- Email: [email protected]
Key evaluation techniques in the process of gas reservoir being converted into underground gas storage
PETROLEUM EXPLORATION AND DEVELOPMENT Volume 44, Issue 5, October 2017 Online English edition of the Chinese language journal Cite this article as: PETROL. EXPLOR. DEVELOP., 2017, 44(5): 840–849.
RESEARCH PAPER
Key evaluation techniques in the process of gas reservoir being converted into underground gas storage ZHENG Dewen1, 2, XU Hongcheng1, 2,*, WANG Jieming1, 2, SUN Junchang1, 2, ZHAO Kai1, 2, LI Chun1, 2, SHI Lei1, 2, TANG Ligen1, 2 1. PetroChina Research Institute of Petroleum Exploration & Development, Langfang 065007, China; 2. Key Lab of Oil and Gas Underground Gas Storage Engineer of China National Petroleum Corporation, Langfang 065007, China
Abstract: Due to the significant differences in development modes and operation rules of underground gas storage (UGS) and gas reservoir, the design of UGS construction has its own particularity and complexity. Key evaluation techniques in the process of gas reservoir being converted into underground gas storage were proposed and field application was analyzed. The construction and operation experience of the first batch commercial UGS in China was summarized, the mechanisms of multi-cycle injection and production with large flux in short-term was examined and some concepts were proposed such as the dynamic sealing of traps, the effective pore volume of UGS and the high velocity unstable seepage flow with finite supply. Four key technologies of UGS, i.e., trap sealing evaluation, gas storage parameter design, well pattern optimization and monitoring programs design were created. Preservation condition, storage capacity, effective injection & production and safe operation technology problems of UGS were solved respectively. The geological program design technology system of UGS construction in a gas field was gradually enriched and improved. These technologies have successfully guided geological plan design and implementation of UGS construction in a gas reservoir, the effects of gas storage and peaking capacity of the ramp-up cycles were great, and the actual dynamic was very consistent with design indicators. Key words: underground gas storage in gas field; dynamic sealing; gas storage parameter; productivity evaluation; injection-production well pattern; monitor scheme
Introduction The underground gas storage in a gas reservoir is a kind of economic and effective gas storing and peak shaving facility, which plays an irreplaceable role in the safe and stable supply of gas. Main gas storage technologies are maturing abroad, but still in the infancy in China. At the end of the 20th century, six gas storages were built by transforming sandstone gas reservoirs in Dagang area successively. These gas storages have worked for 16 injection-withdrawal cycles, with a working gas volume of 18.6×108 m3, making major contribution to the winter gas peak shaving of Beijing area. This storage group is the first batch of commercial gas storages built in China with limited design and running experiences, current evaluation shows that they have lower working gas volume ratio than the designed value, inadequate monitoring and higher gas loss. Therefore, the design, evaluation, and operation of gas storages must be improved. Large scale construction of gas storages in China started in 2010, when six gas storages, Hutubi in Xinjiang, Xiangguosi in the Southwest Oilfield, and Shuang6 in Liaohe Oilfield etc, were built by
transforming gas reservoirs, marking the 2nd stage of gas storage construction in China[1]. Based on the construction and operation experience of the 1st batch of gas storages, design concept has been updated, and breakthroughs have been made in evaluation of gas storage, forming the basic gas storage construction system. Aimed at four bottlenecks in gas storage construction, i.e. sealing conditions, storage capacity, efficient injection and withdrawal, and safety monitoring, this study proposes four technologies: dynamic sealing evaluation of trap, effective storage capacity design, optimization of injection-withdrawal well pattern and monitoring scheme design technologies. These technologies have been used to guide the design of gas storage geologic plan and gas storage operation, and the results are analyzed.
1.
Dynamic evaluation of the trap sealing property
Different from development of a gas reservoir in which the static sealing of cap rock and fault is the major concern, during the operation of gas storage, the formation pressure goes up and down alternately, inducing periodic disturbance of
Received date: 24 Aug. 2016; Revised date: 15 Jun. 2017. * Corresponding author. E-mail: [email protected] Foundation item: Supported by the PetroChina Science and Technology Major Project (2015E-4002). Copyright © 2017, Research Institute of Petroleum Exploration and Development, PetroChina. Published by Elsevier BV. All rights reserved.
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
ground stress field, deformation and fatigue damage of cap rock, and activation of fault, so the trap may lose sealing capacity, and consequently gas may leak massively[25]. Therefore, the mechanism and main control factors and potential risks associated with dynamic sealing failure of cap rock and fault must be evaluated comprehensively by considering the stress disturbance to the cap rock and fault caused by high speed gas injection and withdrawal during gas storage operation with the time before gas storage construction as the node, to guide the pressure design and lower risks in gas storage operation. 1.1.
Mechanisms of dynamic trap sealing failure
According to the features of ground stress disturbance in gas storage operation, the root causes of trap sealing failure are microscopic pore structure change and macroscopic mechanical deformation and damage of rock under alternating stress. Therefore, four kinds of sealing failure mechanisms, cap rock capillary sealing failure, tensile damage, shear damage and shear slip activation of fault, have been pointed out. (1) Cap rock capillary sealing failure. Under alternating stress, microscopic pore structure may change and micro-fissures may expand in the cap rock, or the strong heterogeneous disturbance caused by high speed injection and withdrawal may lead to change of the original water driving system in the cap rock, causing cap rock capillary sealing failure and the exacerbation of initial slow diffusion of gas into leak. (2) Tensile damage. High speed gas injection may lead to high pressure in local area, when higher than the minimum horizontal stress, the pressure would cause tensile damage to the cap rock. (3) Shear damage. Ground stress disturbance in high speed injection and withdrawal may cause stress concentration in the abrupt structural change area, and consequently sliding deformation and shear damage in the abrupt structural change area or lithologic change weak plane of the cap rock. (4) Shear slip of fault. When the shear stress acting on the fault plane is higher than the critical value due to ground stress disturbance during high speed injection and withdrawal, fault running through the cap rock will slip, breaking the integrity of the cap rock and causing sealing failure. 1.2.
Evaluation for trap dynamic sealing property
The pressure limit a gas storage trap can withstand can be estimated by calculating the pressure withstanding limits of the cap rock and fault at the critical point of capillary sealing failure, tensile and shear damages, and shear slip of fault, then the upper pressure limit of the gas storage can be designed scientifically, and the room for upper pressure limit increase evaluated. 1.2.1.
Capillary sealing capacity of cap rock
The capillary sealing capacity evaluation experiment of gas storage cap rock follows the same idea with gas reservoir de-
velopment experiment, but simulates quite different working conditions. In the conventional method, on the basis of comprehensive geologic study and porosity, permeability and microscopic pore structure test of core, core samples are selected to test the critical breakthrough pressure by gas breakthrough pressure experiment. As ground stress disturbance during gas storage operation would cause deformation and micro-pore structure change of cap rock, on the basis of conventional static breakthrough pressure test, core fatigue damage experiment under alternating load needs to be conducted according to the actual ground stress and designed pressure range, to figure out the critical breakthrough pressure of cap rock core after cyclic injection and withdrawal and dynamic capillary pressure withstanding limit of cap rock. 1.2.2.
Tensile damage risk to cap rock
The high speed gas injection and withdrawal during gas storage operation can aggravate the effect of reservoir heterogeneity, especially during gas injection, the bottom hole pressure may exceed the designed upper pressure limit of the gas storage. Pressure in local parts may be higher than the minimum horizontal stress, causing tensile damage to cap rock. Gas storages transformed from gas reservoirs of shallow burial depth especially have much higher tensile damage risk than shear damage risk. Therefore, in the evaluation of cap rock tensile damage risk, the trap ground stress must be tested accurately, especially for depleted gas reservoirs. The ground stress of reservoir and cap rock can be tested by hydraulic fracturing or leak off test and AE Kaiser effect experiment to evaluate tensile damage risk. In order to take the effects of cap rock shape and rock mechanic heterogeneity into consideration, the dynamic 3D trap geo-mechanical model including reservoir and cap rock is built to predict the 3D stress field distribution in the cap rock under different working conditions with numerical simulation. The ratio of minimum horizontal stress under any formation pressure to original minimum horizontal stress of cap rock is defined as tensile damage safety index of the cap rock. When it is larger than zero, the cap rock is not damaged, when it is less than zero, the cap rock is damaged. This way, the tensile damage risk of cap rock under different gas storage operating pressure ranges can be evaluated quantitatively and visually by geomechanical numerical simulation to find out the pressure limit the cap rock can withstand. 1.2.3.
Shear damage risk of cap rock
Rock mechanical heterogeneity caused by triaxial major stress difference, local stress concentration due to high speed injection and withdrawal, lithologic changes and sedimentary bedding may lead to shear damage of cap rock along mechanical weak plane. For gas storages with large burial depth, and complex structure shape and lithology, the shear damage risk of cap rock is higher than tensile damage risk. Triaxial compression rock mechanical experiment and 3D
841
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
geo-mechanical modeling are used to evaluate the shear damage risk of cap rock quantitatively. According to ground stress of cap rock and operation pressure of the gas storage, samples were selected for uniaxial and triaxial compression test under different confining pressures. The shear damage envelop of cap rock was fitted based on Mohr-Coulomb strength rule (Fig. 1). Then according to the range of cap rock ground stress and disturbance range of the gas storage operation, the shear damage strength limit of cap rock was determined. Since the samples used in lab rock mechanical experiment are all small scale homogeneous cores, the effects of stress concentration at abrupt structural change area and heterogeneity of rock mechanics due to different lithologies can’t be reflected by the experiment. Therefore, it is necessary to quantitatively evaluate the dynamic sealing failure risk under different injection-withdrawal conditions by the shear damage safety factor of cap rock through 3D geo-mechanical numerical simulation on the basis of geologic study of trap and rock mechanical experiment of reservoir and cap rock. The calculation of the safety factor is also based on the classical Mohr-Coulomb strength rule, when the shear stress at a point is higher than the shear strength, shear damage occurs. The ratio of difference between the critical shear stress of cap rock at which shear damage occurs and the maximum shear stress at an injection-withdrawal condition to the critical shear stress is defined as the safety factor (Fig. 1). When it is greater than zero, no shear damage happens to the cap rock, when it equals zero, shear damage occurs to the cap rock. 1.2.4.
