GasWaterRock Interactions and Implications for Geoenvironmental Issues
View this Special IssueResearch Article  Open Access
Investigation on the Relationship between Wellhead Injection Pressure and Injection Rate for Practical Injection Control in CO_{2} Geological Storage Projects
Abstract
The existing investigations on the maximum allowable wellhead injection pressure have found the upper limit of wellhead injection pressure, which, however, cannot provide a practical operational designing scheme of wellhead injection parameters for CO_{2} geological storage projects. Therefore, this work firstly proposes the complete constraint conditions of wellbore injection to realize the whole process of forward and inverse calculations of wellbore pressure and then applies it to explore the relationship between wellhead injection pressure and injection rate. The results show that the wellhead injection pressure and the injection rate are a pair of mutually constrained physical quantities. For a certain injection project, the allowable wellhead injection pressure and injection rate separately form a continuous interval. Change of one quantity within its allowable interval will also change the other within its interval, both jointly forming a closed region. Thus, controlling the wellhead injection parameters in this closed region can simultaneously ensure the effectiveness and safety of injection. Subsequently, this work further studies the factors of impacting the relationship between wellhead injection pressure and injection rate and finds that all the temperature of injected fluid, the parameters of saturation, and the characteristic parameters of reservoirs only change their upper and lower limits to some extent but have no essential effects on their relationship. Application of this theory and method in Shenhua CCS demonstration project obtained the relationship diagram of wellhead injection pressure and injection rate, which found that its actual injection parameters just fall into the closed region of the relationship diagram, effectively verifying the reliability of this work.
1. Introduction
In the ChinaUS joint announcement on climate change, 2014, China intended to achieve the peaking of CO_{2} emissions around 2030 and to make best efforts to peak early [1]. To achieve this goal, carbon capture and storage (CCS), as one of the most effective approaches for greatly reducing CO_{2} level in the atmosphere [2–5], is expected to gradually advance from demonstration project to commercialization. The general procedure of CO_{2} geological storage includes site selection, well drilling, injection, monitoring, and evaluation [6, 7], among which the most key step is CO_{2} injection. Therefore, it is very important to control wellbore injection parameters [8, 9]. Due to the great burial depth of the storage sites (means high temperature and high pressure), CO_{2} injection (resulting complex twophase flow in pores and cracks), and the complexity and uncertainty of geological conditions, a great challenge to the effectiveness and safety of CO_{2} injection is induced [2, 10].
Currently, it is widely accepted by engineers that controlling the maximum bottom hole pressure is practical and reliable to avoid the strata fracturing [11–13]. Based on that, the more convenient method is to control the wellhead injection pressure [13], which has been verified sufficiently in In Salah [14] and Ketzin and Shenhua CCS projects [13, 15]. To obtain the maximum allowable wellhead injection pressure for CCS projects, Carroll and Lui [16] and Hashim and Maloka [17] analyzed the main influential factors of wellhead pressure, and Streit and Hillis [18], Rutqvist et al. [10], and Gapillou et al. [19] proposed some approximate estimation methods and engineering experience. Subsequently, Bai et al. [8] developed a new fast explicit finite difference method (FEFDM) for calculating the wellhead pressure under assumption of single reservoir, and Wu et al. [20] presented a new estimation method of wellbore pressure from the perspective of engineering. However, many CCS projects still use the typical approach referring from the field of enhanced oil recovery (EOR) and sour gas reinjection for the advantages of simplicity and extensive application experience [21], although there are four obvious shortcomings [13]. Therefore, Bai et al. [13] developed a methodology for determining the maximum allowable wellhead injection pressure based on the precious work recently, which includes the control conditions for bottom hole safety, the inversion method of wellhead pressure, and the design coefficients on the basis of engineering experience and analogy. It is available for multireservoirs injection simultaneously. However, it is worthy to concern that the above investigations are mainly focused on the wellhead injection pressure, especially on the maximum allowable wellhead injection pressure, which ignores the wellhead injection rate. Although the injection rate is usually set as the project target, it is also a significant parameter of injection control. What is more, no investigation has addressed the relationship between wellhead injection pressure and injection rate.
Thus, as to a determined target injection flow, the above methods of designing wellhead injection pressure can only provide the maximum allowable wellhead injection pressure. However, the actual wellhead injection pressure must be less than its maximum value in consideration of safety, so how much should the applicable wellhead injection pressure be? Could any wellhead injection pressure bellowing the maximum allowable wellhead injection pressure ensure that the target injection flow will enter the reservoirs completely? Of course, the answer is no. That is to say, the applicable wellhead injection pressure not only has an upper limit but also has a lower limit. Only the pressure of ranging from the lower limit to the upper limit can ensure that the target injection flow will enter the reservoirs completely. Another question is how does the applicable wellhead injection pressure change when the target injection flow changes? It would be transferred to the schema of constant pressure controlling; thus the problem is how to determine the applicable target injection flow under knowing the wellhead injection pressure. Apparently, to answer these questions exactly, it is necessary to investigate the wellhead injection pressure and the injection rate simultaneously and to master the internal relationship between wellhead injection pressure and injection rate clearly. In addition, to facilitate the application in projects, it is worthy to study their influential factors.
