Multi-injection parameter collaborative optimization method for efficient coal-based carbon sequestration and yield increase

Abstract

The invention discloses a multi-injection parameter collaborative optimization method for efficient coal-based carbon sequestration and yield increase, and belongs to the technical field of natural gas extraction. The method comprises the following steps: measuring, collecting and fitting the physical property parameters of the coal seam which cannot be mined under different reservoir depths; a mixture gas injection yield increasing and carbon sequestration scheme based on physical property parameters and a ComsolMultiphysics numerical calculation model are constructed, and a database of numerical solutions of breakthrough time, CH4 accumulative recovery amount, N2 accumulative injection amount, CO2 accumulative injection amount and the like corresponding to different injection parameters is formed; designing an economic objective function comprising factors such as gas supply and compression cost, carbon sequestration and methane output income, injection pressure and the like; and based on the database and the economic objective function, in combination with a BP neural network and a genetic algorithm, determining an optimal injection parameter combination, influence factor sensitivity and weight under different reservoir parameters. According to the method, efficient coal-based carbon sequestration can be achieved, the recovery efficiency is increased, and the carbon sequestration and yield increasing cost and energy consumption are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of natural gas exploitation, and in particular to a method for collaboratively optimizing multiple injection parameters for efficient coal-based carbon fixation and production increase. Background Art

[0002] Coalbed methane (CBM) is the main hazard factor that causes a series of disasters such as gas explosions, coal and gas outbursts, and is also a source of environmental pollution. The greenhouse effect of its emissions in the air is 25 times that of carbon dioxide. However, this unconventional natural gas can be used as a non-renewable clean energy. The conventional CBM recovery rate is poor. In order to improve mining efficiency, related technologies such as fracturing permeability enhancement and heat injection production increase have been rapidly developed.

[0003] Unmineable coal seams have the characteristics of difficult reservoir transformation and CH 4 The low production and other characteristics usually make it impossible to obtain economic benefits through conventional mining technology. 2 When the coal seam is exposed to the sun, it is easy to cause the pore structure of the coal seam to expand and the permeability to decrease, thus affecting the desorption and output of methane. 2 Injection can not only prevent the permeability of coal seams from decreasing, but also displace methane through pressure difference, increase the desorption rate of methane, and thus increase coalbed methane production. 2 / N 2 Mixed gas injection technology has unique advantages, not only can CO 2 Sealing in coal seams can reduce greenhouse gas emissions and increase coalbed methane production. 2 With N 2 The mixed injection forms a relatively optimized coal-based production increase and carbon fixation pathway, achieving the dual goals of resource and environmental benefits.

[0004] The current means of increasing production are a series of conceptual carbon fixation and production increase methods proposed with the core goal of improving recovery rate or increasing carbon storage, such as the publication number "CN115539130A" and the name "A method for enhancing coalbed methane mining and CO 2 The Chinese patent for the method of sealing 2 The foam pulsation fracturing method is used to fractur e the unmineable coal seams. 2 The storage provides a wide adsorption space, and then the CO 2 Injection to displace coalbed methane to enhance the extraction process to achieve CO 2The method only empirically gives relevant technical parameters such as the corresponding fracturing parameters, the distance between injection wells and production wells, and does not fully consider the matching between the actual reservoir conditions and the technical parameters. The additional fracturing will increase the cost. The Chinese patent with the publication number "CN118167250A" and the name "A method for extracting gas from high-gas low-permeability coal seams" uses gas injection to promote the extraction of coal seam CH 4 Optimize key parameters, establish key control parameters, and select extraction methods based on sensitivity analysis. 2 Coal seam CH 4 The design of the extraction and injection scheme provides a basis, but this method focuses more on coal seam gas recovery and does not consider the impact of costs.

