in case 1-4 are similar. This also reaffirms the point that horizontal wells should be located in the ML for this NGH reservoir condition of the Shenhu area to achieve the best gas production capacity.

In addition, the average Q_d of cases 1-2~4 is about 5.3 × 10⁴ m³, 3.2 × 10⁴ m³, and 2.5 × 10⁴ m³, respectively. The average Q_d and V of cases 1-2~4 are about 2.67 times, 1.62 times, and 1.27 times that of case 1-1, respectively. The cumulative gas production of the horizontal wells when they are horizontally parallel and exceeded that of a single horizontal well twice, which proves that the synergistic effect can be fully exerted when exploited by using the dual horizontal well system to obtain greater benefits.

3.2. Different Spacing of Dual Horizontal Well Systems

In Section 3.1, the results show that the gas production capacity improves the most obviously when the two wells of the dual horizontal well system are horizontally parallel. Consequently, because of this situation, the influence of the spacing between two wells on the gas production capacity was further analyzed. Six cases (cases 2-1~6) were established, and the spacings are 100 m, 200 m, 300 m, 400 m, 500 m, and 600 m, respectively. The production parameters are shown in

Table 5. Production parameters for the cases of adjustment of spacing.

No.	Spacing/m	Pw/MPa	Time/d
Case 2-1	100	4.5	500
Case 2-2	200	4.5	500
Case 2-3	300	4.5	500
Case 2-4	400	4.5	500
Case 2-5	500	4.5	500
Case 2-6	600	4.5	500

. The simulation results are shown in Figure 5.

The results show that the gas production behaviors of cases 2-1~6 are similar in the first 20 days of exploitation, during which the Q_d is high in the early exploitation and then decreases rapidly. For cases 2-1~3, it enters the stage of stable gas production after Q_d drops to about 8.7 × 10⁴ m³, 9.8 × 10⁴ m³, and 10.8 × 10⁴ m³, respectively; the larger the spacing the earlier the enter time. When the spacing is 400 m or more, Q_d began to rise when it dropped to the lowest and finally began to enter this stage around 50 d after exploitation. Meanwhile, when the spacing is small, the NGH between the wells can quickly disassociate and gradually flow into the production well. When the spacing increases, the NGH in the central part of the wells does not have time to start to disassociate until the pressure drop area gradually expands; natural gas will gradually be generated. The six Q_d curves showed an overall upward trend with an increase in the spacing in 20~200 d. Meanwhile, the growth rate is largest when the spacing increases from 100 to 200 m, and as the spacing continues to increase, the growth rate begins to decrease and the Q_d curves almost coincide for cases 2-5,6. The increase in spacing helps to a certain extent in terms of the V. But the increase slows down and even decreases to a certain extent over time when the spacing increases above 400 m. From the perspective of the overall daily gas production, the Q_d of cases 2-4~5 is the largest at 11.6 × 10⁴ m³, and the Q_d of case 2-1~3 gradually decreased to only about 9.5 × 10⁴ m³. This may be due to the phenomenon of “excessive overlap” in the exploitation area affected by pressure drop when the well spacing of the two wells is within 300 m and the synergistic effect between the two wells cannot be fully used and failed to maximize the area affected by the pressure drop. When the spacing is increased above 400 m, this phenomenon will be alleviated. There is not much difference in the V of case 2-4~6, they are all about 3.1 × 10⁷ m³ which is 1.17 times higher than that of case 2-1. Combining Q_d and V, and considering the investment and related technical feasibility for the characteristics of NGH reservoirs in the Shenhu area, the spacing of 400 m is more appropriate.

3.3. Different Production Pressure of Dual Horizontal Well Systems

The production pressure is the main control parameter for NGH exploitation by depressurization. Usually, increasing the magnitude of pressure reduction can significantly improve the production capacity of NGH to some extent [2]. In Section 3.1 and Section 3.2, the production pressure is 4.5 Mpa. Consequently, the production behavior of the dual horizontal well systems at different production pressures is further explored in this section. The layout option of the dual horizontal well system is still horizontally parallel with a spacing of 400 m to exclude the influence of other factors. A total of four cases are established; the production pressures are 4.5 Mpa, 3.5 Mpa, 2.5 Mpa, and 1.5 Mpa, respectively.

Table 6. Production parameters for the cases of different production pressure.