Risk of fault instability due to shear slip
Fault instability due to shear slip has similar principle with shear damage of cap rock, but the fault is the geologic crushed zone, the largest mechanical weak plane, so the cohesion is generally ignored. Geomechanical study shows[79] during the ground stress disturbance caused by gas storage injection-withdrawal, when the shear stress acting on the fault plane is greater than the product of friction coefficient and effective normal stress, the fault will slip, losing sealing capacity. The slip and instability of far field fault will cause substantial formation deformation, and affect wellbore integrity in turn. Therefore, the mechanical stability of fault will directly dictate its dynamic sealing property. Limited by experiment modeling equipment and prepara-
Fig. 1.
Mohr circle of cap rock shear damage.
tion of fault sample etc, currently the stability of fault is mainly evaluated by geo-mechanical numerical simulation. By building mechanical grid model of fault, the Coulomb damage function is taken as the main index to evaluate the stability of fault under different injection-withdrawal conditions, and find out the pressure limit a sealing trap can withstand through finite numerical simulation. When the Coulomb damage function is less than zero, equal to zero and greater than zero, the fault is stable, at critical state, and loses stability due to shear slip respectively.
2.
Design of gas storage capacity parameters
During low speed depletion development of a gas reservoir, a single well can control a large area and large reserves, and a high proportion of gas-bearing pores is involved in gas seepage. In contrast, a gas storage is often transformed from a gas reservoir in the late stage of development or even a depleted gas reservoir when the gas reservoir turns smaller in storage space and poorer in seepage condition, coupled with high speed repeated injection and withdrawalal under high pressure difference, the gas storage will see worsening of storage and seepage conditions, shrink of sing well control area and reduction of gas-bearing pore proportion involved in seepage. Therefore, the effective capacity of a gas storage isn’t simply equal to the dynamic geologic reserves of the gas reservoir. Fine geologic study and injection-withdrawal seepage experiment are needed to find out the mechanism and main control factors of the micro-seepage in the gas storage, and quantify the effective pore volume, so the gas storage capacity prediction model can be established to guide the design of storage capacity. 2.1.
Analysis of gas-bearing pore space
The life expectancy of gas reservoirs developed at low speed with depletion drive is 20-30 years in general, the gas reservoir features large gas drainage radius, infinite fluid supply range, and high geologic reserve producing degree. In contrast, in the operation of gas storage, gas is injected and recovered at high speed in cycle of 1 year, fluid far from the well can’t flow to the well bottom in time, so the well control radius drops significantly. In addition, to meet the requirement of winter peak shaving, the single well proration can be as high as 5-10 times the rate of gas reservoir development, so after the well in a gas storage is opened, the gas flow reaches high speed pseudosteady state flow soon, the effect of formation heterogeneity increases and the single well control range reduces further. Clearly, compared with the gas reservoir development period, the single well in gas storage has much lower control degree on sandbody, which has been proved by the Banqiao gas storage group. Taking Ban 876 as an example, the angle between the gas storage capacity curve and vertical axis is much smaller than that between the gas reservoir pressure drop curve and vertical axis (Fig. 2), that means the gas recovered from the gas storage is less than gas
842
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
Fig. 2. Comparison of gas storage capacity curves with the pressure drop curve during gas reservoir development.
produced from the gas reservoir under a unit formation pressure drop. Obviously, the dynamic geologic reserves of gas reservoir can’t be fully produced during the short high speed injection-withdrawal cycle of gas storage[1011]. Besides, invasion of edge or bottom water, retrograde gas condesate loss, plastic deformation of reservoir caused by stress sensitivity, and gas isolated by water-gas interlocking all can make the gas-bearing pore volume reduce further in the process of gas reservoir development and gas storage operation. 2.2. Evaluation method of effective pore volume after gas storage construction Based on results of geologic, dynamic and seepage mechanism researches, and the distribution features and variation trend of fluid in vertical direction during gas reservoir development and gas storage operation, the gas storage operation profile is simplified into four zones and four contacts (Fig. 3). For a gas storage transformed from a depleted water-invaded gas reservoir, the gas-water contact in the gas storage has moved upward from the original GWC0 of the gas reservoir to GWC1 at the beginning of gas storage construction. When the gas storage operates between the upper and lower pressure limits, a relative stable transition zone will form due to the injection-withdrawal breathing effect, with its top corresponding to the gas-water contact at lower pressure limit GWC2, and its bottom corresponding to the gas-water contact at upper pressure limit GWC3. As during the gas reservoir development and gas storage operation, no formation water invaded, the zone above GWC1 is a pure gas zone. Since gas injection pushes water downward, after multiple rounds of gas injection, the zone between GWC1 and GWC2 becomes a pure gas zone. These two zones are the major gas storage zones, where a high proportion of gasbearing pores involves in the gas seepage. During the operation of gas storage, the gas-water transition zone experiences water in and out and alternating displacement, with lower operation efficiency, but pore space in this zone involved in seepage is not negligible. In comparison, the watered-out zone is occupied by formation water during the whole process of gas reservoir development and gas storage
Fig. 3.
Evaluation of effective pore space in gas storage by zone.
operation, unable to be used to form working gas. In the gas-water seepage zone of the gas storage, the seepage mechanism is complex, and controlled by reservoir physical properties and heterogeneity, wettability, and capillary pressure etc microscopically. For highly heterogeneous reservoirs, water channeling or selective water invasion during high speed gas production blocks the gas in small pores and micro-fissures in the form of slipstream. Due to the additional resistance produced by the Jamin effect, the uncontinuous flow gas will be stuck, forming confined gas, and there is some confined gas in unconnected pores or their blind end [12-15]. Moreover, during gas storage operation, gas displacing water still mainly work in large pores and throats, as water is tiny pores and throats is hard to displace. Also bottom and edge water invasion and retrograde condensation of condensate gas will take up some pore space, and plastic deformation caused by reservoir stress sensitivity can lead to drop of reservoir porosity and permeability. The gas-bearing pore volume calculated by the gas reservoir development dynamic method minus the immobile pore space of each zone is the effective pore volume of each zone, and the total effective pore volume of the gas storage can be expressed as: Veff V0 V1 V2 V3 V4 (1) 2.3.
Prediction model of effective gas storage volume
After the gas storage is built, the gas stored in the storage includes gas left before gas storage construction and injected gas, and the effective gas storage volume is the sum of effetive gas storage volume before the storage construction and cumulative injected gas volume: Gr Gr0 Ginj (2) From the underground pore volume balance principle, as the formation pressure increases after gas storage construction, the effective gas storage volume after storage construction
843
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
corresponds to effective pore volume under the pressure at this point. Thus, the effective pore volume of the gas storage under the formation pressure can be calculated by equation (2). For gas storage transformed from water invaded reservoir, on the basis of effective pore volume of the gas storage, the effective pore volume of the gas storage under different pressures can be worked out by considering the irreducible water volume and rock elastic deformation during gas storage construction and operation. Therefore the effective pore volumes at upper and lower pressure limits can be calculated. Then the storage capacity parameters (effective storage volume, cushion gas volume, working gas volume, and additional cushion gas volume) can be worked out by the equations below:
Gmax Gmin
Veff ,max Bg,max Veff ,min Bg,min
(3) (4)
Gwg Gmax Gmin
(5)
Gadd Gmin Gr0
(6)
3. Design of injection-withdrawal well pattern of a gas storage The layout of injection-withdrawal well pattern of a gas storage should consider the high speed injection and withdrawal feature of the gas storage, and meet the requirement of controlling the gas storage volume effectively and uneven peak shaving gas production. Therefore, based on evaluation of single well productivity, well control and analysis of market gas demand, the minimum well number meeting the gas storage parameter design will be determined, proper well pattern density will be proposed, and the layout of injectionwithdrawal well pattern should consider the structure position, reservoir features and fluid distribution etc comprehensively. 3.1. Main factors affecting injectivity and productivity of wells in a gas storage The gas injectivity and productivity of wells in a gas storage under high speed unsteady seepage flow are affected by near wellbore seepage condition and well controlled storage volume largely. The seepage conditions near wellbore which directly determine the instantaneous injectivity and productivity of the gas well, are affected by the reservoir effective permeability, well bottom perfection degree, reservoir producing degree and drainage area etc primarily. Therefore, removing pollution and connecting reservoir through reservoir micro-stimulation can enhance reservoir permeability, meanwhile, with the injection and withdrawal going on, the formation seepage conditions will become better. Big tubing diameter should be adopted as far as possible to increase wellbore drainage area, and the combination of vertical well, horizontal well and highly deviated well should be used to increase perforated thickness
and area of reservoir, and single well productivity. In addition, sanding out of reservoir, liquid loading in wellbore, tubing string corrosion, and wellhead minimum pressure etc all may restrict the play of the formation seepage capacity, so optimization at multiple nodes of formation, wellbore and wellhead should be carried out to get the reasonable production under different tubing diameters, well types, formation pressures and wellhead pressure. The control radius of a gas storage well under high speed unstable percolation directly determine the effective fluid supply flow. At present, the injection-withdrawal cycle of gas storages in China is around half a year, less than one fiftyth of the gas reservoir development span, so there is no time for fluid far from the well to flow to the well bottom. Apparently, in the limited well control scope, the culmulative gas a well can absorb and produce in a cycle are limited, restricting gas injectivity and productivity of the well. 3.2. Optimization of production of wells in a gas storage at multiple nodes Different from nodal analysis for gas reservoir development, gas storage operation focuses more on multiple node optimization of formation, wellbore and wellhead, to obtain a reasonable daily gas injection/production rate under high speed unstable seepage flow. In terms of fluid flow in reservoirs, due to repeated injection and withdrawal, the reservoirs may decrease in effective permeability due to stress sensitivity or sanding. For the pure gas zone formed by gas displacing water, after multiple cycles of gas injection, the reservoir seepage conditions would be improved, and gas productivity enhanced. For the gas-water transition zone, with the rise of water saturation after multiple rounds of operation, the gas relative permeability would drop and tend to be stable in late stage. As the reservoir seepage situation of wells changes constantly during multiple injection-withdrawal process, the productivity of well changes accordingly[16]. Therefore, the productivity prediction equation for gas wells in storage is established by correcting key parameters such as effective permeability and stress sensitivity of reservir and natural gas property etc. For vertical pipe flow, besides consideration of critical liquid carrying speed and wellbore erosion, to meet the requirement of large volume huff and puff, large size tubing and horizontal well and highly deviated well should be adopted as far as possible. With respect to wellhead exporting conditions, the major aim is to meet the lowest pressure needed for gas to be sent into the long exporting pipeline, and the boosted export should be avoided as far as possible. If economically feasible, the gas export pressure can be lowered properly to enhance the peak shaving capacity of the gas storage. Finally, the productivity equation is modified under the condition that the gas storage reaches stable injection-withdrawal state. By combining the equation with vertical pipe
844
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
flow equation, and taking the critical sanding pressure difference, critical liquid carrying volume, erosion flow and wellhead minimum export pressure as constraint conditions, comprehensive nodal pressure analysis is conducted for different tubing diameters, well types at different formation pressures and wellhead pressures to get the reasonable production of the gas storage at different pressures.