Therefore, this work will firstly improve the constraint conditions of wellbore injection from the perspective of flow rate to realize the whole process of forward and inverse calculations of wellbore pressure and then explore the relationship between wellhead injection pressure and injection rate and its influential factors. Finally, we apply this method to Shenhua CCS demonstration project to find the feasible ranges of wellhead injection pressure and injection rate and to verify the reliability of this work simultaneously.
2. Theory and Method
2.1. Constraint Conditions of Wellbore Injection
Figure 1 shows the schematic of injection wellbore and its related reservoirs. There is one injection well and layers of reservoircaprock combination units from top to bottom. The wellbore is divided into segments by the reservoircaprock unit. The reservoircaprock units, reservoirs, caprocks, and well segments are numbered as from top to bottom. In the following, the subscript denotes the corresponding physical quantities of the th reservoircaprock unit, the superscript indicates the physical quantities within the reservoir, and the superscript marks the physical quantities within the well segments.
As described in Bai et al. [13], when CO_{2} is injected at a given wellhead injection pressure [Pa] into the injection well, the flow not only should satisfy its related fluid dynamics equations but also should not damage the stratum. Moreover, the injection rate [kg/s] should reach the preset target injection flow. These are enough for the inverse calculation of wellhead injection parameters based on the bottom hole conditions, whereas for the forward calculation of wellbore pressure and flow rate distribution on the basis of wellhead injection parameters, they are not enough because the forward calculation is unable to ensure that the injected fluid will enter reservoirs completely only from the perspective of pressure. Therefore, to realize the process of forward calculation, must be less than the total available capacity of all the reservoirs and the actual flow rate of each reservoir also must be below its available capacity; then the effectiveness and safety of injection can be ensured.
Hence, the complete constraint conditions of wellbore injection arewhere is the formation pressure (Pa), is the capillary pressure (Pa), is the pressure on the interface of wellbore and reservoir (Pa), is the maximum allowable pressure of reservoirs (Pa), is the target injection flow rate (kg/s), denotes the actual flow rate of the reservoir (kg/s), and denotes the available capacity of reservoir (kg/s).
The first inequality of (1) is the constraint condition about pressure. The maximum allowable pressure is a reduced value of fracturing pressure [22]; details can be found in Bai et al. [13]. The other two inequalities represent the constraint conditions about flow rate. And the second one is focused on the total flow rate, which requires the injection rate to be higher than the design target but lower than the total available capacity of all the reservoirs. The third one requires the actual flow rate of each reservoir to be within its available capacity. Obviously, when the third inequality is satisfied, the right half of the second one is met naturally, while the left half of the second one could be confirmed in the feasibility study of project. Therefore, the constraint conditions about flow rate are equivalent to the third inequality in essence. Furthermore, in the actual calculations, only the available capacity of the bottom reservoir needs to be checked because the capacity of the other reservoirs is just equal to their actual flow rate, which is calculated by the following calculation method. Therefore, they always satisfy (1). It should be noted that the above constraint conditions did not involve the caprocks, which are beyond our discussion.
2.2. Calculation Method
As mentioned above, inverting wellhead parameters based on the bottom hole conditions only can find their maximum value, which, however, can only be used as the upper limit of injection control. Therefore, in practical operation, it is necessary to directly calculate the wellbore pressure and flow rate distribution according to the given wellhead injection parameters and then to judge whether the injection is safe and effective. It is a forward calculation process. The fast explicit finite difference model (EFDM) from wellhead to bottom can be derived based on the continuity equation of steady flow, the motion equation of vertical wellbore, and the state equation of fluid [8, 9] as follows:where is the wellbore pressure (Pa); is the mass flow rate of wellbore cross section (kg/m^{2}/s); is the differential element (m); is the acceleration of gravity (m/s^{2}); is the interior diameter of injection tube (m); is the molar mass of gas (kg/mol); is the universal gas constant; is the friction coefficient; is the compression factor; and is the temperature (). The subscript is the number of differential elements within a certain well segment.
CO_{2} is injected from wellhead through wellbore into the reservoirs; due to the presence of brine in the reservoir pores, the flow changes into twophase flow [23–25]. In accordance with the twophase flow theory, the fluidrock characters determine the relative permeability of each phase [26–29]. The functional relationship of the flow rate flowing into the th reservoir and the pressure can be obtained under the assumption of steady flow [27], orwhere is the maximum influence radius of flow in the reservoir (m); and are the maximum radii of CO_{2} plume at the bottom and top of the reservoir (m), respectively; is the tubing radius (m); is the absolute permeability of the reservoir (m^{2}); is the thickness of the reservoir (m); is the porosity of the reservoir; is the density of CO_{2} (kg/m^{3}); is the mobility of CO_{2} (m·s/kg); and is the mobility of brine (m·s/kg).
