Numerical Simulation of Vertical Well Depressurization-Assisted In Situ Heating Mining in a Class 1-Type Hydrate Reservoir

Abstract

In situ electric heating is an important method used to increase production capacity during the extraction of natural gas hydrates. This work numerically evaluated the sensitivity of different heating parameters on gas production behavior with a vertical well depressurization in the Shenhu Sea area hydrate reservoir, the production pressure difference of 4 MPa, and continuous depressurization for 1080 days. The results showed that the in situ electric heating method can effectively enhance production capability by promoting hydrate dissociation and eliminating secondary hydrates. Compared with scenarios without heating, implementing whole wellbore heating (100 W/m) increases cumulative gas production (V_g) by 118.56%. When intermittent heating is applied to the local wellbore (15 m) located in the three-phase layer (with an interval of 30 days) and stops heating in advance at 480 days, there is no significant difference in V_g compared to the whole wellbore heating case, and the cumulative heat input is only 4.76%. We recommend considering intermittent heating of the local wellbore and stopping heating in advance during vertical well depressurization as this approach significantly reduces heating energy consumption while simultaneously improving production capacity.

1. Introduction

Natural gas hydrates (NGHs) are compact crystalline substances from the combination of water and gas [1]. Deep sea and permafrost geological environments are suitable places for the formation and occurrence of hydrates [1,2,3]. NGH resources are considered a significant alternative energy source for future global sustainable development [3,4]. The recent offshore NGH trial production completed in the South China Sea (SCS) and Japan’s Nankai Trough verified the efficiency of the depressurization method, but their production capacity is still far below the standard for industrial development [5,6,7,8,9]. To achieve industrial development of NGHs, various studies likewise contribute relevant information and background on environmentally friendly, safe, and efficient exploitation of NGHs (e.g., Okwananke et al., 2019; Zhang et al., 2019; Farahani et al., 2021) [10,11,12]. Currently, the most promising method for achieving industrial development is the complex structured well or well-pattern production mode combined with the composite production method (such as assisted-thermal extraction) and/or reservoir stimulation technologies (such as hydraulic fracturing) [13]. The depressurization-assisted thermal extraction method can be divided into two categories as follows: one is fluid-assisted thermal mining methods (such as hot water injection, geothermal assistance, etc.), and the other is non-fluid assisted thermal mining methods (such as electric heating, microwave heating, in situ backfill heating, etc.) [13]. The fluid-assisted thermal mining method is prone to forming high-pressure injection zones, which inhibits hydrate dissociation and makes injection in low-permeability reservoirs problematic; the method also has the problem of significant heat loss [13]. The non-fluid-assisted thermal mining method is less likely to cause reservoir instability and other issues and is thus especially suitable for low-permeability reservoirs [13]. In recent years, the non-fluid-assisted thermal mining method has attracted a lot of attention and research effort. For example, Zheng et al. (2018) conducted a sensitivity analysis on the dissociation front of hydrates under the conditions of depressurization combined with wellbore heating [14]. Zhao et al. (2021) used depressurization assisted with the heating method to extract Class 2-type hydrate reservoirs, and their numerical results confirmed the efficiency of the method [15]. Xu et al. (2021) numerically validated the effectiveness of extracting NGHs by depressurization combined with in situ backfilling CaO powder [16]. Li et al. (2021) found that when using the five-point method to extract hydrate reservoirs, the recovery rate of hot water injection was almost the same as that of electric heating combined with water injection, with only wellbore heating suitable for low-permeability hydrate reservoir extraction [17]. Zhao et al. (2021) analyzed the production performance of the NGHs in the SCS using a mathematical model of low-frequency electric heating and depressurization (LF-EHAD). Their results showed that after more than 5 years of continuous production, the gas production of LF-EHAD is 2.37 times that of depressurization [18]. Liu et al. (2022) found that intermittent heating has higher energy utilization efficiency than continuous heating in depressurization-assisted thermal for extracting hydrates [19]. Zhao et al. (2022) numerically validated the effectiveness of using LF-EHAD to develop Class 1-type hydrate deposits with five-point wells [20]. Wang et al. (2022) found that microwaves can quickly respond to the energy demand for hydrate dissociation [21]. Zhang et al. (2022) discovered that combining brine flooding and electric heating can result in low heat loss and enhanced thermal convection [22]. Zhang et al. (2023) numerically compared the recovery efficiency of different assisted heating methods under dual horizontal well depressurization mining. They found that LF-EHAD can promote the dissociation of hydrates and the heat distribution is more uniform than wellbore electric heating, avoiding wellbore instability caused by continuous high temperatures [23]. Previous research work has effectively improved the technical feasibility of the non-fluid-assisted thermal method in NGH mining. The group well mining scheme is an inevitable choice for the future industrial development of hydrates, and the vertical well is an important basic well type of the group well mining scheme. It is necessary to further study the application strategy of assisted heating methods in vertical well depressurization.