Considering reservoir effective thickness, gas saturation and porosity etc, equation (7) can be simplified as:
3.3. Design of well pattern density for a gas storage under the high speed unstable seepage condition
By using well production from multi-node optimization, the number of injection and withdrawal wells of the gas storage can be calculated according to days of gas production, single well reasonable production rate at the end of gas recovery, and designed working gas volume:
The essence of reasonable well pattern design for a gas storage is to figure out the minimum well number satisfying the designed indexes of storage. The experience of gas storage operation shows wells have higher gas injection capacity than gas production capacity, and longer gas injection time than gas production time. Therefore, gas production capacity is the key dictating the number of wells needed. It is suggested to determine the reasonable number of wells by considering the well control radius under unstable seepage flow, single well coordinated production and monthly gas demand for peak shaving. 3.3.1. Statistics on finite well control radius under high speed unstable seepage flow As one of the first batch commercial gas storages built in China, the Banqiao storage group has been operated for 16 cycles, accumulating a lot of operation dynamic data. By diagnosing and evaluating the gas production performance of 71 wells over multiple cycles with the rate transient analysis software RTA, the relationship chart between well control radius and reservoir effective permeability has been compiled (Fig. 4), which shows the two have good correlation. If the effective permeability of a reservoir is given, the well control radius can be obtained by analogue. The minimum number of wells needed to control the gas storage capacity can be obtained by diving the gas storage capacity by average storage capacity per well:
n1
Gmax Gwk
(7)
Fig. 4. Relationship between single well control radius and reservoir effective permeability of the Banqiao storage group.
n1
S πRe2
(8)
3.3.2. Formation-wellbore-wellhead multi-node coordinated production method
n2 104 3.3.3.
Gwg
(9)
tqg
Estimation of monthly gas demand
The daily gas production in each month can be calculated by using the working gas volume of the gas storage and the monthly uneven demand coefficent. The mathematical model is built by coupling periodical cumulative gas production, formation pressure and the reasonable gas production under this pressure. The maximum number of injection and withdrawal wells are those at the peak of gas production and the end of gas recovery:
n3 max 104 Gp
10
3.3.4. wells
4
Gp
m m / tq m m / tq j
j
end
max
g, max
,
g,end
(10)
Reasonable number of injection and withdrawal
Since the gas storage operation must satisfy well control material balance, single well productivity and uneven gas demand at the same time, the number of injection and withdrawal wells needed for the storage take the largest of the above three results: n max( n1 , n2 , n3 ) (11)
4.
Design of gas storage monitoring scheme
A scientific and reasonable monitoring scheme is the guarantee for safe and smooth running of a gas storage. Dynamic monitoring of a gas storage over the whole life span is a high concern abroad, and a complete monitoring system centering on well engineering, trap sealing and gas storage performance has been established (Fig. 5). The monitoring methods and means include conventional temperature and pressure test, special logging, micro-seismic, productivity test and interferometry synthetic aperture radar (InSAR) system. Enough monitoring wells deployed on the boundary and inside of the gas storage, plus space-earth observation system and GPS constitute an integrated space-earth monitoring network. The monitoring wells should be 0.5-1 times even 1.5 times of injection and withdrawal wells, and mostly newly drilled, with a small number of old wells. With a whole process monitoring installed and running, we can get first-
845
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
of inner diameter, perforation quality and string structure etc are checked regularly, and wells and positions of poor cementing quality are monitored especially. The well structure quality of old and new wells are generally monitored by cementing acoustic logging, acoustic variable density log and electromagnetic defect logging etc, and only wells meeting the integrity standard can be used for the gas storage. 4.2.2.
Fig. 5. Monitoring well net of a gas storage transformed from a gas reservoir.
hand data on the safety and operation performance of the gas storage. 4.1.
Monitoring system
The monitoring system consists of three aspects, well engineering, trap sealing and performance of the gas storage, and can monitor the whole process from construction to operation of the gas storage. (1) Well engineering monitoring. Quality test of old well structure before gas storage construction, monitoring and formation testing of newly drilled and completed well, and downhole technical condition monitoring are conducted to ensure integrity of gas wells in the gas storage. (2) Monitoring trap sealing property. The cap rock, fault system, spill point, surrounding reservoir and overlying permeable layer and shallow aquifer in the gas-bearing area are monitored for gas leak, to ensure the gas injected into the gas storage is held securely. (3) Operation performance monitoring of the inner gas storage. Production performance of injection and withdrawal wells, gas reservoir temperature and fluid property, gas-water contact and fluid migration, and productivity of injection and withdrawal wells etc are monitored to guide gas storage expansion and production, proration and working system adjustment. 4.2. 4.2.1.
Monitoring modes Well engineering monitoring
Before the construction of a gas storage, pressure test can be used to find out the casing corrosion degree and cementing quality of old wells. The drilling of new wells should be monitored according to geologic design, and the formation testing of new wells should follow relevant codes of the petroleum industry. The downhole technical conditions during gas storage operation are monitored by building a casing monitoring system, the casing, damage and corrosion of collar, variation
Monitoring of trap sealing property
The sites with thinner cap rock, higher lateral and vertical sealing risks on both sides of control faults of the gas storage area are monitored by old wells or newly drilled wells, from which, pressure is observed and samples are taken for analysis regularly. The sites of cap rock with possible lithologic change, thinning and more fractures are monitored by old wells or newly drilled wells, from which, pressure is observed and samples are taken for analysis regularly. So are the perimeter of the gas storage, spill point of trap and shallow aquifer etc. In addition, geochemical monitoring and radon-thoron micro-element test are used abroad to identify gas leak directly. Microseismic technique is employed to interpret the microseismic events caused by alternating stress during gas injection and withdrawal, to tell the strength of stress-strain, and judge the risk of trap sealing failure. The newly emerging space-earth observation technique, for example the InSAR and GPS combined system can monitor the ground subsidence around the gas storage area. 4.2.3.
Monitoring of inner performance
(1) Monitoring of injection and withdrawal well performance. The items need to be monitored in gas production well include choke, oil, gas and water production, tubing and casing pressure and wellhead temperature etc. Things needed to monitored in gas injection wells include gas injection rate, pressure and temperature of compressor, and casing and tubing pressure and wellhead temperature etc. (2) Monitoring of temperature, pressure and fluid property in gas reservoirs. Old or new wells are used to monitor static temperature and geothermal gradient, static pressure and pressure gradient of the reservoirs, and fluid samples are taken from the wells for analysis to get information on the variation of fluid properties in the reservoirs over mulitple cycles of injection and withdrawal. (3) Monitoring of gas-water contact and fluid migration. Old wells or new wells are deployed at the structural mediumhigh positions, transition zone and perimeter of the storage to monitor fluid migration direction and variation of gas-water contact. Gas-water contact device and seismic tomography can be used to detect the gas-water contact; induced polarization electric logging, high precision gravity test, pulse spectrum gamma logging etc can be employed to detect gas-water contact and fluid migration direction; and 4D seismic and
846
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
tracer can be applied to monitor fluid migration direction and variation trend of gas flooding front. (4) Productivity monitoring. Systematic well test and transient well test of different zones are conducted in representative wells at different gas injection (withdrawal) periods. In 2-3 cycles, all wells will be tested in turn, to get well productivity, effective permeability, formation pressure, well integrity coefficient, formation connectivity and formation boundary property etc, to figure out the variation of gas injectivity and productivity of wells over multiple cycles. For commingled injection or production, the gas injection profile logging is needed to know the gas absorption of single layer, and gas production profile logging is needed to find out the gas recovery and produced fluid components of individual layers.
5. 5.1.