On the basis of (2) and (3), to realize the multireservoir injection simultaneously, the equilibrium condition of flow at the reservoir node is still required. It could be simplified into the flow rate allocation under the condition of constant pressure because all junctions of reservoirs and wellbore are regarded as the equivalent nodes of flow rate allocation and the wellbore pressure between the top and bottom of the reservoir is assumed to be a constant in this work. Hence, the equilibrium condition of flow of the th equivalent node is that the inflow amount of CO_{2} equals the sum of outflow to the next well segment and inflow into the th reservoir. Then, combined with (3), the flow rate allocation relationship at the equivalent node can be deduced as follows [13]:
When (4) is applied to the th reservoir (the bottom reservoir), due to the absence of outflow, the second term of right hand should be considered as zero. Then, according to (4), the available capacity of the bottom reservoir can be determined. And the actual flow rate of the bottom reservoir is obtained by (4) with . So that when the latter is not more than the former, the constraint conditions about flow rate are satisfied.
2.3. Calculation Procedure
There are two basic control models on fluid injection, namely, the constant pressure and the constant flow rate. For constant pressure, the allowable interval of flow rate should be solved; and as for constant flow rate, the problem is how to obtain the allowable interval of pressure. However, the results under different control models are the same. Therefore, this work chooses the control model of constant pressure as an example to explore the relationship between wellhead injection pressure and injection rate. The calculation procedures of forward calculation process are listed as follows:(1)Set .(2)Set the initial .(3)Compute the wellbore pressure according to (2); compute the flow rate of flow into the reservoir according to (3) and renew the flow rate of flow to the next well segment according to (4) when the calculation reaches to the reservoir.(4)Judge whether the computed results at the equivalent node meet the constraint condition of (1). If “Yes,” execute the subsequent calculation; if “No,” go to Step (5). After finishing all calculation process, if the calculated results meet all the corresponding inequality conditions in (1), the corresponding and are the feasible wellhead injection parameters.(5)Renew ; repeat Step (3) to Step (4) until both the upper and lower limits of feasible under the condition of are found; if there is no feasible (indicating that cannot achieve injection), go to Step (6).(6)Renew ; repeat Step (2) to Step (5) to find the upper and lower limits of feasible .(7)Plot the relationship diagram between and according to the results.
3. Relationship between Wellhead Injection Pressure and Injection Rate
3.1. Parameters and Injection Schemes
In the following, this work supposed an analysis case to explore the relationship between and . There are four reservoirs and one injection well with a depth of 2000 meters in this case. Table 1 lists the reservoircaprock combination units and the characteristic parameters of the reservoirs. The values of thickness, permeability, and porosity of reservoirs are determined by engineering experience, which represent the typical characteristic (small thickness, low porosity, and low permeability) of reservoirs in China. They could be updated for research.

First, it is necessary to present the wellbore pressure distribution for a determined under different to analyze the characteristics of the wellbore pressure distribution, as shown in Figure 2. Here is given as 13 MPa, while are 0.1, 1, and 6 kg/s.
It is clear from Figure 2 that the wellbore pressure increases approximately linearly with the well depth; the bigger is, the more obviously the pressure increases; when enlarges to a certain value, the wellbore pressure above the reservoirs barely changes with ; the wellbore pressure distribution changes linearly and continuously with , showing no inflection point; thus it is appropriate to use the variation trend of wellbore pressure at one point to represent the variation trend of pressure in the whole wellbore. Therefore, in the following, the change in pressure on the interface of wellbore and reservoir is adopted to replace that of the whole wellbore.
Table 2 shows the simulation injection schemes. When is great enough, the lower limit of the constraint conditions about pressure expressed by the left part of the first inequality in (1) is clearly met, so the four injection schemes (a), (b), (c), and (d) in Table 2 are adopted to study the variation rule of pressure on the interface of wellbore and reservoir with in the condition of no reservoir fracturing. The calculation results are shown in Figures 3(a)–3(d). Subsequently, with reducing, the constraint condition about pressure to prevent the reservoir from fracturing can be naturally satisfied and the lower limit of the constraint conditions about pressure should be considered. The four injection schemes (e), (f), (g), and (h) in Table 2 are adopted for that and the calculation results are shown in Figures 3(e)–3(h).
 
Notes. The calculation step of in (a), (b), (c), and (d) is 0.1 kg/s; for (g) it is 0.02 kg/s; for (e), (f), and (h) it is 0.05 kg/s. 
(a) MPa
(b) MPa
(c) MPa
(d) MPa
(e) MPa
(f) MPa
(g) MPa
(h) MPa
3.2. Results Analysis
There are two types of curves shown in Figures 3(a), 3(b), 3(c), and 3(d): the pressure curves and the flow rate curves. The former are used to analyze whether the calculated results meet the upper limit of the constraint conditions about pressure, and the latter are used to analyze whether the calculated results meet the constraint conditions about flow rate. The specific analyses are shown below.
The dense data segments in the figure result from onetenth of calculation step, which verified that no oscillation singularities existed in the calculations. It aims to accurately find the intersection and its corresponding .
In Figures 3(a), 3(b), and 3(c), with changing, there is an interval that ; as for the other reservoirs, the similar intervals basically disappear or are included in the interval of the first reservoir, which indicates that the critical reservoir [13, 30] is the first one under the above described injection conditions. Hence, only the intersections of the curve and the curve [] are shown in Figures 3(a)–3(d) as the basis for the following discussions.