No.	Spacing/m	Pw/MPa	Time/d
Case 3-1	400	4.5	500
Case 3-2	400	3.5	500
Case 3-3	400	2.5	500
Case 3-4	400	1.5	500

shows specific production parameters.

As shown in Figure 6, Q_d increases significantly as the production pressure decreases. Meanwhile, the overall trend of the Q_d curves of cases 3-1~4 is similar. Q_d dropped to the lowest value and rose when exploitation was carried out for about 10~14d. The lower the production pressure, the earlier the rising time will be, and the greater the increase rate. This process will last for about 30 d. Subsequently, Q_d entered a stable stage lasting about 150 d. In this stage, Q_d does not change much overall, but there will be a trend of first falling and then rising, which is most obvious in case 3-4. Within 50~200 d of exploitation, the gas production is between 26.8 × 10⁴ m³~32.3 × 10⁴ m³, then quickly dropped to 18.8 × 10⁴ m³ and maintain a slow declining trend until the exploitation ended. Although there are still obvious differences in Q_d under different production pressures in the middle and late stages of exploitation, none of them can meet the minimum requirements of commercial production. In addition, the V of cases 3-2~4 reaches a maximum of 8.6 × 10⁷ m³. Compared with case 3-1, the V of cases 3-2~4 increased to 1.44 times, 2.02 times, and 2.75 times, respectively. Consequently, for the Shenhu area, short-term exploitation can obtain the greatest benefits. Comparing the V curves of case 3-1 and case 3-4, the cumulative gas production at 110 d of exploitation has exceeded the cumulative gas production of the entire exploitation time (500 d), which again shows that reducing production pressure is an effective way to increase production capacity. Nevertheless, reducing production pressure not only leads to wellbore damage but also causes uncontrollable changes in the surrounding formations. Consequently, in order to achieve the commercial exploitation of NGH by this means, it should not only increase the wellbore strength by various measures but also develop a specific depressurization strategy based on the mechanical properties of NGH reservoirs.

4. The Dual Horizontal Well Systems Combined with Auxiliary Measures

4.1. Well Wall Heating

Scholars have found that ice and secondary NGH will be generated during the exploitation process through laboratory experiments and numerical simulations which will block the wellbore and affect the exploitation efficiency. The reason is that NGH dissociation is an endothermic reaction, and at the same time, is subject to the “Joule Thomson effect”. In addition, the lower the production pressure, the more easily this problem arises because the faster decomposition of the NGH and the more heat absorbed. Fetisov et al. proposed a pipeline temperature control system to achieve the best gas transportation state [30]. Heating the well wall during exploitation also can effectively solve this problem. At present, the cable heating technology provided can achieve a cable heating power of about 10~150 w/m by technical service providers. Consequently, six cases were established to investigate the variation of gas production behavior at different heating powers (50 w/m, 100 w/m, 150 w/m) combined with different production pressures (p = 4.5 Mpa, 1.5 Mpa).

Table 7. Production parameters for the cases of well wall heating.

No.	Spacing/m	Pw/MPa	Heating Power/(w·m−1)	Time/d
Case 4-1	400	4.5	0	500
Case 4-2	400	4.5	50 w/m	500
Case 4-3	400	4.5	100 w/m	500
Case 4-4	400	4.5	150 w/m	500
Case 4-5	400	1.5	0	500
Case 4-6	400	1.5	50 w/m	500
Case 4-7	400	1.5	100 w/m	500
Case 4-8	400	1.5	150 w/m	500

shows the specific case design.

Figure 7 shows the variation of Q_d and V over time. Results show that heating the well wall can promote gas production capacity to an extent, regardless of the production pressure. This phenomenon is especially evident at the beginning of exploitation due to the rapid dissociation of NGH in the vicinity of the exploitation well caused by the heating of the well wall. However, after the peak, although there is a certain increase in gas production efficiency with the increase in heating power, it is almost negligible which indicates that the well wall heating has a limited range of action and can only promote the dissociation of NGH around the well. But it is enough to prevent ice and secondary hydrates from blocking the wellbore. At the end of exploitation, the V of cases 4-2~4, and 4-5~8 only increased about 5~6% compared without heating the well wall, indicating that the improvement in NGH production by well wall heating is small. It also shows that wellbore plugging may not be the main factor affecting the long-term production efficiency of NGH. Consequently, for the Shenhu area, the exploitation method of the dual horizontal well system depressurization combined with a heating well wall is not appropriate.