[0005] The coordinated optimization of multiple injection parameters refers to the realization of a cooperative balance between multiple injection parameters and reservoir physical parameters in the process of coal-based carbon fixation and production increase by adjusting the ratio of injected gas mixture, injection pressure, temperature, well spacing, etc., so as to achieve a win-win situation of carbon fixation and production increase. Therefore, how to adapt the optimal injection parameters according to the physical parameters to achieve a balance between cost-efficiency and recovery rate and increase carbon storage is a problem that needs to be solved. Summary of the invention

[0006] In order to overcome the defects in the prior art, the present invention proposes a multi-injection parameter collaborative optimization method for efficient coal-based carbon fixation and production increase. Through the collaborative optimization of injection parameters, efficient coal-based carbon sequestration and increased recovery rate are achieved, and costs and energy consumption are reduced.

[0007] The technical solution adopted by the present invention is: a method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase, comprising the following steps:

[0008] Step S1, measuring, collecting and fitting the physical property parameters of unmineable coal seams at different reservoir depths H; the physical property parameters include reservoir temperature T 0 , reservoir pressure p 0 , reservoir permeability k 0 and vertical ground stress ;

[0009] Step S2, establishing a conventional extraction numerical model that meets the on-site reservoir parameters and production conditions, and comparing it with the on-site gas production rate to verify the reliability of the model;

[0010] Step S3: After the model is verified, a corresponding mixed gas injection plan is formulated according to the reservoir characteristics of different reservoir depths H, where the injection parameters include the injection pressure P inj , mixed gas ratio c, injection temperature T inj Well spacing ;

[0011] Step S4, establishing a single variable control numerical calculation model for injection parameters according to the corresponding mixed gas injection scheme;

[0012] Step S5: Count the numerical calculation results under each injection scheme, including the breakthrough time , CH 4 Cumulative harvest volume 、N 2 Cumulative injection volume , CO 2 Cumulative injection volume and CO 2 Cumulative harvest volume , and then build a database;

[0013] Step S6: Design a breakthrough time The economic objective function , the economic objective function Including CO2 supply costs , Nitrogen supply cost , Mixture compression cost Well construction cost , injection cost , CH 4 income , Carbon storage benefits and gas separation costs ; The objective function formula is as follows:

[0014]

[0015] In the formula, Indicates the amount of gas, m 3 , the subscript p represents production, inj represents injection, h represents reservoir thickness, and m represents meters; represents the breakthrough time, d represents days; Represents the total amount of produced gas, m 3 ;

[0016] Step S7: Taking the reservoir depth H and the injection parameters under the corresponding injection scheme as independent variables and the numerical calculation result data under each injection scheme as dependent variables, construct and verify the BP neural network; write a genetic algorithm to seek the optimal economic objective function , and determine the optimal injection parameter combination and the influence weight of each factor under different reservoir depth H conditions.

[0017] As a further improvement of the present invention, in step S1, the physical property parameters of the unmineable coal seams at different reservoir depths H are measured, collected and fitted, specifically:

[0018] Step S11, establishing a coordinate system with reservoir depth as the horizontal coordinate and physical property parameters as the vertical coordinate;

[0019] Step S12: plot the tested and collected physical property parameters in a coordinate system and perform parameter fitting.

[0020] As a further improvement of the present invention, in step S2, in establishing a conventional extraction numerical model that meets the field reservoir parameters and production conditions, the production conditions include the production well pressure and temperature, the production well pressure is 0.15Mpa, the production temperature is the initial reservoir temperature, and the average error rate between the numerical solution of the gas production rate and the actual field value is no more than 15% as the model verification threshold.

[0021] As a further improvement of the present invention, in step S3, the injected mixed gas is CO 2 and N 2 , the injection scheme is specifically as follows:

[0022] Step S31: The reservoir depth H is 300-1200m, and the step length is increased to a. 1 ;

[0023] Step S32: injection pressure P inj Reservoir pressure P 0 1.1~2.0 times of the original value, increase the step size to a 2 ;

[0024] Step S33, CO 2 、N 2 CO in mixed gas 2 The proportion c, c value is between 0.0 and 1.0, and the increase step is a 3 ;

[0025] Step S34, injection temperature T inj Greater than or equal to reservoir temperature T 0 , the difference between the two does not exceed 30℃ of reservoir temperature, and the step size is a 4 ;

[0026] Step S35: Well spacing Between 150 and 500 m, increase the step length to a 5 ;

[0027] As a further improvement of the present invention, increase the step length a 1 、a 2 、a 3 、a 4 、a 5 The values ​​are 50, 0.1, 0.1, 5, and 50 respectively.