Case study Main geologic characteristics of the gas storage
The gas storage, one of six storages started to build in 2010, is large in burial depth (3 580 m), large in reserve scale (126×108 m3), medium in reservoir physical property (with a porosity of 19.4% and permeability of 48.6×103 μm2), and strongly heterogeneous. The storage is transformed from a deep sandstone lean condensate gas reservoir with bottom water controlled by lithologic and structure jointly. There are 3 major faults with big fault throw (200 m) and long extension (20 km) in the gas-bearing area. Affected by underwater distributary channel environment, the reservoir is medium in physical property, highly variable in sand distribution, strong in heterogeneity, which would impact producing of reserve negatively. In addition, with the development of the gas reservoir, edge water invaded into the reservoir, forming water and gas seepage zone with complex seepage mechanism, which will lower producing degree further. 5.2. Key parameters in gas storage construction evaluation In consideration of the large faults in the gas reservoir trap and the sealing property of the reservoir during its own formation process, the original formation pressure 34 MPa is taken as the upper limit pressure of the gas storage. By taking reservoir physical property and heterogeneity, edge water invasion, retrograde condensation loss and the effect of high speed unstable seepage flow on pore space into account, the immobile pore volume is estimated at 432×104 m3, accounting for 10% of the original gas-bearing pore volume. Thus, the gas storage is estimated to have an effective pore volume of 4 018×104 m3, a storage capacity of 107×108 m3, working gas volume of 45×108 m3, cushion gas volume of 62×108 m3, and additional cushion gas volume of 16.5×108 m3. Referring to the relationship chart between well control radius and reservoir effective permeability (Fig. 4), the average well control radius from analogue is 400 m. In
consideration of the injection and withdrawal capacity of well and peak shaving demand, the number of injection and withdrawal wells are decided at 30, including 29 vertical wells and one horizontal well. The gas storage has an average daily gas injection rate of 1 550×104 m3 and average daily gas production rate of 2 800×104 m3. At last, four conventional monitoring wells are designed to survey the sealing of cap rock and fault, gas-water contact and fluid migration, temperature and fluid property etc constantly. Moreover, micro-seismic monitoring is deployed in six shallow wells and three medium depth wells at the outer side of the boundary fault and inside the gas storage to keep watching the sealing of the trap. 5.3.
Performance of the storage
The gas storage has operated for four cycles so far, the storage indexes obtained from dynamic data are in good agreement with the indexes in the design plan (Fig. 6). The gas storage has been running well on the whole, with higher storage capacity recovery level and fast production rate. Fig. 7 shows the storage capacity at the end of the fourth cycle is 96×108 m3, 90% of the original storage capacity; and its peak shaving capacity has increased by four times from the
Fig. 6. Reviewed storage capacity curve of the storage after multiple cycles of operation.
847
Fig. 7.
Predicted indexes of the storage over the years.
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
initial 2.7×108 m3 to 14.0×108 m3. It is expected to reach the designed storage capacity after two years (Fig. 7), basically consistent with the capacity reaching cycle of six years in the design plan, and the capacity reaching cycle of 5-8 years of storages abroad, which proves the geologic design plan of the storage is scientific and reliable.
6.
n—reasonable number of injection and withdrawal wells; n1, n2, n3—numbers of injection and withdrawal wells obtained from limited well control radius method, formation-wellbore-wellhead multi-node coordinated production method, and monthly uneven gas demand estimation method; qg—reasonable production of single well, 104 m3/d; qg,end—reasonable gas well production under the formation pres-
Conclusions
sure at the end month of the peak shaving gas production, 104 m3/d;
In light of the differences between gas storage operation and gas reservoir development, four key technologies for gas storage design, evaluation of trap dynamic sealing, storage parameter design, optimization of injection and withdrawal well pattern, and monitoring plan design have been established through the study, to solve preservation conditions, capacity, high efficient injection and withdrawal, and safe operation of the gas storage, which enriches and improves the storage construction evaluation system. Through reviewing the construction and operation experience of the Banqiao storage group, we have updated the gas storage design concept, and advanced the ideas of effective storage capacity, limited well control under high speed unstable seepage flow, and variation of well productivity in the storage over operation cycle, laying theoretical foundation for storage parameter design and injection and withdrawal well pattern deployment. An application case shows that the gas storage operation indexes are highly consistent with the designed indexes, proving the key technologies for gas storage geologic plan design are scientific and reliable. These technologies can provide theoretical basis and technical support for similar gas storage construction in China.
qg,max—reasonable gas well production under the formation pressure at the highest gas production month of the peak shaving, 104 m3/d; Re—radius controlled by one well, m; S—effective gas-bearing area of the reservoir, m2; t —days of gas production, d; V0 —underground pore volume corresponding to the reservoir dynamic geologic reserves, m3; V1—immobile pore volume in the pure gas zone before gas storage construction, m3; V2—immobile pore volume in the pure gas zone formed by gas displacing water, m3; V3—immobile pore volume in gas-water transit zone, m3; V4—immobile pore volume in watered-out zone, m3; Veff —effective pore volume of gas storage, m3; Veff,max—effective pore volume corresponding to the upper pressure limit, 108 m3; Veff,min—effective pore volume corresponding to the lower pressure limit, 108 m3; σ—effective normal stress, MPa; σ1, σ3—maximum and minimum effective normal stress, MPa; τ—shear stress, MPa; τm—maximum shear stress under an injection-withdrawal condition, MPa;
Nomenclature
m —critical shear stress at which shear damage occurs, MPa.
Bg,max—gas volume factor corresponding to the pressure upper
References
limit; Bg,min—gas volume factor corresponding to the pressure lower lim-
[1]
it;
DING Guosheng, LI Chun, WANG Jieming, et al. The status quo and technical development direction of underground gas
Gadd—additional cushion gas volume, 108 m3;
storage in China. Natural Gas Industry, 2015, 35(11): 107–112.
Ginj—cumulative gas injection volume, 108 m3;
[2]
Gmax—storage capacity, 108 m3;
BRUNO M S, KANG L, DIESSL J, et al. Development of improved caprock integrity analysis and risk assessment tech-
Gmin—cushion gas volume, 108 m3;
niques. Energy Procedia, 2014, 63: 4708–4744.
Gp—peak shaving gas demand, 108 m3;
[3]
ORLIC B, WASSING B B T, GEEL C R. Field scale geome-
Gr—storage effective capacity, 108 m3;
chanical modeling for prediction of fault stability during un-
Gr0—effective storage capacity before storage construction, 108
derground gas storage operations in a depleted gas field in the
m3;
Netherlands. ARMA 13-300, 2013.
Gwg—working gas volume, 108 m3;
[4]
TENTHOREY E, VIDAL-GILBERT S, BACKÉ G, et al.
Gwk—average storage capacity controlled by one well, 108 m3;
Modelling the geomechanics of gas storage: A case study from
K—reservoir effective permeability, 103 μm2;
the Iona gas field, Australia. International Journal of Green-
mend—uneven factor of the end month of peak shaving gas production;
house Gas Control, 2013, 13: 138–148. [5]
ANSAR S, HAIGH R, KHOSRAVI N, et al. Caprock integrity
mj —uneven factor of the jth month;
case study for non-thermal polymer flooding project using 4D
mmax—uneven factor of month with high peak shaving gas produc-
reservoir coupled geomechanical simulation. SPE 157912,
tion;
2012.
848
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
[6]
[7]
BILDSTEIN O, JULLIEN M, CRÉDOZ A, et al. Integrated
[11] XU Hongcheng, WANG Jieming, LI Chun, et al. Inventory
modeling and experimental approach for caprock integrity,
verification of underground gas storage based on a flooded
risk analysis, and long term safety assessment. Energy Proce-
and depleted gas reservoir. Natural Gas Industry, 2010, 30(8):
dia, 2009, 1(1): 3237–3244.
79–82.
TEATINI P, CASTELLETTO N, GAMBOLATI G. 3D ge-
[12] CIESLINSKI J T, MOSDORF R. Gas bubble dynamics ex-
omechanical modeling for CO2 geological storage in faulted
periment and fractal analysis. International Journal of Heat
formations: A case study in an offshore northern Adriatic reservoir, Italy. International Journal of Greenhouse Gas Control, [8]
2014, 22(3): 63–76.
forecast in underground gas storage based on modified mate-
VIDAL-GILBERT S, TENTHOREY E, DEWHURST D, et al.
rial balance equation. Petroleum Exploration and Develop-
Geomechanical analysis of the Naylor Field, Otway Basin,
[9]
and Mass Transfer, 2005, 48(9): 1808–1818. [13] TANG Ligen, WANG Jieming, BAI Fengjuan, et al. Inventory
ment, 2014, 41(4): 480–484.
Australia: Implications for CO2 injection and storage. Interna-
[14] SHI Lei, LIAO Guangzhi, XIONG Wei, et al. Gas water per-
tional Journal of Greenhouse Gas Control, 2010, 4(5):
colation mechanism in underground storage built on water
827–839.
drive sandstone gas reservoir. Natural Gas Industry, 2012,
TEATINI P, CASTELLETTO N, FERRONATO M, et al.
32(9): 85–87.
Geomechanical response to seasonal gas storage in depleted
[15] DING Yunhong, ZHANG Qian, ZHENG Dewen, et al. Seep-
reservoirs: A case study in the Po River basin, Italy. Journal of
age laws in converting a micro fissure-pore carbonate gas reservoir into UGS. Natural Gas Industry, 2015, 35(1): 109–114.
Geophysical Research, 2011, 116(F2): 1–21. [10] XU Hongcheng, WANG Jieming, QU Ping, et al. A prediction
[16] TANG Ligen, WANG Jieming, DING Guosheng, et al.
model of storage capacity parameters of geologically-com-
Downhole inflow-performance forecast for underground gas
plicated reservoir-type underground gas storage. Natural Gas
storage based on gas reservoir development data. Petroleum
Industry, 2015, 35(1): 103–108.