On-site testing is difficult and costly, and numerical simulation methods are the best choice for studying the large-scale development of NGHs. Some well-known simulators include TOUGH+HYDRATE, CMG-STARS, MH21-HYDRES, STOMP-HYD, etc. [24,25,26,27]. Except for TOUGH+HYDRATE and CMG-STARS, which are widely used, other simulators are limited to use by development departments because of technical confidentiality and intellectual property reasons. TOUGH+HYDRATE is specifically developed for hydrate exploitation simulation and has the best applicability, although its disadvantage is weak pre- and post-processing. CMG is not specifically developed for simulating hydrate extraction, and its description of hydrate characteristics may be insufficient, but it has more advantages in pre- and post-processing. This work adopted TOUGH+HYDRATE and numerically analyzed the effects of different heating parameters such as heating power, heating location, heating time, and heating mode on gas production in a Class 1-type hydrate reservoir with a single vertical well depressurization and proposed an efficient application strategy for assisted in situ electric heating. The research results provide a reference for hydrate extraction in the Shenhu Sea area.

2. Methodology

2.1. Site Description

Class 1-type hydrate reservoirs consist of an upper hydrate layer and an underlying two-phase fluid layer containing free gas and water. The SHSC4 (Figure 1) well’s NGH reservoir is classified as Class 1-type with three layers as follows: (1) the gas hydrate-bearing layer (GHBL) contains water and hydrates, with a buried depth of 201–236 m below seafloor (mbsf); (2) the three-phase layer (TPL) contains water, free gas, and hydrates, with a buried depth of 236–251 mbsf; and (3) the free gas layer (FGL) contains water and free gas, with a buried depth of 251–278 mbsf [8,28,29].

2.2. Simulator

TOUGH+HYDRATE V1.0 comprehensively considers depressurization, inhibitors, thermal stimulation, and combined methods. It includes equilibrium and kinetic models for describing the dissociation and formation of hydrates, considering components including water, methane, salt, and hydrate. The phase states include water, gas, ice, and hydrate. By adopting the fully implicit equation form, continuously updating physical property parameters, and decoupling the nonlinear equations of heat and flow using the Newton iteration method, phase change calculations can be efficiently completed [24]. The efficiency of this code has been experimentally and on-site verified [31,32]. We adopted the parallel version of this code and the equilibrium model in this work [33,34]. During the simulation process, it was assumed that Darcy’s law is applicable, and geological mechanical responses were ignored. The main equations of the code are as follows [24]:

• Multi-components and -phases.

Multicomponent (κ): methane (m), water (w), salt (i), and hydrate (h); multiphase (β): hydrate (H), aqueous (A), gas (G), and ice (I).

2.

Mass conservation equation.