[0028] As a further improvement of the present invention, the breakthrough time is set in the numerical calculation model of step S4 , specifically:

[0029] Step S41: To ensure that the production well has a sufficient concentration of CH 4 , with the gas production rate of the mixed gas and CH 4 The breakthrough time is when the yield ratio is equal to 50% ;

[0030] Step S42: When production > 7000d or CH 4 Gas production rate <1000m 3 / d, the corresponding production time is mandatory to be set as the breakthrough time ,d stands for day.

[0031] As a further improvement of the present invention, in step S6, the injection cost Including injection pressure costs and injection temperature cost , the specific formula is as follows:

[0032] =10+2(T inj -T 0 )($ / d)

[0033] =10+50(pp 0 )($ / d)

[0034] Where $ / d is US dollars per day.

[0035] As a further improvement of the present invention, in step S7, the BP neural network is specifically: the BP neural network includes 10 hidden layers, the number of iterations is 100, the iteration is terminated when the mean square error is less than 0.1, the model training group is 80% of the number of all injection schemes, and the prediction group is 20% of the number of all injection schemes.

[0036] The beneficial effects of the present invention are:

[0037] (1) Based on the synergistic effect of multiple injection parameters, the present invention takes into account the differences in the influence of different injection parameters on the basis of reservoir parameters, and realizes the cooperative balance among the injection parameters under the constraints of the economic objective function including different categories of cost, benefit and other factors, thereby ensuring the maximization of the benefits of increased production and coal-based carbon sequestration, further reducing the energy consumption of injection and production processes, and saving energy and protecting the environment;

[0038] (2) The present invention has strong applicability and scalability. When designing and planning reservoir exploitation, the injection parameters of different regions, including injection pressure, mixed gas ratio, injection temperature, and well spacing, can be zoned according to the non-uniformity of reservoir parameters. Injection parameters can be collaboratively optimized during reservoir exploitation design and planning, and can be optimized once or multiple times during production according to changes in reservoir parameters, such as once every 500 or 1,000 days of production. At the same time, the economic objective function The cost parameters involved are not fixed and may change with market fluctuations, such as different CH 4 Price, CO 2 Changes in factors such as storage efficiency will lead to different optimal injection plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] Figure 1 It is a process flow chart of the multi-injection parameter collaborative optimization method for efficient coal-based carbon fixation and production increase of the present invention;

[0041] Figure 2 is a diagram showing the relationship between reservoir physical property parameters and burial depth in the embodiment;

[0042] Figure 3 Figure 1 is a numerical calculation model diagram of the mixed gas injection to increase carbon fixation in the embodiment;

[0043] Figure 4 Figure II is a numerical calculation model diagram of the mixed gas injection to increase carbon fixation in the embodiment;

[0044] Figure 5 This is a model verification diagram for gas production rate at different well spacings in the embodiment;

[0045] Figure 6 It is the BP neural network partial prediction-actual verification diagram;

[0046] Figure 7 The parameter combination diagram injected under the optimal objective function in the embodiment;

[0047] Figure 8 For different CH in the embodiment 4 Price injection parameter combination chart. DETAILED DESCRIPTION

[0048] The multi-injection parameter collaborative optimization method for efficient coal-based carbon fixation and production increase of the present invention is described in detail below with a specific embodiment.

[0049] Based on the unmineable coal seams with a depth of about 800m in my country, the specific physical properties are based on the fitting parameters. Ordinary coalbed methane extraction in the area generally adopts vertical gas wells with a bottom hole pressure of 0.15MPa and a well spacing of 300~500m.