Exploration and Development, 2016, 43(1): 127–130.
849
RESEARCH PAPER
Key evaluation techniques in the process of gas reservoir being converted into underground gas storage ZHENG Dewen1, 2, XU Hongcheng1, 2,*, WANG Jieming1, 2, SUN Junchang1, 2, ZHAO Kai1, 2, LI Chun1, 2, SHI Lei1, 2, TANG Ligen1, 2 1. PetroChina Research Institute of Petroleum Exploration & Development, Langfang 065007, China; 2. Key Lab of Oil and Gas Underground Gas Storage Engineer of China National Petroleum Corporation, Langfang 065007, China
Abstract: Due to the significant differences in development modes and operation rules of underground gas storage (UGS) and gas reservoir, the design of UGS construction has its own particularity and complexity. Key evaluation techniques in the process of gas reservoir being converted into underground gas storage were proposed and field application was analyzed. The construction and operation experience of the first batch commercial UGS in China was summarized, the mechanisms of multi-cycle injection and production with large flux in short-term was examined and some concepts were proposed such as the dynamic sealing of traps, the effective pore volume of UGS and the high velocity unstable seepage flow with finite supply. Four key technologies of UGS, i.e., trap sealing evaluation, gas storage parameter design, well pattern optimization and monitoring programs design were created. Preservation condition, storage capacity, effective injection & production and safe operation technology problems of UGS were solved respectively. The geological program design technology system of UGS construction in a gas field was gradually enriched and improved. These technologies have successfully guided geological plan design and implementation of UGS construction in a gas reservoir, the effects of gas storage and peaking capacity of the ramp-up cycles were great, and the actual dynamic was very consistent with design indicators. Key words: underground gas storage in gas field; dynamic sealing; gas storage parameter; productivity evaluation; injection-production well pattern; monitor scheme
Introduction The underground gas storage in a gas reservoir is a kind of economic and effective gas storing and peak shaving facility, which plays an irreplaceable role in the safe and stable supply of gas. Main gas storage technologies are maturing abroad, but still in the infancy in China. At the end of the 20th century, six gas storages were built by transforming sandstone gas reservoirs in Dagang area successively. These gas storages have worked for 16 injection-withdrawal cycles, with a working gas volume of 18.6×108 m3, making major contribution to the winter gas peak shaving of Beijing area. This storage group is the first batch of commercial gas storages built in China with limited design and running experiences, current evaluation shows that they have lower working gas volume ratio than the designed value, inadequate monitoring and higher gas loss. Therefore, the design, evaluation, and operation of gas storages must be improved. Large scale construction of gas storages in China started in 2010, when six gas storages, Hutubi in Xinjiang, Xiangguosi in the Southwest Oilfield, and Shuang6 in Liaohe Oilfield etc, were built by
transforming gas reservoirs, marking the 2nd stage of gas storage construction in China[1]. Based on the construction and operation experience of the 1st batch of gas storages, design concept has been updated, and breakthroughs have been made in evaluation of gas storage, forming the basic gas storage construction system. Aimed at four bottlenecks in gas storage construction, i.e. sealing conditions, storage capacity, efficient injection and withdrawal, and safety monitoring, this study proposes four technologies: dynamic sealing evaluation of trap, effective storage capacity design, optimization of injection-withdrawal well pattern and monitoring scheme design technologies. These technologies have been used to guide the design of gas storage geologic plan and gas storage operation, and the results are analyzed.
1.
Dynamic evaluation of the trap sealing property
Different from development of a gas reservoir in which the static sealing of cap rock and fault is the major concern, during the operation of gas storage, the formation pressure goes up and down alternately, inducing periodic disturbance of
Received date: 24 Aug. 2016; Revised date: 15 Jun. 2017. * Corresponding author. E-mail: [email protected] Foundation item: Supported by the PetroChina Science and Technology Major Project (2015E-4002). Copyright © 2017, Research Institute of Petroleum Exploration and Development, PetroChina. Published by Elsevier BV. All rights reserved.
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
ground stress field, deformation and fatigue damage of cap rock, and activation of fault, so the trap may lose sealing capacity, and consequently gas may leak massively[25]. Therefore, the mechanism and main control factors and potential risks associated with dynamic sealing failure of cap rock and fault must be evaluated comprehensively by considering the stress disturbance to the cap rock and fault caused by high speed gas injection and withdrawal during gas storage operation with the time before gas storage construction as the node, to guide the pressure design and lower risks in gas storage operation. 1.1.
Mechanisms of dynamic trap sealing failure
According to the features of ground stress disturbance in gas storage operation, the root causes of trap sealing failure are microscopic pore structure change and macroscopic mechanical deformation and damage of rock under alternating stress. Therefore, four kinds of sealing failure mechanisms, cap rock capillary sealing failure, tensile damage, shear damage and shear slip activation of fault, have been pointed out. (1) Cap rock capillary sealing failure. Under alternating stress, microscopic pore structure may change and micro-fissures may expand in the cap rock, or the strong heterogeneous disturbance caused by high speed injection and withdrawal may lead to change of the original water driving system in the cap rock, causing cap rock capillary sealing failure and the exacerbation of initial slow diffusion of gas into leak. (2) Tensile damage. High speed gas injection may lead to high pressure in local area, when higher than the minimum horizontal stress, the pressure would cause tensile damage to the cap rock. (3) Shear damage. Ground stress disturbance in high speed injection and withdrawal may cause stress concentration in the abrupt structural change area, and consequently sliding deformation and shear damage in the abrupt structural change area or lithologic change weak plane of the cap rock. (4) Shear slip of fault. When the shear stress acting on the fault plane is higher than the critical value due to ground stress disturbance during high speed injection and withdrawal, fault running through the cap rock will slip, breaking the integrity of the cap rock and causing sealing failure. 1.2.
Evaluation for trap dynamic sealing property
The pressure limit a gas storage trap can withstand can be estimated by calculating the pressure withstanding limits of the cap rock and fault at the critical point of capillary sealing failure, tensile and shear damages, and shear slip of fault, then the upper pressure limit of the gas storage can be designed scientifically, and the room for upper pressure limit increase evaluated. 1.2.1.
Capillary sealing capacity of cap rock
The capillary sealing capacity evaluation experiment of gas storage cap rock follows the same idea with gas reservoir de-
velopment experiment, but simulates quite different working conditions. In the conventional method, on the basis of comprehensive geologic study and porosity, permeability and microscopic pore structure test of core, core samples are selected to test the critical breakthrough pressure by gas breakthrough pressure experiment. As ground stress disturbance during gas storage operation would cause deformation and micro-pore structure change of cap rock, on the basis of conventional static breakthrough pressure test, core fatigue damage experiment under alternating load needs to be conducted according to the actual ground stress and designed pressure range, to figure out the critical breakthrough pressure of cap rock core after cyclic injection and withdrawal and dynamic capillary pressure withstanding limit of cap rock. 1.2.2.
Tensile damage risk to cap rock
The high speed gas injection and withdrawal during gas storage operation can aggravate the effect of reservoir heterogeneity, especially during gas injection, the bottom hole pressure may exceed the designed upper pressure limit of the gas storage. Pressure in local parts may be higher than the minimum horizontal stress, causing tensile damage to cap rock. Gas storages transformed from gas reservoirs of shallow burial depth especially have much higher tensile damage risk than shear damage risk. Therefore, in the evaluation of cap rock tensile damage risk, the trap ground stress must be tested accurately, especially for depleted gas reservoirs. The ground stress of reservoir and cap rock can be tested by hydraulic fracturing or leak off test and AE Kaiser effect experiment to evaluate tensile damage risk. In order to take the effects of cap rock shape and rock mechanic heterogeneity into consideration, the dynamic 3D trap geo-mechanical model including reservoir and cap rock is built to predict the 3D stress field distribution in the cap rock under different working conditions with numerical simulation. The ratio of minimum horizontal stress under any formation pressure to original minimum horizontal stress of cap rock is defined as tensile damage safety index of the cap rock. When it is larger than zero, the cap rock is not damaged, when it is less than zero, the cap rock is damaged. This way, the tensile damage risk of cap rock under different gas storage operating pressure ranges can be evaluated quantitatively and visually by geomechanical numerical simulation to find out the pressure limit the cap rock can withstand. 1.2.3.
Shear damage risk of cap rock
Rock mechanical heterogeneity caused by triaxial major stress difference, local stress concentration due to high speed injection and withdrawal, lithologic changes and sedimentary bedding may lead to shear damage of cap rock along mechanical weak plane. For gas storages with large burial depth, and complex structure shape and lithology, the shear damage risk of cap rock is higher than tensile damage risk. Triaxial compression rock mechanical experiment and 3D
841
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
geo-mechanical modeling are used to evaluate the shear damage risk of cap rock quantitatively. According to ground stress of cap rock and operation pressure of the gas storage, samples were selected for uniaxial and triaxial compression test under different confining pressures. The shear damage envelop of cap rock was fitted based on Mohr-Coulomb strength rule (Fig. 1). Then according to the range of cap rock ground stress and disturbance range of the gas storage operation, the shear damage strength limit of cap rock was determined. Since the samples used in lab rock mechanical experiment are all small scale homogeneous cores, the effects of stress concentration at abrupt structural change area and heterogeneity of rock mechanics due to different lithologies can’t be reflected by the experiment. Therefore, it is necessary to quantitatively evaluate the dynamic sealing failure risk under different injection-withdrawal conditions by the shear damage safety factor of cap rock through 3D geo-mechanical numerical simulation on the basis of geologic study of trap and rock mechanical experiment of reservoir and cap rock. The calculation of the safety factor is also based on the classical Mohr-Coulomb strength rule, when the shear stress at a point is higher than the shear strength, shear damage occurs. The ratio of difference between the critical shear stress of cap rock at which shear damage occurs and the maximum shear stress at an injection-withdrawal condition to the critical shear stress is defined as the safety factor (Fig. 1). When it is greater than zero, no shear damage happens to the cap rock, when it equals zero, shear damage occurs to the cap rock. 1.2.4.