The mass conservation equation of the code is as follows:

$${\displaystyle \frac{d}{dt}}{\int }_{V_n}{M}^{\kappa }dV={\int }_{{\Gamma }_n}{F}^{\kappa }\cdot \mathrm{n}d\Gamma +{\int }_{V_n}{q}^{\kappa }dV$$

(1)

The component’s mass accumulation, flux, and source/sink ratio are represented by $M^{\kappa }$, $F^{\kappa }$, and $q^{\kappa }$ in the above equation.

3.

Energy conservation equation.

The energy conservation equation of the code is as follows:

$${\displaystyle \frac{d}{dt}}{\int }_{V_n}{M}^{\theta }dV={\int }_{{\Gamma }_n}{F}^{\theta }\cdot \mathrm{n}d\Gamma +{\int }_{V_n}{q}^{\theta }dV$$

(2)

The heat component, heat accumulation, flux, and source/sink ratio are represented by θ, $M^{\theta }$, $F^{\theta }$, and $q^{\theta }$ in the above equation.

2.3. Model Construction

The size of the model is 1000 × 1000 × 117 m in (x, y, z) (Figure 2a). The thicknesses of the underburden (UB) and overburden (OB) are set to be 20 m [35,36]. This work created twelve simulated cases to assess the impacts of various heating parameters, with each case including a 70 m long vertical well deployed at the center of the model (model’s −21 and −91 m) with a wellbore radius of 0.1 m. The reservoir properties used in the simulation are summarized in

Table 1. Detail setting of simulation cases.

ID	Heating Location	Heating Power	Heating Mode	Time to SHIA
Case 01	-	-	-	-
Case 02	WWH	100 W/m	CH	-
Case 03	WWH	200 W/m	CH	-
Case 04	WWH	300 W/m	CH	-
Case 05	WWH	400 W/m	CH	-
Case 06	WWH	500 W/m	CH	-
Case 07	LWH	100 W/m	CH	-
Case 08	LWH	100 W/m	CH and SHIA	120 d
Case 09	LWH	100 W/m	CH and SHIA	240 d
Case 10	LWH	100 W/m	CH and SHIA	360 d
Case 11	LWH	100 W/m	CH and SHIA	480 d
Case 12	LWH	100 W/m	IH and SHIA	480 d

Note: WWH is whole wellbore heating; LWH is local wellbore heating; CH is continuous heating; SHIA is stop heating in advance; IH is intermittent heating.

.

The model is discretized into 37 × 37 × 81 grids in (x, y, z), with a total of 110,889 grids (Figure 2b). Here, the x-y plane discretization is carried out separately. To capture subtle changes in the physical properties of the reservoir around the wellbore, the mesh near the wellbore is appropriately refined, the minimum grid size near the wellbore is set to be 1.0 m and then extruded along the z-axis into a 3D discretized grid. The grid size of the hydrate reservoir along the z-axis is set to be 1.0 m.

2.4. Model Initialization

To obtain the model’s initial pressure, temperature, and saturation steady-state conditions (Figure 3), we separately simulated three subdomains (GHBL, TPL, and FGL). We achieved consistent heat flux between the subdomains’ contact surfaces by fine-tuning the geothermal gradient. Finally, we merged them to complete the initialization of the model [37,38,39,40,41,42,43]. The capillary pressure calculation adopted the van Genuchten model, where the maximum reference aqueous saturation S_mxA is 1, the irreducible aqueous saturation S_irA is 0.30, the porosity distribution index λ is 0.45, and the initial capillary pressure P₀ is 1 × 10⁴ Pa [24,44,45,46]. The capillary force varies with the change in aqueous saturation S_A. The relative permeability calculation adopted the Stone model, where K_rA is the aqueous phase relative permeability and K_rG is the gas phase relative permeability. The irreducible aqueous saturation S_irA and irreducible gas saturation S_irG are 0.30 and 0.03, respectively, and the aqueous phase permeability reduction index nA and gas phase permeability reduction index nG are 3.5 and 2.5, respectively [24,44,45,46]. K_rA varies with the change in aqueous saturation S_A, while K_rG varies with the change in gas saturation S_G. The above model parameter settings are based on previous research work (Li et al., 2018; Cao et al., 2023; Sun et al., 2019; Ma et al., 2020) that verified the applicability of these models and parameters in this sea area [44,45,46]. During the simulation process, the wellbore grids were considered as a “pseudo porous medium” and given a fixed production pressure difference of 4MPa. The porosity of the wellbore grids is 1, and the permeability is 10,000 D, without considering capillary force, the relative permeability varies linearly with the saturation of each phase and has extremely low residual gas saturation [43]. The assumption was made that the OB and UB layers had a permeability of 2.0 mD and porosity of 0.3, respectively [44]. The detailed parameters of the model are summarized in