[0050] like Figure 1 As shown, the multi-injection parameter collaborative optimization method for efficient coal-based carbon fixation and production increase of the present invention has the following specific steps:

[0051] Step S1, measuring, collecting and fitting the basic physical property parameters of unconventional reservoirs at different reservoir depths H, mainly including: reservoir temperature T 0 , reservoir pressure p 0 , reservoir permeability k 0 and vertical ground stress At the same time, a coordinate system is established to draw a physical property parameter image, with the reservoir depth as the horizontal coordinate and the main physical property parameters as the vertical coordinate, such as Figure 2 shown. Figure 2 In the figure, (a) is the relationship between temperature and burial depth, (b) is the relationship between permeability and burial depth, (c) is the relationship between pressure and burial depth, and (d) is the relationship between vertical stress and burial depth. The main physical property parameters at different reservoir depths H are fitted.

[0052] In order to ensure the simplicity of the invention, linear fitting is used, where:

[0053] The relationship between reservoir temperature and burial depth is: y = 0.0257x + 13.522;

[0054] The relationship between reservoir pressure and burial depth is: y = 0.0066x + 0.262;

[0055] The relationship between reservoir permeability and burial depth is y = -0.1173x + 104.04;

[0056] Relationship between vertical stress and burial depth: y = 0.0208x + 3.7876.

[0057] The basic physical parameters corresponding to the burial depth of 800m are shown in Table 1.

[0058]

[0059] Formulate corresponding injection schemes according to reservoir characteristics. Injection parameters include injection pressure p, mixture ratio c, injection temperature T inj , well spacing and reservoir depth H, the injection pressure p is respectively the reservoir pressure p 0 1.1~2.0 times, with a spacing of 0.1; CO 2 、N 2 CO in the mixture 2The proportion c, c ranges from 0.0 to 1.0, with a step interval of 0.1; the injection temperature T inj Should be greater than or equal to the reservoir temperature T 0 , the difference between the two does not exceed 30℃ of reservoir temperature, the step length is 5; the well spacing , that is, the projection distance between the injection well and the production well is between 150 and 500 m, with a step size of 50, and a total of 6160 numerical calculation schemes.

[0060] The injection scheme parameters are shown in Table 2.

[0061]

[0062] Table 2 shows the parameters corresponding to the injection scheme, which adopts the single variable control method, that is, there are 10 levels of injection pressure, 11 levels of mixture ratio, 7 levels of injection temperature, 8 levels of well spacing, and 1 level of reservoir burial depth in the embodiment. Therefore, the total number of numerical model schemes is 10×11×7×8×1=6160.

[0063] Step S2: Use Comsol Multiphysics software to establish a conventional extraction numerical model that meets actual conditions, such as Figure 3 and Figure 4 As shown in the figure, for the case without injection wells, the production well pressure is 0.15 MPa, the production time is 6000 days, and the error rate between the numerical solution of the recovery rate and the actual solution is no more than 15%. Figure 5 shown.

[0064] Step S3: After the model is verified, a corresponding mixed gas injection plan is formulated according to the reservoir characteristics at different reservoir depths H. The injection parameters include the injection pressure P inj , mixed gas ratio c, injection temperature T inj和 Well spacing .

[0065] Step S4: establishing a single variable control numerical calculation model for injection parameters according to the corresponding mixed gas injection scheme.

[0066] To ensure that there is sufficient CH concentration on the production well side 4 , with N 2 +CO 2 Gas production rate and CH 4 The yield ratio is equal to 50% as the breakthrough time, while the production is >7000d or CH 4 Gas production rate <1000m 3 / d Forced well closure, d represents days.

[0067] Step S5: Statistically summarize the breakthrough time under different injection schemes , CH 4 Cumulative harvest volume 、N 2 Cumulative injection volume , CO 2 Cumulative injection volume , CO 2 Cumulative harvest volume .

[0068] Step S6: Design an economic objective function to achieve breakthrough time , including the cost of CO2 supply , Nitrogen supply cost , Mixture compression cost Well construction cost , injection cost 、CH4 income , Carbon storage benefits , Gas separation costs Etc. The specific formula is:

[0069]

[0070] In the formula, Indicates the amount of gas, m 3 , the subscript represents production, inj represents injection, h represents reservoir thickness, m, represents the breakthrough time, and d represents days.