Risk of fault instability due to shear slip
Fault instability due to shear slip has similar principle with shear damage of cap rock, but the fault is the geologic crushed zone, the largest mechanical weak plane, so the cohesion is generally ignored. Geomechanical study shows[79] during the ground stress disturbance caused by gas storage injection-withdrawal, when the shear stress acting on the fault plane is greater than the product of friction coefficient and effective normal stress, the fault will slip, losing sealing capacity. The slip and instability of far field fault will cause substantial formation deformation, and affect wellbore integrity in turn. Therefore, the mechanical stability of fault will directly dictate its dynamic sealing property. Limited by experiment modeling equipment and prepara-
Fig. 1.
Mohr circle of cap rock shear damage.
tion of fault sample etc, currently the stability of fault is mainly evaluated by geo-mechanical numerical simulation. By building mechanical grid model of fault, the Coulomb damage function is taken as the main index to evaluate the stability of fault under different injection-withdrawal conditions, and find out the pressure limit a sealing trap can withstand through finite numerical simulation. When the Coulomb damage function is less than zero, equal to zero and greater than zero, the fault is stable, at critical state, and loses stability due to shear slip respectively.
2.
Design of gas storage capacity parameters
During low speed depletion development of a gas reservoir, a single well can control a large area and large reserves, and a high proportion of gas-bearing pores is involved in gas seepage. In contrast, a gas storage is often transformed from a gas reservoir in the late stage of development or even a depleted gas reservoir when the gas reservoir turns smaller in storage space and poorer in seepage condition, coupled with high speed repeated injection and withdrawalal under high pressure difference, the gas storage will see worsening of storage and seepage conditions, shrink of sing well control area and reduction of gas-bearing pore proportion involved in seepage. Therefore, the effective capacity of a gas storage isn’t simply equal to the dynamic geologic reserves of the gas reservoir. Fine geologic study and injection-withdrawal seepage experiment are needed to find out the mechanism and main control factors of the micro-seepage in the gas storage, and quantify the effective pore volume, so the gas storage capacity prediction model can be established to guide the design of storage capacity. 2.1.
Analysis of gas-bearing pore space
The life expectancy of gas reservoirs developed at low speed with depletion drive is 20-30 years in general, the gas reservoir features large gas drainage radius, infinite fluid supply range, and high geologic reserve producing degree. In contrast, in the operation of gas storage, gas is injected and recovered at high speed in cycle of 1 year, fluid far from the well can’t flow to the well bottom in time, so the well control radius drops significantly. In addition, to meet the requirement of winter peak shaving, the single well proration can be as high as 5-10 times the rate of gas reservoir development, so after the well in a gas storage is opened, the gas flow reaches high speed pseudosteady state flow soon, the effect of formation heterogeneity increases and the single well control range reduces further. Clearly, compared with the gas reservoir development period, the single well in gas storage has much lower control degree on sandbody, which has been proved by the Banqiao gas storage group. Taking Ban 876 as an example, the angle between the gas storage capacity curve and vertical axis is much smaller than that between the gas reservoir pressure drop curve and vertical axis (Fig. 2), that means the gas recovered from the gas storage is less than gas
842
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
Fig. 2. Comparison of gas storage capacity curves with the pressure drop curve during gas reservoir development.
produced from the gas reservoir under a unit formation pressure drop. Obviously, the dynamic geologic reserves of gas reservoir can’t be fully produced during the short high speed injection-withdrawal cycle of gas storage[1011]. Besides, invasion of edge or bottom water, retrograde gas condesate loss, plastic deformation of reservoir caused by stress sensitivity, and gas isolated by water-gas interlocking all can make the gas-bearing pore volume reduce further in the process of gas reservoir development and gas storage operation. 2.2. Evaluation method of effective pore volume after gas storage construction Based on results of geologic, dynamic and seepage mechanism researches, and the distribution features and variation trend of fluid in vertical direction during gas reservoir development and gas storage operation, the gas storage operation profile is simplified into four zones and four contacts (Fig. 3). For a gas storage transformed from a depleted water-invaded gas reservoir, the gas-water contact in the gas storage has moved upward from the original GWC0 of the gas reservoir to GWC1 at the beginning of gas storage construction. When the gas storage operates between the upper and lower pressure limits, a relative stable transition zone will form due to the injection-withdrawal breathing effect, with its top corresponding to the gas-water contact at lower pressure limit GWC2, and its bottom corresponding to the gas-water contact at upper pressure limit GWC3. As during the gas reservoir development and gas storage operation, no formation water invaded, the zone above GWC1 is a pure gas zone. Since gas injection pushes water downward, after multiple rounds of gas injection, the zone between GWC1 and GWC2 becomes a pure gas zone. These two zones are the major gas storage zones, where a high proportion of gasbearing pores involves in the gas seepage. During the operation of gas storage, the gas-water transition zone experiences water in and out and alternating displacement, with lower operation efficiency, but pore space in this zone involved in seepage is not negligible. In comparison, the watered-out zone is occupied by formation water during the whole process of gas reservoir development and gas storage
Fig. 3.
Evaluation of effective pore space in gas storage by zone.
operation, unable to be used to form working gas. In the gas-water seepage zone of the gas storage, the seepage mechanism is complex, and controlled by reservoir physical properties and heterogeneity, wettability, and capillary pressure etc microscopically. For highly heterogeneous reservoirs, water channeling or selective water invasion during high speed gas production blocks the gas in small pores and micro-fissures in the form of slipstream. Due to the additional resistance produced by the Jamin effect, the uncontinuous flow gas will be stuck, forming confined gas, and there is some confined gas in unconnected pores or their blind end [12-15]. Moreover, during gas storage operation, gas displacing water still mainly work in large pores and throats, as water is tiny pores and throats is hard to displace. Also bottom and edge water invasion and retrograde condensation of condensate gas will take up some pore space, and plastic deformation caused by reservoir stress sensitivity can lead to drop of reservoir porosity and permeability. The gas-bearing pore volume calculated by the gas reservoir development dynamic method minus the immobile pore space of each zone is the effective pore volume of each zone, and the total effective pore volume of the gas storage can be expressed as: Veff V0 V1 V2 V3 V4 (1) 2.3.
Prediction model of effective gas storage volume
After the gas storage is built, the gas stored in the storage includes gas left before gas storage construction and injected gas, and the effective gas storage volume is the sum of effetive gas storage volume before the storage construction and cumulative injected gas volume: Gr Gr0 Ginj (2) From the underground pore volume balance principle, as the formation pressure increases after gas storage construction, the effective gas storage volume after storage construction
843
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
corresponds to effective pore volume under the pressure at this point. Thus, the effective pore volume of the gas storage under the formation pressure can be calculated by equation (2). For gas storage transformed from water invaded reservoir, on the basis of effective pore volume of the gas storage, the effective pore volume of the gas storage under different pressures can be worked out by considering the irreducible water volume and rock elastic deformation during gas storage construction and operation. Therefore the effective pore volumes at upper and lower pressure limits can be calculated. Then the storage capacity parameters (effective storage volume, cushion gas volume, working gas volume, and additional cushion gas volume) can be worked out by the equations below:
Gmax Gmin
Veff ,max Bg,max Veff ,min Bg,min
(3) (4)
Gwg Gmax Gmin
(5)
Gadd Gmin Gr0
(6)
3. Design of injection-withdrawal well pattern of a gas storage The layout of injection-withdrawal well pattern of a gas storage should consider the high speed injection and withdrawal feature of the gas storage, and meet the requirement of controlling the gas storage volume effectively and uneven peak shaving gas production. Therefore, based on evaluation of single well productivity, well control and analysis of market gas demand, the minimum well number meeting the gas storage parameter design will be determined, proper well pattern density will be proposed, and the layout of injectionwithdrawal well pattern should consider the structure position, reservoir features and fluid distribution etc comprehensively. 3.1. Main factors affecting injectivity and productivity of wells in a gas storage The gas injectivity and productivity of wells in a gas storage under high speed unsteady seepage flow are affected by near wellbore seepage condition and well controlled storage volume largely. The seepage conditions near wellbore which directly determine the instantaneous injectivity and productivity of the gas well, are affected by the reservoir effective permeability, well bottom perfection degree, reservoir producing degree and drainage area etc primarily. Therefore, removing pollution and connecting reservoir through reservoir micro-stimulation can enhance reservoir permeability, meanwhile, with the injection and withdrawal going on, the formation seepage conditions will become better. Big tubing diameter should be adopted as far as possible to increase wellbore drainage area, and the combination of vertical well, horizontal well and highly deviated well should be used to increase perforated thickness
and area of reservoir, and single well productivity. In addition, sanding out of reservoir, liquid loading in wellbore, tubing string corrosion, and wellhead minimum pressure etc all may restrict the play of the formation seepage capacity, so optimization at multiple nodes of formation, wellbore and wellhead should be carried out to get the reasonable production under different tubing diameters, well types, formation pressures and wellhead pressure. The control radius of a gas storage well under high speed unstable percolation directly determine the effective fluid supply flow. At present, the injection-withdrawal cycle of gas storages in China is around half a year, less than one fiftyth of the gas reservoir development span, so there is no time for fluid far from the well to flow to the well bottom. Apparently, in the limited well control scope, the culmulative gas a well can absorb and produce in a cycle are limited, restricting gas injectivity and productivity of the well. 3.2. Optimization of production of wells in a gas storage at multiple nodes Different from nodal analysis for gas reservoir development, gas storage operation focuses more on multiple node optimization of formation, wellbore and wellhead, to obtain a reasonable daily gas injection/production rate under high speed unstable seepage flow. In terms of fluid flow in reservoirs, due to repeated injection and withdrawal, the reservoirs may decrease in effective permeability due to stress sensitivity or sanding. For the pure gas zone formed by gas displacing water, after multiple cycles of gas injection, the reservoir seepage conditions would be improved, and gas productivity enhanced. For the gas-water transition zone, with the rise of water saturation after multiple rounds of operation, the gas relative permeability would drop and tend to be stable in late stage. As the reservoir seepage situation of wells changes constantly during multiple injection-withdrawal process, the productivity of well changes accordingly[16]. Therefore, the productivity prediction equation for gas wells in storage is established by correcting key parameters such as effective permeability and stress sensitivity of reservir and natural gas property etc. For vertical pipe flow, besides consideration of critical liquid carrying speed and wellbore erosion, to meet the requirement of large volume huff and puff, large size tubing and horizontal well and highly deviated well should be adopted as far as possible. With respect to wellhead exporting conditions, the major aim is to meet the lowest pressure needed for gas to be sent into the long exporting pipeline, and the boosted export should be avoided as far as possible. If economically feasible, the gas export pressure can be lowered properly to enhance the peak shaving capacity of the gas storage. Finally, the productivity equation is modified under the condition that the gas storage reaches stable injection-withdrawal state. By combining the equation with vertical pipe
844
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
flow equation, and taking the critical sanding pressure difference, critical liquid carrying volume, erosion flow and wellhead minimum export pressure as constraint conditions, comprehensive nodal pressure analysis is conducted for different tubing diameters, well types at different formation pressures and wellhead pressures to get the reasonable production of the gas storage at different pressures.