With changing, all the pressure curves show the same varying trend of upward convex parabola with obvious peak. Based on the above analysis and with Figure 3(a) as an example, the intersections P1 and P2 of the curve and the curve divide into three intervals. It looks like smaller than P1 or greater than P2 would meet the upper limit of the constraint conditions about pressure. However, in fact, this is not true, because the calculations are based on the assumption that the flow in the wellbore reaches to steady state at 10 days after CO_{2} injection; if greatly expanding the time, even a tiny will make the wellbore pressure exceed the allowable pressure at some points. Therefore, as for whose P1 is located before the peak point, less than P_{1} could not be used as the design scheme of the gas injection projects. Moreover, whether higher than P_{2} can be used as the design scheme still needs to be determined by the constraint conditions about flow rate, which can be judged by the flow rate curves of the last reservoir. Apparently, at the left side of intersection F1 of the flow rate curves meet this condition, while at the right side does not. Therefore, for determined , its corresponding that can be used as the design scheme of gas injection must belong to the interval of [P2, F1]. Among the above schemes, in Figure 3(a), F1 is smaller than P2, indicating that there is no feasible under this ; that is, this has exceeded the maximum allowable wellhead injection pressure; in Figure 3(b), F1 is equal to P2, indicating that this is the maximum allowable wellhead injection pressure; in Figure 3(c), F1 is greater than P2, corresponding to the general cases; in Figure 3(d), only F1 exists but there is no intersection of pressure curves, revealing that no matter how changes, the constraint conditions about pressure will be always met under this ; therefore the corresponding interval of flow rate is . It is predictable that the feasible interval of remains with continuous reduction of , although F1 will decrease with declining.
Figures 3(e)–3(h) are basically consistent with Figures 3(a)–3(d) except that the pressure curves in Figures 3(e)–3(h) show the lower limit of the constraint conditions about pressure. According to Figures 3(e)–3(h), as for each injection scheme, the intersections of curve and curve correspond to the minimum , indicating that the injection to the first reservoir is most easily realized. Therefore, the lower limit of the constraint conditions about pressure can be transferred into that as long as ; the injection is achievable. Similarly, conclusion is only focused on the constraint conditions about pressure, and the constraint conditions about flow rate still need to be met. In summary, the feasible interval of is in Figure 3(e), in Figure 3(f), and F1 in Figure 3(g) and does not exist in Figure 3(h), revealing that in Figure 3(g) is corresponding to the minimum achievable wellhead injection pressure. It is predictable that gradually increasing from the minimum achievable wellhead injection pressure will expand the feasible range of , which is consistent with the relationship between and in the consideration of no reservoir fracturing. In the process of one increasing with the other decreasing, they will coincide eventually.
The two abovementioned types of injection schemes represent two limit zones in the relationship diagram between and , respectively. Similarly, the middle and can also be obtained, plotting them in one figure, that is, the relationship diagram between and , as shown in Figure 4, where the red and blue lines represent the upper and lower limits of the interval of flow rate under constant pressure injection, respectively; both of them jointly form a closed region. And the closed region forms the wellhead injection parameters domain of simultaneously meeting the constraint conditions about pressure and flow rate. Hence theoretically, as long as controlling the wellhead injection parameters falling in the closed region, the safe and effective injection can be achieved. In addition, the left intersection of curves represents the lower limit of constraint conditions about pressure, where the CO_{2} fluid is just able to be injected into the reservoir completely. And the right one represents the upper limit of constraint conditions about pressure, in which just no reservoir would fracture. The feasible flow rate corresponding to these two intersections is unique and determined. Therefore, the wellhead injection parameters should be away from the intersections and close to the middle region as much as possible when designing the injection schemes, which will bring a greater allowable interval for pressure and flow rate, and then the projects will be safer.
4. Influential Factors of Wellhead Injection Pressure and Injection Rate
4.1. Overview
The influential factors can be classified into two parts: the humancontrollable factors and the engineering geological factors, as shown in Figure 5. The former mainly refer to the temperature of injected fluid determined by the engineers, and the latter include the factors of impacting wellbore heat transfer, impacting reservoir capacity, and determining constraint conditions about pressure. And the last five in Figure 5 also can be considered as the characteristic parameters of reservoirs.
As for the geothermal gradient, it was studied by Lui et al. [31] in consideration of the single reservoir and they found that it affects the wellbore pressure distribution by impacting the wellbore heat transfer and only has slight effect when is small. Obviously, the impact of earth’s surface temperature is weaker. Since the wellbore heat transfer has nothing to do with the number of reservoirs, it is acceptable that the effect of geothermal parameters on the wellbore pressure distribution can be neglected. Thereafter, the analysis case in Section 3.1 is used to study the effect of the other abovementioned factors.
4.2. Temperature of Injected Fluid
Take the situation of °C as the standard case, compared with the cases of = −10, 0, 10, and 20°C, respectively, to study the influence of on the relationship of and with other parameters unchanged. Figure 6 shows the calculated results.