[0071] Step S7, injection pressure p, mixture ratio c, injection temperature T inj , well spacing , reservoir depth H as independent variable, and objective function The variables involved are dependent variables, including breakthrough time , CH 4 Cumulative harvest volume 、N 2 Cumulative injection volume , CO 2 Cumulative injection volume , CO 2 Cumulative harvest volume , construct a BP neural network. The BP neural network contains 10 hidden layers, the number of iterations is 100, and the iteration is terminated when the mean square error is less than 0.1. The model training group is 80% of the total number of solutions, that is, 4928, and the prediction group is 20% of the total number of solutions, that is, 1232. The variance of the predicted value and the actual value of each dependent variable is used to verify the effectiveness of the model. Due to the large number of prediction groups, some prediction results are shown in Figure 6 As shown, R 2 It is above 0.96, indicating the effectiveness of the model. Figure 6In the figure, (a) is the breakthrough time corresponding to some samples, (b) is the cumulative CH4 production corresponding to some samples, and (c) is the cumulative CO 2 Injection volume, (d) is the cumulative N corresponding to some samples 2 Injection volume.

[0072] On the basis of verifying the correctness of BP neural network, the optimal objective function value is sought through Matlab's own module and legacy algorithm. The genetic algorithm learning rate is 1e-3, the number of iterations is 500, and the calculation tolerance is 1e -6 。 Collect and organize the economic objective function proposed in step S6 The cost parameters involved are shown in Table 3.

[0073]

[0074] To ensure the correctness of the optimal objective function value, the optimization algorithm of each solution is executed 1000 times, and the economic objective function is taken Maximum value, CH 4 , CO 2 The price is 0.5$ / m 3 1$ / m 3 , the economic objective function obtained The maximum value is 26.28$ / m 3 .

[0075] Economic objective function determined by genetic algorithm The maximum value is obtained by using the back propagation BP neural network to predict the optimal injection parameter combination and the weight of influencing factors under different reservoir depths H. The injection pressure, well spacing, temperature, and mixture ratio are 10.0 MPa, 175.6, 34 °C, and 0.97, respectively, and the weights are 0.30, 0.20, 0.15, and 0.45, respectively. The corresponding recovery factor is 79.86%. Figure 7 As shown, it can provide a basis for the selection of main control parameters when optimizing injection parameters.

[0076] At the same time, the model can also analyze different CH 4 The impact of price on the optimal combination, the carbon storage price is fixed at 0.4$ / m 3 , CH 4 The prices are 0.2, 0.5, 1$ / m respectively 3 , different CH 4The optimal injection parameter combination corresponding to the price, injection pressure, well spacing, temperature, and mixture ratio are 10MPa-219.2m-309.5K or 36.4℃-100%, 10MPa-187.5m-309.7K or 36.5℃-100%, 10MPa-174.8m-305.5K or 32.4℃-83.8%, respectively. 4 The recovery factors are 55.94%, 77.76%, and 80.00%, respectively, and the objective functions are 3.36, 8.74, and 16.63$ / m 3 ,like Figure 8 shown.

[0077] Under different injection parameter combinations, CO 2 It is possible to reach a supercritical state, and the synergistic optimization of injection parameters under different phase conditions can be further analyzed.

[0078] The multi-injection parameter collaborative optimization method for efficient coal-based carbon fixation and production increase of the present invention significantly improves the recovery rate and carbon fixation rate, reduces the occurrence of very high fuel gas, reduces greenhouse gas emissions, and achieves high efficiency, energy saving, and environmental protection. At the same time, the economic objective function The cost parameters involved are not fixed and can change with market fluctuations. In addition, the injection parameters can be optimized before mining or adjusted during mining. The specific method remains unchanged, only the initial physical property parameters are changed, which reflects the reliability and scalability of this method.

[0079] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited thereto. Various changes that can be made within the knowledge scope of technicians in the relevant technical field without departing from the spirit of the present invention are all within the protection scope of the claims of the present invention.