Table 2. Model parameters.

Parameter	Value and Unit
OB and UB thickness [35,36]	20 m
GHBL thickness [8,44,45,46]	35 m
TPL thickness [8,44,45,46]	15 m
FGL thickness [8,44,45,46]	27 m
OB and UB initial permeability	2.0 mD
GHBL initial permeability [8,44,45,46]	2.9 mD
TPL initial permeability [8,44,45,46]	1.5 mD
FGL initial permeability [8,44,45,46]	7.4 mD
Wellbore radius [44,45,46]	0.1 m
Salinity [44,45,46]	3.5%
GHBL and TPL initial hydrate saturation [8,44,45,46]	Extracted from logging curve (Figure 2a)
FGL initial free gas saturation [8,44,45,46]	Extracted from logging curve (Figure 2a)
OB and UB porosity	0.30
GHBL porosity [8,44,45,46]	0.35
TPL porosity [[8,44,45,46]	0.33
FGL porosity [8,44,45,46]	0.32
Grain density [44,45,46]	2600 kg/m3
Geothermal gradient [44,45,46]	43.653 °C/km
Grain specific heat [44,45,46]	1000 J·kg−1·K−1
Gas composition [44,45,46]	100% CH4
Dry thermal conductivity [24,44,45,46]	1.0 W·m−1·K−1
Wet thermal conductivity [24,44,45,46]	3.1 W·m−1·K−1
Composite thermal conductivity [24,44,45,46]	KΘ = kdry + (SA1/2 + SH1/2)(kwet − kdry) + φSIλI
Capillary pressure model [24,44,45,46]	$P_{cap}=-P_0{\left[{\left(S^{\ast }\right)}^{-{\displaystyle \frac{1}{\mathsf{\lambda }}}}-1\right]}^{1-\mathsf{\lambda }},S^{\ast }={\displaystyle \frac{\left(S_A-S_{irA}\right)}{\left(S_{mxA}-S_{irA}\right)}}$
Maximum reference aqueous saturation SmxA [24,44,45,46]	1
Porosity distribution index λ [24,44,45,46]	0.45
Initial capillary pressure P0 [24,44,45,46]	104 Pa
Relative permeability model [24,44,45,46]	KrA $=\left[\right(S_A-$SirA$)/(1-$SirA$){]}^{\mathrm{nA}},K_{\mathrm{rG}}=[(S_G-$SirG$)/(1-$SirA)]nG
Permeability reduction index for aqueous nA [24,44,45,46]	3.5
Permeability reduction index for gas nG [24,44,45,46]	2.5
Residual gas saturation SirG [24,44,45,46]	0.03
Residual aqueous saturation SirA [24,44,45,46]	0.30
Permeability reduction model [24,44,45,46]	${\displaystyle \frac{k}{k_0}}=F_{\varphi S}={\left({\displaystyle \frac{\varphi -{\varphi }_c}{{\varphi }_0-{\varphi }_c}}\right)}^n$
Critical porosity φc [24,44,45,46]	0.05
Permeability reduction index n [24,44,45,46]	3

.