According to Figure 6, it is clear that the relationship diagrams of and under different are accordant in shape and the areas of different closed region are roughly equal. With increasing, the closed region only shows a tendency of uniformly shifting to the right. Therefore, it could be concluded that has no substantial impact on the relationship of and . Furthermore, with increasing, the allowable exhibits a linearly decreasing trend, and the allowable shows a linearly increasing trend. The reason is that a rise in causes the wellbore pressure to reduce, therefore leading to an increase in the allowable . Subsequently all the constraint conditions about pressure and flow rate changed, especially for the upper limit of constraint condition about pressure.
4.3. Saturation Parameters and
In the above cases, the saturation parameters and are set as 0.55 and 0.999, respectively. To analyze their impact on the relationship between and , firstly it is necessary to design different values of and . Here, the Van GenuchtenMualem Model [32] was adopted to inversely derive the parameters and . Table 3 lists the design values of and and Figures 7 and 8 show the calculation results.

Figures 7 and 8 show that the parameters and just change the ranges of allowable and to some extent and have no substantial effects on the relationship between and , which is similar to . With the decrease of , the allowable and gradually reduce, the upper limit of gradually closes to its lower limit, and the effects of are just contrary to those of because the CO_{2} domain and brine domain of twophase flow are mutually restrictive. The changes of the right intersection of the upper and lower limits of in Figures 7 and 8 have a common feature; that is, the intersection moves strictly along the track of lower limit of with the variable changing. And it differs slightly from that of the left intersection, which suggested that the change of and just alters the constraint conditions about the flow rate and the lower limit of pressure; however, it has no impact on the constraint conditions about the upper limit of pressure.
4.4. Thickness, Permeability, and Porosity of Reservoirs
The thickness, permeability, and porosity of reservoirs as their inherent characteristic parameters are invariant for a given storage site, but they directly determine the capacity of the reservoir. Thus their influence that aims to generalize the above obtained conclusions to other projects is analyzed. Secondly, with CO_{2} injected into the reservoir, this will cause a series of physical and chemical reactions. However, a change in load and chemical reactions with minerals will generally cause changes in the permeability and porosity of reservoirs [33–35]. Limited by the existing testing technology, the impacts of thickness, permeability, and porosity have not been taken into consideration. Tables 4–6 list the different design parameters of reservoirs, and the corresponding calculation results are shown in Figures 9–11.



On the basis of Figures 9 and 10, the impacts of reservoir thickness and permeability on the relationship between and have common features: both only affect the range rather than the essence of and . With the thickness or permeability increasing, the allowable and increase at similar ratio, ascribed to their similar status in (3). Of course, there are also some differences about the impact of reservoir thickness and permeability: curves B and C in the figures are not exactly the same, because the value of is also affected by the thickness but not by the permeability. Curves B and C in Figure 9 do not overlap, revealing that increase in thickness of the upper reservoir has greater effect on enlarging the allowable and in the multireservoir injection simultaneously, compared with increasing the thickness of the lower reservoir. It indicates that the injected fluid majorly enters the upper reservoirs, which is consistent with the conclusions of Rutqvist and Tsang [36], Liu et al. [31], and Xie et al. [37, 38] based on TOUGH2. Furthermore, as described above, the critical reservoir in this case is the first reservoir, which means that increase in thickness of the critical reservoir is most effective for the increase of total available capacity of the entire storage site. The effect of reservoir porosity on and is limited and could be ignored according to Figure 11. The traces of intersections of upper and lower limit of in Figures 9, 10, and 11 are in agreement with those in Figures 7 and 8, indicating that the changes caused by these five parameters are similar in essence; that is, they all indirectly change the constraint conditions about flow rate by changing the capacity of reservoirs.
4.5. Formation Pressure and Fracturing Pressure
The formation pressure in essence also affects the capacity of reservoirs; therefore it has some similarity to the five abovementioned parameters. However, it is also different from them because it also determines the lower limit of constraint conditions about pressure. Thus, it is discussed together with the fracturing pressure. Table 7 lists the different pressure conditions and Figure 12 shows the corresponding calculation results.

Comparison of curves A, B, and C shows that, with the formation pressure increasing, both allowable and decline, and the right intersection of upper and lower limits of still moves along the lower limit of . It indicates that the formation pressure also indirectly changes the constraint conditions about flow rate by affecting the capacity of reservoirs, so there is a same feature with the five aforementioned parameters with similar characteristics. Moreover, the left part of closed region shifts toward the right when the formation pressure increases, which embodies the fact that the formation pressure directly determines the lower limit of constraint conditions about pressure. Obviously, a rise in the lower limit of pressure will naturally increase the allowable . Comparison of curves A, D, and E shows that, with the fracturing pressure increasing, both allowable and will increase, and only the right boundary with the initial trend and shape extends to the right. It suggests that the fracturing pressure only changes the upper limit of constraint conditions about pressure and does not affect the other conditions and parameters.
5. Case Study: Shenhua CCS Demonstration Project
In Section 3, the relationship diagram between and was obtained through an analysis case; in Section 4, relevant factors affecting the relationship and their influential rules were found. All these indicate that the theory and method presented in this paper are scientific and effective in solving engineering problems. As follows, this work will apply the theory and method in Shenhua CCS demonstration project to obtain the ranges of and which simultaneously meet the constraint conditions of pressure and flow rate in the actual project.