Claims

1. A method for the coordinated optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase, characterized in that: The following steps are involved: Step S1, measuring, collecting and fitting the physical property parameters of the unmineable coal seam at different reservoir depths H; the physical property parameters include reservoir temperature T0, reservoir pressure p0, reservoir permeability k0 and vertical ground stress ; Step S2, establishing a conventional extraction numerical model that meets the on-site reservoir parameters and production conditions, and comparing it with the on-site gas production rate to verify the reliability of the model; Step S3: After the model is verified, a corresponding mixed gas injection plan is formulated according to the reservoir characteristics of different reservoir depths H, where the injection parameters include the injection pressure P inj , mixed gas ratio c, injection temperature T inj Well spacing ; Step S4, establishing a single variable control numerical calculation model for injection parameters according to the corresponding mixed gas injection scheme; Step S5: Count the numerical calculation results under each injection scheme, including the breakthrough time , Cumulative CH4 recovery , N2 cumulative injection amount , Cumulative CO2 injection volume and cumulative CO2 recovery , and then build a database; Step S6: Design a breakthrough time The economic objective function , the economic objective function Including CO2 supply costs , Nitrogen supply cost , Mixture compression cost Well construction cost , injection cost 、CH4 income , Carbon storage benefits and gas separation costs ; The objective function formula is as follows: ， In the formula, Indicates the amount of gas, m 3 , the subscript p represents production, inj represents injection, h represents reservoir thickness, and m represents meters; represents the breakthrough time, d represents days; Represents the total amount of produced gas, m 3 ; Step S7: Taking the reservoir depth H and the injection parameters under the corresponding injection scheme as independent variables and the numerical calculation result data under each injection scheme as dependent variables, construct and verify the BP neural network; write a genetic algorithm to seek the optimal economic objective function , and determine the optimal injection parameter combination and the influence weight of each factor under different reservoir depth H conditions.

2. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: In step S1, the physical property parameters of unmineable coal seams at different reservoir depths H are measured, collected and fitted, specifically: Step S11, establishing a coordinate system with reservoir depth as the horizontal coordinate and physical property parameters as the vertical coordinate; Step S12: plot the tested and collected physical property parameters in a coordinate system and perform parameter fitting.

3. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: In step S2, a conventional extraction numerical model that meets the field reservoir parameters and production conditions is established. The production conditions include the production well pressure and temperature. The production well pressure is 0.15 MPa, and the production temperature is the initial reservoir temperature. The average error rate between the numerical solution of the gas production rate and the actual field value is no more than 15% as the model verification threshold.

4. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: In step S3, the injected mixed gas is CO2 and N2, and the injection scheme is specifically as follows: Step S31, the reservoir depth H is 300-1200m, and the increase step is a1; Step S32: injection pressure P inj They are 1.1~2.0 times of the reservoir pressure P0, respectively, and the increase step is a2; Step S33, the proportion of CO2 in the CO2 / N2 mixture c, where c is between 0.0 and 1.0, and the increment step is a3; Step S34, injection temperature T inj is greater than or equal to the reservoir temperature T0, the difference between the two does not exceed 30°C of the reservoir temperature, and the increase step is a4; Step S35: Well spacing Between 150 and 500 m, increase the step length to a5.

5. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 4 is characterized in that: The values ​​of a1, a2, a3, a4, and a5 in the increasing step are 50, 0.1, 0.1, 5, and 50 respectively.

6. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: Set the breakthrough time in the numerical calculation model of step S4 , specifically: Step S41: To ensure that there is sufficient concentration of CH4 on the production well side, the breakthrough time is set at a ratio of the mixed gas production rate to the CH4 yield of 50%. ; Step S42: When production > 7000d or CH4 production rate < 1000m 3 / d, the corresponding production time is mandatory to be set as the breakthrough time ,d stands for day.

7. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: In step S6, the cost is injected Including injection pressure costs and injection temperature cost , the specific formula is as follows: =10+2(T inj -T0)($ / d) =10+50(p-p0)($ / d) Where $ / d is US dollars per day.

8. The method for collaborative optimization of multiple injection parameters for efficient coal-based carbon fixation and production increase according to claim 1 is characterized in that: In step S7, the BP neural network is specifically as follows: the BP neural network includes 10 hidden layers, the number of iterations is 100, the iteration is terminated when the mean square error is less than 0.1, the model training group is 80% of all injection schemes, and the prediction group is 20% of all injection schemes.

Images