2.5. Model Validation

The model’s effectiveness was validated by the data from China’s first round of offshore NGH trial production. A vertical wellbore was placed in the center of the model with a length of 70 m (model’s −21 and −91 m) and a pressure difference of 3 MPa [47]. As shown in Figure 4, the fitting results indicate that the gas production rate Q_g and cumulative gas production V_g both are within the acceptable range. We used the same model and mesh discretization as the model validation in this study.

3. Results and Analysis

3.1. Gas and Water Production

Figure 5a depicts the trends in Q_g and V_g both for the cases without and with whole wellbore heating (WWH) and continuous heating (CH) within 1080 days of production. The Q_g in the heating case (heating power set to 100 W/m) is higher than in the case without heating, and in the case assisted with heating, the V_g increases from 3.420 × 10⁶ to 4.055 × 10⁶ ST m³, an increase of 118.56%. The simulation results show that heating technology has a good effect on increasing the production of hydrate depressurization mining. It is worth noting that after approximately 420 days of production, the Q_g of the case without heating begins to exceed that of the heating case, and then the Q_g of both is consistent around 900 days. This may be caused by the following reasons: firstly, around 420 days, the bottom warm water of the case without heating began to enter the wellbore, and the high-saturation secondary hydrates formed at the TPL began to accelerate dissociation. Secondly, in the heating case, because the high saturation secondary hydrates were not formed around the wellbore, and the gas around the wellbore was quickly extracted, while the gas further away from the wellbore slowly into the wellbore. Thirdly, the non-fluid heating method can only heat the reservoir at a certain distance around the wellbore, and the heating effect gradually weakened in the later stage. The water production rate Q_w and gas-to-water ratio R_gw evolutions within 1080 days of production are depicted in Figure 5b. The Q_w curve trends in every case are in line with the Q_g and show that heating has a positive effect on gas production as well as making it easier to produce water, which raises the Q_w. The heating results in a slightly higher R_gw than the case without heating in the first 720 days.

3.2. Reservoir Physical Property Evolution

3.2.1. Pore Pressure

Figure 6 presents the evolution diagram of pore pressure (P) for the cases without and with heating within 1080 days. In the pressure field diagram of the case without heating, it can be observed that the wellbore located in the TPL is severely affected by secondary hydrates, resulting in limited pressure propagation. Especially at 360 days, the saturation of the secondary hydrates is highest and the pressure limitation is most obvious. In the heating case, the wellbore located in the TPL did not form significant high-saturation secondary hydrates throughout the entire production cycle, so its pressure propagation was not restricted.

3.2.2. Reservoir Temperature

Figure 7 presents the evolution diagram of reservoir temperature (T) for the cases without and with heating within 1080 days. Initially (60 days), a low-temperature area in the TPL of the reservoir close to the wellbore can be observed in the case without heating. This is because there is a substantial Joule–Thomson effect due to the significant amount of highly saturated free gas entering the wellbore from the TPL, and the wellbore contact area is only around 9.4 m². With the warm bottom water entering the wellbore, the reservoir temperature around the wellbore gradually increases during the latter period (360 days to 1080 days). In the temperature field diagram (60 days) of the heating case, this phenomenon can also be observed. A heating power of 100 W/m is not enough to completely offset the Joule–Thomson cooling effect located in the TPL. Therefore, a corresponding low-temperature region is also observed here. However, as production progresses, the temperature of the reservoir around the wellbore gradually increases because of the weakening of the Joule–Thomson effect, sustained electrical heating, and bottom warm water entering the wellbore.

3.2.3. Hydrate and Gas Saturation

Figure 8 depicts the evolution of hydrate saturation (S_h) for both cases without and with heating within 1080 days. According to the hydrate saturation field diagram for the case without heating, high-saturated secondary hydrates were formed in the reservoir near the wellbore in the TPL after 60 days of production and secondary hydrate saturation peaked about 360 days later. The temperature of the reservoir near the wellbore rises as bottom warm water enters the wellbore, causing the high-saturation secondary hydrates in the TPL to dissociate faster, boosting the gas production rate Q_g. Only a limited quantity of low-saturated secondary hydrates accumulated in the reservoir around the TPL’s wellbore during the first production stage (60 days) and peaked at roughly 360 days, as seen in the heating case’s hydrate saturation field diagram. However, in the later stage, all secondary hydrates in this region were decomposed. Under a 4 MPa pressure difference, 100 W/m of heating power is insufficient to avoid secondary hydrate formation around the wellbore. Meanwhile, it is noticed that the hydrate dissociation front advances quicker and deeper in the heating condition. After 1080 days of production, the heating case has a wider range of hydrate dissociation in the GHBL, around 2–5 m.