Considering reservoir effective thickness, gas saturation and porosity etc, equation (7) can be simplified as:
3.3. Design of well pattern density for a gas storage under the high speed unstable seepage condition
By using well production from multi-node optimization, the number of injection and withdrawal wells of the gas storage can be calculated according to days of gas production, single well reasonable production rate at the end of gas recovery, and designed working gas volume:
The essence of reasonable well pattern design for a gas storage is to figure out the minimum well number satisfying the designed indexes of storage. The experience of gas storage operation shows wells have higher gas injection capacity than gas production capacity, and longer gas injection time than gas production time. Therefore, gas production capacity is the key dictating the number of wells needed. It is suggested to determine the reasonable number of wells by considering the well control radius under unstable seepage flow, single well coordinated production and monthly gas demand for peak shaving. 3.3.1. Statistics on finite well control radius under high speed unstable seepage flow As one of the first batch commercial gas storages built in China, the Banqiao storage group has been operated for 16 cycles, accumulating a lot of operation dynamic data. By diagnosing and evaluating the gas production performance of 71 wells over multiple cycles with the rate transient analysis software RTA, the relationship chart between well control radius and reservoir effective permeability has been compiled (Fig. 4), which shows the two have good correlation. If the effective permeability of a reservoir is given, the well control radius can be obtained by analogue. The minimum number of wells needed to control the gas storage capacity can be obtained by diving the gas storage capacity by average storage capacity per well:
n1
Gmax Gwk
(7)
Fig. 4. Relationship between single well control radius and reservoir effective permeability of the Banqiao storage group.
n1
S πRe2
(8)
3.3.2. Formation-wellbore-wellhead multi-node coordinated production method
n2 104 3.3.3.
Gwg
(9)
tqg
Estimation of monthly gas demand
The daily gas production in each month can be calculated by using the working gas volume of the gas storage and the monthly uneven demand coefficent. The mathematical model is built by coupling periodical cumulative gas production, formation pressure and the reasonable gas production under this pressure. The maximum number of injection and withdrawal wells are those at the peak of gas production and the end of gas recovery:
n3 max 104 Gp
10
3.3.4. wells
4
Gp
m m / tq m m / tq j
j
end
max
g, max
,
g,end
(10)
Reasonable number of injection and withdrawal
Since the gas storage operation must satisfy well control material balance, single well productivity and uneven gas demand at the same time, the number of injection and withdrawal wells needed for the storage take the largest of the above three results: n max( n1 , n2 , n3 ) (11)
4.
Design of gas storage monitoring scheme
A scientific and reasonable monitoring scheme is the guarantee for safe and smooth running of a gas storage. Dynamic monitoring of a gas storage over the whole life span is a high concern abroad, and a complete monitoring system centering on well engineering, trap sealing and gas storage performance has been established (Fig. 5). The monitoring methods and means include conventional temperature and pressure test, special logging, micro-seismic, productivity test and interferometry synthetic aperture radar (InSAR) system. Enough monitoring wells deployed on the boundary and inside of the gas storage, plus space-earth observation system and GPS constitute an integrated space-earth monitoring network. The monitoring wells should be 0.5-1 times even 1.5 times of injection and withdrawal wells, and mostly newly drilled, with a small number of old wells. With a whole process monitoring installed and running, we can get first-
845
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
of inner diameter, perforation quality and string structure etc are checked regularly, and wells and positions of poor cementing quality are monitored especially. The well structure quality of old and new wells are generally monitored by cementing acoustic logging, acoustic variable density log and electromagnetic defect logging etc, and only wells meeting the integrity standard can be used for the gas storage. 4.2.2.
Fig. 5. Monitoring well net of a gas storage transformed from a gas reservoir.
hand data on the safety and operation performance of the gas storage. 4.1.
Monitoring system
The monitoring system consists of three aspects, well engineering, trap sealing and performance of the gas storage, and can monitor the whole process from construction to operation of the gas storage. (1) Well engineering monitoring. Quality test of old well structure before gas storage construction, monitoring and formation testing of newly drilled and completed well, and downhole technical condition monitoring are conducted to ensure integrity of gas wells in the gas storage. (2) Monitoring trap sealing property. The cap rock, fault system, spill point, surrounding reservoir and overlying permeable layer and shallow aquifer in the gas-bearing area are monitored for gas leak, to ensure the gas injected into the gas storage is held securely. (3) Operation performance monitoring of the inner gas storage. Production performance of injection and withdrawal wells, gas reservoir temperature and fluid property, gas-water contact and fluid migration, and productivity of injection and withdrawal wells etc are monitored to guide gas storage expansion and production, proration and working system adjustment. 4.2. 4.2.1.
Monitoring modes Well engineering monitoring
Before the construction of a gas storage, pressure test can be used to find out the casing corrosion degree and cementing quality of old wells. The drilling of new wells should be monitored according to geologic design, and the formation testing of new wells should follow relevant codes of the petroleum industry. The downhole technical conditions during gas storage operation are monitored by building a casing monitoring system, the casing, damage and corrosion of collar, variation
Monitoring of trap sealing property
The sites with thinner cap rock, higher lateral and vertical sealing risks on both sides of control faults of the gas storage area are monitored by old wells or newly drilled wells, from which, pressure is observed and samples are taken for analysis regularly. The sites of cap rock with possible lithologic change, thinning and more fractures are monitored by old wells or newly drilled wells, from which, pressure is observed and samples are taken for analysis regularly. So are the perimeter of the gas storage, spill point of trap and shallow aquifer etc. In addition, geochemical monitoring and radon-thoron micro-element test are used abroad to identify gas leak directly. Microseismic technique is employed to interpret the microseismic events caused by alternating stress during gas injection and withdrawal, to tell the strength of stress-strain, and judge the risk of trap sealing failure. The newly emerging space-earth observation technique, for example the InSAR and GPS combined system can monitor the ground subsidence around the gas storage area. 4.2.3.
Monitoring of inner performance
(1) Monitoring of injection and withdrawal well performance. The items need to be monitored in gas production well include choke, oil, gas and water production, tubing and casing pressure and wellhead temperature etc. Things needed to monitored in gas injection wells include gas injection rate, pressure and temperature of compressor, and casing and tubing pressure and wellhead temperature etc. (2) Monitoring of temperature, pressure and fluid property in gas reservoirs. Old or new wells are used to monitor static temperature and geothermal gradient, static pressure and pressure gradient of the reservoirs, and fluid samples are taken from the wells for analysis to get information on the variation of fluid properties in the reservoirs over mulitple cycles of injection and withdrawal. (3) Monitoring of gas-water contact and fluid migration. Old wells or new wells are deployed at the structural mediumhigh positions, transition zone and perimeter of the storage to monitor fluid migration direction and variation of gas-water contact. Gas-water contact device and seismic tomography can be used to detect the gas-water contact; induced polarization electric logging, high precision gravity test, pulse spectrum gamma logging etc can be employed to detect gas-water contact and fluid migration direction; and 4D seismic and
846
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
tracer can be applied to monitor fluid migration direction and variation trend of gas flooding front. (4) Productivity monitoring. Systematic well test and transient well test of different zones are conducted in representative wells at different gas injection (withdrawal) periods. In 2-3 cycles, all wells will be tested in turn, to get well productivity, effective permeability, formation pressure, well integrity coefficient, formation connectivity and formation boundary property etc, to figure out the variation of gas injectivity and productivity of wells over multiple cycles. For commingled injection or production, the gas injection profile logging is needed to know the gas absorption of single layer, and gas production profile logging is needed to find out the gas recovery and produced fluid components of individual layers.
5. 5.1.