4.2. Heat Injection

The productivity of NGH could be improved to a certain extent by combining heat injection with depressurization. Yu et al. investigated the gas production capacity of a dual horizontal well system (well length is 1000 m, well spacing is 90 m), based on the NGH reservoir conditions in the Nankai Trough, Japan [31]. The simulation results proved that the average daily NGH production capacity in this area can reach as high as 8.6 × 10⁵ m³ under the conditions of heat injection at 40 °C and a heat injection rate of 2 kg/(s·m). Consequently, in this paper, we conducted a similar study for the Shenhu area. A total of nine cases were established to explore the influence of dual horizontal well system depressurization combined with a heat injection on gas production behavior. Meanwhile, the results were compared to the single depressurization. The layout option of the dual horizontal well system is still horizontally parallel with a spacing of 400 m. Meanwhile, the results were compared to the single depressurization. Among them, the injection well is located in the middle of the two wells in the horizontal direction. In the vertical direction, it is located in the middle of NBL, ML, and FGL, respectively to study the influence when they are located in different layers. The injection temperature is 40 °C, and the injection rates are 40 m³/d, 80 m³/d, and 120 m³/d, respectively. The specific production parameters are shown in

Table 8. Production parameters for the cases of heat injection.

Injection Well Location	No.	Spacing/m	Water Injection Rate/(m3·d−1)	Time/d
/	Case 5-0	400	/	500
Middle of NBL	Case 5-1	400	40	500
	Case 5-2	400	80	500
	Case 5-3	400	120	500
Middle of ML	Case 5-4	400	40	500
	Case 5-5	400	80	500
	Case 5-6	400	120	500
Top of FGL	Case 5-7	400	40	500
	Case 5-8	400	80	500
	Case 5-9	400	120	500

.

Figure 8 shows the simulation results. The Q_d curves of cases 5-1~9 trends are similar. The average daily gas production of all cases is around 6.4 × 10⁴ m³. Results show that the production capacity is slightly improved when the injection well is located in the NBL, but it is reduced when the injection well is located in the ML and FGL. The reason may be the low permeability of the reservoirs in the Shenhu area, and the heated fluid is not easy to diffuse. As a result, a high-pressure area is formed around the injection well, which is not conducive to NGH dissociation. Consequently, taking measures to increase the permeability of the reservoir could solve this problem when using the dual horizontal well system depressurization combined with heat injection to exploit. In addition, the use of non-fluid heating can not only overcome the potential engineering geological risks of heat injection but also improve the production efficiency of NGH. In terms of cumulative gas production, the V only increases by 0.5% even if the injection well is located in the NBL and the injection rate is 120 m³/d. Combining Q_d and V, taking the dual horizontal well system depressurization combined with heat injection to exploit is not necessary.

5. Conclusions

In this paper, the production behavior of dual horizontal well systems and their combination with various auxiliary measures are studied, based on the second field test data in the Shenhu area of the South China Sea. The conclusions are as below:

(1)

The production capacity can be significantly improved for the NGH occurrence characteristics in the Shenhu area when using the dual horizontal well systems to exploit, compared with a single horizontal well.

(2)

The production efficiency is the highest when the horizontal wells are horizontally parallel and the spacing is 400 m. There is an “excessive overlap” phenomenon in the pressure drop-affected areas of the two wells when the spacing is less than 300 m, which obstructs the further improvement of production efficiency. The synergistic effect between the two wells could be fully utilized when the spacing is 400 m or more. The cumulative gas production has reached 2.7 times that of a single horizontal well. Consequently, it is recommended to consider the geological conditions and investment to determine a reasonable spacing when exploitation.

(3)

The effect of reducing production pressure is significant in improving exploitation capacity. However, the daily gas production will drop sharply when exploitation is carried out for about 200 d. Therefore, during short-term exploitation, choosing a lower production pressure is more appropriate.

(4)

The dual horizontal well systems depressurization combined with auxiliary measures such as well wall heating and heat injection have a limited impact on increasing production. Consequently, these two exploitation methods are less helpful in breaking the threshold of commercial exploitation of NGH in the Shenhu area and are not recommended to use.