Figure 9 presents the evolution diagram of gas saturation (S_g) for the cases without and with heating within 1080 days. Comparing the gas saturation field diagrams, it can be found that because of heating, the reservoir around the wellbore maintains a smooth flow channel. At the same time, more gas accumulates around the wellbore of the heating case, and most of the high-saturation free gas near the wellbore is extracted around 360 days. In the field diagram of the case without heating, it can be observed that the high-saturation secondary hydrates generated in the reservoir around the TPL block the flow channel here and force the high-saturation free gas near the wellbore located in the TPL to migrate to the upper and lower reservoirs and wellbore, respectively.

4. Discussion

The heating power, heating location, heating time, and heating mode are the primary parameters that can be changed during production. This section discusses the impact of these primary parameters on productivity.

4.1. Effects of Heating Power

Figure 10 depicts the effects of heating power on the V_g and R_gw within 1080 days with WWH&CH. When the heating power is increased from 100 to 200, 300, 400, and 500 W/m, the V_g increases from 4.055 × 10⁶ to 4.120 × 10⁶, 4.161 × 10⁶, 4.197 × 10⁶, and 4.227 × 10⁶ ST m³, increasing by 101.60%, 102.61%, 103.05%, and 104.24%, respectively. The results show that a heating power of 100 W/m is sufficient to prevent the formation of secondary hydrates in the TPL when the production pressure difference is set to 4 MPa. Increasing heating power further does not significantly increase gas output. A heating power of 100 W/m offers an almost identical heating effect as a 500 W/m heating power but with less heating energy consumption. The R_gw in the aforementioned cases stays around 200 after 1080 days of production, showing no discernible change.

To further investigate the relationship between the calorific value of increased production and the additional heat to drive production, we introduced a custom production efficiency index EROI_custom based on previous work [48,49,50,51]. The specific definition is as follows:

EROI_custom = E_O/E_I

(3)

where E_O is the energy output, which is the increased gas production times the heating value of natural gas (approximately 35.8 MJ/m³), and E_I is the energy input, which is the energy consumed by in situ electric heating [48,49,50,51]. Figure 11 shows the EROI_custom curves with different heating power. All curves show a trend of first increasing and then decreasing and tend to remain constant in the later stage. The peak value of the curve indicates that heating has the most significant effect on increasing production in the initial stages of production. Among them, a heating power of 100 W/m has the best energy output-to-input ratio.

4.2. Effects of Heating Location

Figure 12 depicts the effects of heating location on the V_g and R_gw within 1080 days. When adopting local wellbore heating (LWH) and CH, separately heating the 15 m wellbore located in the TPL, the V_g after 1080 days of production is 3.982 × 10⁶ ST m³. Under the same heating power of 100 W/m, compared with WWH&CH, the V_g just decreases by 1.8%. This also indicates that in the Class-1 type hydrate reservoir, when adopting vertical well depressurization production, the secondary hydrates generated in the reservoir around the TPL are the key factor affecting production capacity. The cumulative heat input of LWH&CH is only 21.42% of that of WWH&CH. After 1080 days of production, the R_gw in the above cases remains around 200, and there is no significant difference.

As shown in Figure 13, the EROI_custom curve of local wellbore heating (LWH) is significantly higher than that of whole wellbore heating (WWH), which indicates that LWH heating has a better energy output-to-input ratio.