Case study Main geologic characteristics of the gas storage
The gas storage, one of six storages started to build in 2010, is large in burial depth (3 580 m), large in reserve scale (126×108 m3), medium in reservoir physical property (with a porosity of 19.4% and permeability of 48.6×103 μm2), and strongly heterogeneous. The storage is transformed from a deep sandstone lean condensate gas reservoir with bottom water controlled by lithologic and structure jointly. There are 3 major faults with big fault throw (200 m) and long extension (20 km) in the gas-bearing area. Affected by underwater distributary channel environment, the reservoir is medium in physical property, highly variable in sand distribution, strong in heterogeneity, which would impact producing of reserve negatively. In addition, with the development of the gas reservoir, edge water invaded into the reservoir, forming water and gas seepage zone with complex seepage mechanism, which will lower producing degree further. 5.2. Key parameters in gas storage construction evaluation In consideration of the large faults in the gas reservoir trap and the sealing property of the reservoir during its own formation process, the original formation pressure 34 MPa is taken as the upper limit pressure of the gas storage. By taking reservoir physical property and heterogeneity, edge water invasion, retrograde condensation loss and the effect of high speed unstable seepage flow on pore space into account, the immobile pore volume is estimated at 432×104 m3, accounting for 10% of the original gas-bearing pore volume. Thus, the gas storage is estimated to have an effective pore volume of 4 018×104 m3, a storage capacity of 107×108 m3, working gas volume of 45×108 m3, cushion gas volume of 62×108 m3, and additional cushion gas volume of 16.5×108 m3. Referring to the relationship chart between well control radius and reservoir effective permeability (Fig. 4), the average well control radius from analogue is 400 m. In
consideration of the injection and withdrawal capacity of well and peak shaving demand, the number of injection and withdrawal wells are decided at 30, including 29 vertical wells and one horizontal well. The gas storage has an average daily gas injection rate of 1 550×104 m3 and average daily gas production rate of 2 800×104 m3. At last, four conventional monitoring wells are designed to survey the sealing of cap rock and fault, gas-water contact and fluid migration, temperature and fluid property etc constantly. Moreover, micro-seismic monitoring is deployed in six shallow wells and three medium depth wells at the outer side of the boundary fault and inside the gas storage to keep watching the sealing of the trap. 5.3.
Performance of the storage
The gas storage has operated for four cycles so far, the storage indexes obtained from dynamic data are in good agreement with the indexes in the design plan (Fig. 6). The gas storage has been running well on the whole, with higher storage capacity recovery level and fast production rate. Fig. 7 shows the storage capacity at the end of the fourth cycle is 96×108 m3, 90% of the original storage capacity; and its peak shaving capacity has increased by four times from the
Fig. 6. Reviewed storage capacity curve of the storage after multiple cycles of operation.
847
Fig. 7.
Predicted indexes of the storage over the years.
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
initial 2.7×108 m3 to 14.0×108 m3. It is expected to reach the designed storage capacity after two years (Fig. 7), basically consistent with the capacity reaching cycle of six years in the design plan, and the capacity reaching cycle of 5-8 years of storages abroad, which proves the geologic design plan of the storage is scientific and reliable.
6.
n—reasonable number of injection and withdrawal wells; n1, n2, n3—numbers of injection and withdrawal wells obtained from limited well control radius method, formation-wellbore-wellhead multi-node coordinated production method, and monthly uneven gas demand estimation method; qg—reasonable production of single well, 104 m3/d; qg,end—reasonable gas well production under the formation pres-
Conclusions
sure at the end month of the peak shaving gas production, 104 m3/d;
In light of the differences between gas storage operation and gas reservoir development, four key technologies for gas storage design, evaluation of trap dynamic sealing, storage parameter design, optimization of injection and withdrawal well pattern, and monitoring plan design have been established through the study, to solve preservation conditions, capacity, high efficient injection and withdrawal, and safe operation of the gas storage, which enriches and improves the storage construction evaluation system. Through reviewing the construction and operation experience of the Banqiao storage group, we have updated the gas storage design concept, and advanced the ideas of effective storage capacity, limited well control under high speed unstable seepage flow, and variation of well productivity in the storage over operation cycle, laying theoretical foundation for storage parameter design and injection and withdrawal well pattern deployment. An application case shows that the gas storage operation indexes are highly consistent with the designed indexes, proving the key technologies for gas storage geologic plan design are scientific and reliable. These technologies can provide theoretical basis and technical support for similar gas storage construction in China.
qg,max—reasonable gas well production under the formation pressure at the highest gas production month of the peak shaving, 104 m3/d; Re—radius controlled by one well, m; S—effective gas-bearing area of the reservoir, m2; t —days of gas production, d; V0 —underground pore volume corresponding to the reservoir dynamic geologic reserves, m3; V1—immobile pore volume in the pure gas zone before gas storage construction, m3; V2—immobile pore volume in the pure gas zone formed by gas displacing water, m3; V3—immobile pore volume in gas-water transit zone, m3; V4—immobile pore volume in watered-out zone, m3; Veff —effective pore volume of gas storage, m3; Veff,max—effective pore volume corresponding to the upper pressure limit, 108 m3; Veff,min—effective pore volume corresponding to the lower pressure limit, 108 m3; σ—effective normal stress, MPa; σ1, σ3—maximum and minimum effective normal stress, MPa; τ—shear stress, MPa; τm—maximum shear stress under an injection-withdrawal condition, MPa;
Nomenclature
m —critical shear stress at which shear damage occurs, MPa.
Bg,max—gas volume factor corresponding to the pressure upper
References
limit; Bg,min—gas volume factor corresponding to the pressure lower lim-
[1]
it;
DING Guosheng, LI Chun, WANG Jieming, et al. The status quo and technical development direction of underground gas
Gadd—additional cushion gas volume, 108 m3;
storage in China. Natural Gas Industry, 2015, 35(11): 107–112.
Ginj—cumulative gas injection volume, 108 m3;
[2]
Gmax—storage capacity, 108 m3;
BRUNO M S, KANG L, DIESSL J, et al. Development of improved caprock integrity analysis and risk assessment tech-
Gmin—cushion gas volume, 108 m3;
niques. Energy Procedia, 2014, 63: 4708–4744.
Gp—peak shaving gas demand, 108 m3;
[3]
ORLIC B, WASSING B B T, GEEL C R. Field scale geome-
Gr—storage effective capacity, 108 m3;
chanical modeling for prediction of fault stability during un-
Gr0—effective storage capacity before storage construction, 108
derground gas storage operations in a depleted gas field in the
m3;
Netherlands. ARMA 13-300, 2013.
Gwg—working gas volume, 108 m3;
[4]
TENTHOREY E, VIDAL-GILBERT S, BACKÉ G, et al.
Gwk—average storage capacity controlled by one well, 108 m3;
Modelling the geomechanics of gas storage: A case study from
K—reservoir effective permeability, 103 μm2;
the Iona gas field, Australia. International Journal of Green-
mend—uneven factor of the end month of peak shaving gas production;
house Gas Control, 2013, 13: 138–148. [5]
ANSAR S, HAIGH R, KHOSRAVI N, et al. Caprock integrity
mj —uneven factor of the jth month;
case study for non-thermal polymer flooding project using 4D
mmax—uneven factor of month with high peak shaving gas produc-
reservoir coupled geomechanical simulation. SPE 157912,
tion;
2012.
848
ZHENG Dewen et al. / Petroleum Exploration and Development, 2017, 44(5): 840–849
[6]
[7]
BILDSTEIN O, JULLIEN M, CRÉDOZ A, et al. Integrated
[11] XU Hongcheng, WANG Jieming, LI Chun, et al. Inventory
modeling and experimental approach for caprock integrity,
verification of underground gas storage based on a flooded
risk analysis, and long term safety assessment. Energy Proce-
and depleted gas reservoir. Natural Gas Industry, 2010, 30(8):
dia, 2009, 1(1): 3237–3244.
79–82.
TEATINI P, CASTELLETTO N, GAMBOLATI G. 3D ge-
[12] CIESLINSKI J T, MOSDORF R. Gas bubble dynamics ex-
omechanical modeling for CO2 geological storage in faulted
periment and fractal analysis. International Journal of Heat
formations: A case study in an offshore northern Adriatic reservoir, Italy. International Journal of Greenhouse Gas Control, [8]
2014, 22(3): 63–76.
forecast in underground gas storage based on modified mate-
VIDAL-GILBERT S, TENTHOREY E, DEWHURST D, et al.
rial balance equation. Petroleum Exploration and Develop-
Geomechanical analysis of the Naylor Field, Otway Basin,
[9]
and Mass Transfer, 2005, 48(9): 1808–1818. [13] TANG Ligen, WANG Jieming, BAI Fengjuan, et al. Inventory
ment, 2014, 41(4): 480–484.
Australia: Implications for CO2 injection and storage. Interna-
[14] SHI Lei, LIAO Guangzhi, XIONG Wei, et al. Gas water per-
tional Journal of Greenhouse Gas Control, 2010, 4(5):
colation mechanism in underground storage built on water
827–839.
drive sandstone gas reservoir. Natural Gas Industry, 2012,
TEATINI P, CASTELLETTO N, FERRONATO M, et al.
32(9): 85–87.
Geomechanical response to seasonal gas storage in depleted
[15] DING Yunhong, ZHANG Qian, ZHENG Dewen, et al. Seep-
reservoirs: A case study in the Po River basin, Italy. Journal of
age laws in converting a micro fissure-pore carbonate gas reservoir into UGS. Natural Gas Industry, 2015, 35(1): 109–114.
Geophysical Research, 2011, 116(F2): 1–21. [10] XU Hongcheng, WANG Jieming, QU Ping, et al. A prediction
[16] TANG Ligen, WANG Jieming, DING Guosheng, et al.
model of storage capacity parameters of geologically-com-
Downhole inflow-performance forecast for underground gas
plicated reservoir-type underground gas storage. Natural Gas
storage based on gas reservoir development data. Petroleum
Industry, 2015, 35(1): 103–108.
Exploration and Development, 2016, 43(1): 127–130.
849