The calculation parameters of the reservoircaprock combination units in Shenhua CCS demonstration project are listed in Table 8, including 8 reservoircaprock combination units from top to bottom and one injection well with a depth of 2450 meters. The parameters are mainly determined by the logging interpretation data [13]. As mentioned above, the reservoirs are a kind of formation with small thickness, low porosity, and low permeability. The reservoirs are simplified into homogeneous and isotropous formation because we applied an analytical solution to describe the CO_{2}brine flow in the reservoirs, although the actual reservoirs have strong heterogeneity.

Figure 13 shows the calculation results. and of right intersection are 13.16 MPa and 10.31 kg/s, respectively. If the annual injection time is 300 days, the annual injection mass flow will reach 270,000 tons. Obviously, the preset annual injection of 100,000 tons for Shenhua project is achievable. The actual injection started in March 2011 and stopped in April 2015. Figure 14 shows the monitoring records of wellhead injection pressure during the formal injection stage from March 2011 to April 2014, during which time CO_{2} was injected at a constant injection rate about 4.0 kg/s. Inputting the actual injection parameters into Figure 13 finds that they all fall into the closed region except the records of initial loading process and the other two monitoring records. However, the other two abnormal pressures were induced by some unexpected circumstances, which was controlled by engineers immediately and returned to normal. Hence, it is convincing that the theory and method established by this work can withstand the practice test and are scientifically reliable. Moreover, it is necessary to note that the actual injection is very close to the upper limit curve of at the later period of formal injection. Although it improves the utilization efficiency of cost, any improper operation or uncontrollable factor could possibly lead to an abortive injection and subsequently a series of chain consequences. Therefore, according to the current preset annual injection target, this work recommends to control its at 5~7 MPa in the subsequent injection process.
In addition, according to the conclusions of Section 4 and considering the changes of relevant parameters with time, it is necessary to pay attention to the following points for the actual injection control of Shenhua CCS project. is realtime measurable in the injection process. Thus if rises, it is necessary to properly reduce or increase ; if declines, it is necessary to increase or decrease . Because the parameters of saturation and permeability and fracturing pressure are difficult to measure in operation, it is worthy to investigate the rules that they change with time by laboratorial experiments. Then the wellhead injection parameters should be modified based on the conclusions of experiments. For a given project, the thickness of reservoirs generally does not change over time and the effect of the porosity is small enough and ignorable. As to the impacts of formation pressure, it is unnecessary to consider that because the distant boundary condition is constant pressure in this work.
6. Conclusions
To improve the control theory of the wellhead injection parameters, this work firstly developed the complete constraint conditions of wellbore injection and then studied the relationship and its influential factors between and by applying the EFDM in the vertical well segment and the analytical solution of twophase flow in the horizontal reservoir and obtained the following conclusions.
and are a pair of mutual constraint physical quantities. For a given injection project, the allowable and individually form a continuous interval. When one changes within its allowable range, the other correspondingly alters within its allowable range; both of them jointly form a closed region. Thus, controlling the wellhead injection parameters within this closed region can simultaneously ensure the effectiveness and safety of CO_{2} injection.
Analysis of the factors that affect the relationship between and found that all of them only change their allowable upper and lower limits to some extent but not the essence of their relationship. With the fracturing pressure increasing, their allowable upper limits increase and their lower limits remain unchanged. With the formation pressure increasing, the lower limit of increases and its upper limit remains unchanged; in contrast, the upper limit of decreases. With reservoir thickness and permeability increasing, the allowable upper limits of and significantly increase, and the lower limit of slightly decreases. As enlarges or reduces, the impacts on and are in agreement with that of reservoir thickness decreasing. The impact of reservoir porosity is ignorable. With increasing, the allowable upper and lower limits of enlarge, and those of reduce.
Application of the above theory and method in Shenhua CCS demonstration project obtained the relationship diagram of and and found that its actual injection parameters are just located in the closed region, effectively verifying the reliability of the previous conclusions. Accordingly, it is recommended that if Shenhua project keeps the current annual target injection flow rate invariant, would better be controlled within 5~7 MPa in the subsequent injection process.
Although this work only studies the relationship between and of the injection well, the above theory and method can be generalized to study the production well, which can obtain the similar relationship between wellbore pressure and production flow, and to further explore the effects of abovementioned influential factors on the relationship.
The constraint conditions of this paper do not involve the caprocks, so when including the caprocks, the entry pressure may have a strong impact on the maximum allowable pressure of reservoirs. Then the constraint conditions may change. Therefore, further study needs to address this occasion.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was sponsored by the National Natural Science Foundation of China (Grant no. 41672252).
References
 ChinaU.S. Joint Announcement on climate change, Beijing, China, 2014, http://www.china.org.cn/chinese/201412/09/content_34268965.htm.