4.3. Effects of Heating Time

Figure 14 depicts the effects of the heating time on the V_g and R_gw within 1080 days with LWH&CH. With stop heating in advance (SHIA) at 120, 240, 360, and 480 days, the V_g decreases from 3.982 × 10⁶ to 3.716 × 10⁶, 3.827 × 10⁶, 3.890 × 10⁶, and 3.931 × 10⁶ ST m³, decreasing by 6.68%, 3.89%, 2.31%, and 1.28%, respectively. The simulation results indicate that when LWH&CH is applied to the wellbore located in the TPL and stops heating in advance at 480 days, the V_g just decreases by 1.28% during the 1080-day production cycle. However, the cumulative heat input of LWH&CH&SHIA (480 d) is only 44.44% of the LWH&CH heating production. Compared with the WWH&CH production, the V_g only decreases by 3.06%, but the cumulative heat input is only 9.52%. After 1080 days of production, the R_gw in the above cases remains around 200, and there is no significant difference.

Figure 15 shows the EROI_custom curves of continuous heating and stop heating in advance. Compared with continuous heating, all EROI_custom curves that stop heating in advance show an increasing trend at the time of stopping heating, followed by a decreasing trend, and tend to stabilize in the later stage. Obviously, stop heating in advance has a better energy output-to-input ratio than continuous heating.

4.4. Effects of Heating Mode

Figure 16 depicts the effects of the heating mode on the V_g and R_gw within 1080 days with LWH&SHIA (480 d). When the heating mode changes from CH to intermittent heating (IH) and the heating interval is 30 days, the V_g decreases from 3.931 × 10⁶ to 3.869 × 10⁶ ST m³. The result is consistent with the findings of Li et al., who experimentally studied the dissociation and gas production behavior of hydrates under two hot brine injection modes. Their results indicate that continuous thermal injection produces more gas than intermittent thermal injection [52]. Compared with the LWH&CH&SHIA (480 d) production, the V_g of LWH&IH&SHIA (480 d) just decreases by 1.58% during the 1080-day production cycle, and the cumulative heat input is only 50.00% of the LWH&CH&SHIA (480 d) production. Compared with the WWH&CH production, the V_g only decreases by 4.59%, but the cumulative heat input is only 4.76% of the WWH&CH production. After 1080 days of production, the R_gw in the above cases remains around 200, and there is no significant difference.

As shown in Figure 17, the EROI_custom curve of intermittent heating mode is higher than that of the continuous heating mode, which indicates that the intermittent heating mode has a better energy output-to-input ratio.

5. Conclusions

Based on the comprehensive analysis of the effects of various heating parameters, including heating power, heating location, heating time, and heating mode, on the vertical well depressurization production performance of the Class 1-type NGH reservoir in the Shenhu Sea area, the following results were obtained:

(1)

In situ electric heating has a good effect on increasing production capacity, and its main mechanism is to promote the dissociation of hydrates and eliminate secondary hydrates. When the vertical wellbore with a completion length of 70 m is deployed in the center of the model and continuously mined for 1080 days under a production pressure difference of 4 MPa, the V_g of the whole wellbore heating case is increased to 118.56% compared with the case without heating.

(2)

When intermittent heating is applied to the local wellbore (15 m) located in the TPL (with an interval of 30 days) and stop heating in advance at 480 days, there is no significant difference in the V_g after 1080 days of production compared to the whole wellbore heating, and the cumulative heat input is only 4.76% of the whole wellbore heating. We recommend considering intermittent heating of the local wellbore and stopping heating in advance during vertical well depressurization, which can greatly save heating energy consumption while increasing production capacity.

(3)

Intermittent heating of the local wellbore may be an effective strategy to maintain high production capacity while reducing energy consumption during vertical well depressurization. In the future, we will combine methods such as complex structured wells and reservoir reconstruction to further study the effects of intermittent heating of local wellbore on natural gas and water production in different types of hydrate reservoirs.