 S. Bachu, “Screening and ranking of sedimentary basins for sequestration of CO_{2} in geological media in response to climate change,” Environmental Geology, vol. 44, no. 3, pp. 277–289, 2003. View at: Publisher Site  Google Scholar
 IPCC, “Carbon Dioxide Capture and Storage,” in Intergovernmental Panel on Climate Change, WMO & UNEP, Cambridge University Press, New York, NY, USA, 2005. View at: Google Scholar
 S. Sun, “Geological problems of CO_{2} underground storage and its significance on the mitigation of climate change,” China Basic Science, no. 3, pp. 17–22, 2006. View at: Google Scholar
 M. Zhang and M. J. McSaveney, “Is air pollution causing landslides in China?” Earth and Planetary Science Letters, vol. 481, pp. 284–289, 2018. View at: Publisher Site  Google Scholar
 S. Bachu, “CO_{2} storage in geological media: role, means, status and barriers to deployment,” Progress in Energy and Combustion Science, vol. 34, no. 2, pp. 254–273, 2008. View at: Publisher Site  Google Scholar
 X. Li and Z. Fang, “Status quo of connection technologies of CO_{2} geological storage in China,” Rock and Soil Mechanics, vol. 28, no. 10, pp. 2229–2239, 2007 (Chinese). View at: Google Scholar
 B. Bai, X. Li, M. Liu, L. Shi, and Q. Li, “A fast explicit finite difference method for determination of wellhead injection pressure,” Journal of Central South University, vol. 19, no. 11, pp. 3266–3272, 2012. View at: Publisher Site  Google Scholar
 M. Liu, B. Bai, and X. Li, “A unified formula for determination of wellhead pressure and bottomhole pressure,” Energy Procedia, vol. 37, pp. 3291–3298, 2014. View at: Publisher Site  Google Scholar
 J. Rutqvist, J. Birkholzer, F. Cappa, and C.F. Tsang, “Estimating maximum sustainable injection pressure during geological sequestration of CO_{2} using coupled fluid flow and geomechanical faultslip analysis,” Energy Conversion and Management, vol. 48, no. 6, pp. 1798–1807, 2007. View at: Publisher Site  Google Scholar
 S. Mukhopadhyay, S.Y. Yang, and H.D. Yeh, “Pressure Buildup During Supercritical Carbon Dioxide Injection From a Partially Penetrating Borehole into Gas Reservoirs,” Transport in Porous Media, vol. 91, no. 3, pp. 889–911, 2012. View at: Publisher Site  Google Scholar
 A. Liebscher, F. Möller, A. Bannach et al., “Injection operation and operational pressuretemperature monitoring at the CO_{2} storage pilot site Ketzin, GermanyDesign, results, recommendations,” International Journal of Greenhouse Gas Control, vol. 15, pp. 163–173, 2013. View at: Publisher Site  Google Scholar
 B. Bai, X. Li, H. Wu, Y. Wang, and M. Liu, “A methodology for designing maximum allowable wellhead pressure for CO_{2} injection: application to the Shenhua CCS demonstration project,” China. Greenhouse Gases Science Technology, vol. 7, no. 1, pp. 158–181, 2017. View at: Publisher Site  Google Scholar
 J. Rutqvist, “The Geomechanics of CO_{2} Storage in Deep Sedimentary Formations,” Geotechnical and Geological Engineering, vol. 30, no. 3, pp. 525–551, 2012. View at: Publisher Site  Google Scholar
 H. Wu, B. Bai, X. Li, M. Liu, and Y. He, “An explicit finite difference model for prediction of wellbore pressure and temperature distribution in CO_{2} geological sequestration,” Greenhouse Gases: Science and Technology, vol. 7, no. 2, pp. 353–369, 2017. View at: Publisher Site  Google Scholar
 J. J. Carroll and D. W. Lui, “Density, phase behavior keys to acid gas injection,” Oil & Gas Journal, vol. 95, no. 25, pp. 63–72, 1997. View at: Google Scholar
 E. T. Hashim and I. E. Maloka, “Static bottomhole pressures in wells,” Petroleum Science and Technology, vol. 24, no. 1, pp. 113–116, 2006. View at: Publisher Site  Google Scholar
 J. E. Streit and R. R. Hillis, “Estimating fault stability and sustainable fluid pressures for underground storage of CO_{2} in porous rock,” Energy, vol. 29, no. 910, pp. 1445–1456, 2004. View at: Publisher Site  Google Scholar
 C. Gapillou, S. Thibeau, G. Mouronval, and M. Lescanne, “Building a geocellular model of the sedimentary column at rousse CO_{2}, geological storage site (aquitaine, france) as a tool to evaluate a theorical maximum injection pressure,” Energy Procedia, vol. 1, no. 1, pp. 2937–2944, 2009. View at: Google Scholar
 H. Wu, B. Bai, M. Liu, X. Li, and L. Wang, “The equivalent density method to estimate the wellbore pressure of CO_{2} well,” Special Oil & Gas Reservoirs, vol. 22, no. 3, pp. 114–117, 2015. View at: Google Scholar
 J. J. Carroll, Acid Gas Injection and Carbon Dioxide Sequestration, LLC, Salem, Massachusetts: WileyScrivener press, Hoboken, NJ, USA, 2010.
 M. Zhang and M. J. McSaveney, “Rock avalanche deposits store quantitative evidence on internal shear during runout,” Geophysical Research Letters, vol. 44, no. 17, pp. 8814–8821, 2017. View at: Publisher Site  Google Scholar
 X. Li, K. Hitoshi, and O. Takashi, “CO_{2} aquifer storage and the related rock mechanics issues,” Chinese Journal of Rock Mechanics & Engineering, vol. 22, no. 6, pp. 989–994, 2003. View at: Google Scholar
 J. M. Nordbotten, M. A. Celia, and S. Bachu, “Injection and storage of CO_{2} in deep saline aquifers: Analytical solution for CO_{2} plume evolution during injection,” Transport in Porous Media, vol. 58, no. 3, pp. 339–360, 2005. View at: Publisher Site  Google Scholar
 A. Cihan, J. T. Birkholzer, and Q. Zhou, “Pressure Buildup and Brine Migration During CO2 Storage in Multilayered Aquifers,” Groundwater, vol. 51, no. 2, pp. 252–267, 2013. View at: Publisher Site  Google Scholar
 C.F. Tsang, J. Birkholzer, and J. Rutqvist, “A comparative review of hydrologic issues involved in geologic storage of CO_{2} and injection disposal of liquid waste,” Environmental Geology, vol. 54, no. 8, pp. 1723–1737, 2008. View at: Publisher Site  Google Scholar
 H. Wu, B. Bai, X. Li, S. Gao, M. Liu, and L. Wang, “An explicit integral solution for pressure buildup during CO_{2} injection into infinite saline aquifers,” Greenhouse Gases: Science and Technology, vol. 6, no. 5, pp. 633–647, 2016. View at: Publisher Site  Google Scholar
 H. Sidiq, R. Amin, and T. Kennaird, “The study of relative permeability and residual gas saturation at high pressures and high temperatures,” Advances in GeoEnergy Research, vol. 1, no. 1, pp. 64–68, 2017. View at: Publisher Site  Google Scholar
 S. Wang, X. Han, Y. Dong, and H. Shi, “Mechanisms of reservoir pore/throat characteristics evolution during longterm waterflooding,” Advances in GeoEnergy Research, vol. 1, no. 3, pp. 148–157, 2017. View at: Publisher Site  Google Scholar
 L. Shi, B. Bai, H. Wu, and X. Li, “Evaluating reservoir risks and their influencing factors during CO2 injection into multilayered reservoirs,” Geofluids, vol. 2017, Article ID 6059142, 2017. View at: Publisher Site  Google Scholar
 H. Liu, Z. Hou, P. Were, Y. Gou, and X. Sun, “Simulation of CO_{2} plume movement in multilayered saline formations through multilayer injection technology in the Ordos Basin, China,” Environmental Earth Sciences, vol. 71, no. 10, pp. 4447–4462, 2014. View at: Publisher Site  Google Scholar
 K. Pruess, C. Oldenburg, and G. Moridis, “TOUGH2 User's Guide Version 2,” Tech. Rep. LBNL43134, Lawrence Berkeley National Laboratory, Berkeley, Calif, USA, 1999. View at: Publisher Site  Google Scholar
 B. R. Crawford, D. R. Faulkner, and E. H. Rutter, “Strength, porosity, and permeability development during hydrostatic and shear loading of synthetic quartzclay fault gouge,” Journal of Geophysical Research: Solid Earth, vol. 113, no. 3, Article ID B03207, 2008. View at: Publisher Site  Google Scholar
 J. Cai, B. Yu, M. Zou, and M. Mei, “Fractal analysis of invasion depth of extraneous fluids in porous media,” Chemical Engineering Science, vol. 65, no. 18, pp. 5178–5186, 2010. View at: Publisher Site  Google Scholar
 Z. Zhang, J. Cai, F. Chen, H. Li, W. Zhang, and W. Qi, “Progress in enhancement of CO_{2} absorption by nanofluids: A mini review of mechanisms and current status,” Journal of Renewable Energy, vol. 118, pp. 527–535, 2018. View at: Publisher Site  Google Scholar
 J. Rutqvist and C.F. Tsang, “A study of caprock hydromechanical changes associated with CO_{2} injection into a brine formation,” Environmental Geology, vol. 42, no. 23, pp. 296–305, 2002. View at: Publisher Site  Google Scholar
 J. Xie, K. Zhang, L. Hu, Y. Wang, and M. Chen, “Understanding the carbon dioxide sequestration in lowpermeability saline aquifers in the Ordos Basin with numerical simulations,” Greenhouse Gases: Science and Technology, vol. 5, no. 5, pp. 558–576, 2015. View at: Publisher Site  Google Scholar
 J. Xie, K. Zhang, L. Hu, P. Pavelic, Y. Wang, and M. Chen, “Fieldbased simulation of a demonstration site for carbon dioxide sequestration in lowpermeability saline aquifers in the Ordos Basin, China,” Hydrogeology Journal, vol. 23, no. 7, pp. 1465–1480, 2015. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2018 Bing Bai et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.