Water injection profile inversion model construction method, water absorption capacity determination method and device

Abstract

This article relates to the field of oil and gas field development, and in particular to a method for constructing a water injection profile inversion model, a method for determining water absorption capacity, and a device. It includes obtaining historical initial parameters; using a pre-constructed temperature profile model, and using the parameters of the historical initial parameters that have a large correlation with the wellbore temperature as constraint parameters; the output result obtained after inputting the constraint parameters into the temperature profile model is determined as a training sample data set, and the training sample data set includes historical wellbore temperature, historical wellbore pressure, and historical wellbore flow; using the training sample data set to train the initial machine learning model, and using the trained machine learning model as a water injection profile inversion model, and the output of the water injection profile inversion model is the flow distribution of the wellbore. This scheme quantitatively explains the flow rate of the water injection well along the wellbore and the water absorption of each perforation section, quantifies the degree of utilization of the target layer by perforation in heterogeneous reservoirs, and provides data support for subsequent measures such as water injection well profile adjustment, pressure reduction and injection increase.

Description

Technical Field

[0001] This article relates to the field of oil and gas field development, and in particular to a method for constructing a water injection profile inversion model, a method for determining water absorption capacity, and an apparatus. Background Art

[0002] Waterflooding has been widely used in the development of low-permeability reservoirs and has a good oil displacement effect. Due to the heterogeneity of the reservoir, the water absorption capacity of the formation varies in different perforation sections of the injection well, resulting in uneven oil and water propagation and different sweep efficiency in waterflooding operations.

[0003] In oilfield operations, methods such as isotope tracer profile logging, flowmeter logging, and pulsed neutron oxygen activation logging are commonly used to understand the seepage of injected fluids in the formation near the wellbore. However, these testing methods vary in their applicability to different downhole water injection structures, are difficult to implement, have long testing cycles, and some radioactive elements pose certain pollution and dangers.

[0004] Existing technologies using flowmeter logging, while providing flow rate information, cannot provide continuous real-time wellbore flow distribution. Current reservoir temperature monitoring methods utilize distributed fiber optic temperature sensors (DTS) to rapidly record the temperature profile across the entire wellbore trajectory, analyzing transient temperature information in the wellbore during water injection and after shut-in for reservoir evaluation and water injection monitoring, thereby optimizing water injection design. However, quantitative research on temperature data interpretation is lacking. Studies on DTS temperature data are mostly qualitative, and theoretical interpretation models for temperature are still under development. The sheer volume of data measured by sensors in the continuous time and spatial domains is enormous, and the analysis and processing are complex. Temperature interpretation often relies on manually selected data from a specific moment, introducing a degree of randomness and uncertainty.

[0005] Currently, machine learning algorithms are rarely used in the field of water injection wells. The inversion of water injection profiles mostly adopts conventional numerical simulation methods and conventional inversion methods. These methods have inaccurate calculation results, long testing cycles, and long inversion times.

[0006] To address the problems of long testing cycles in current logging methods, large amounts of DTS temperature monitoring data in water injection wells, and long inversion time, a method for constructing water injection profile inversion models and a method for determining water absorption capacity are needed. Summary of the Invention

[0007] To address the problems of the prior art, this embodiment provides a method for constructing a water injection profile inversion model, a method for determining water absorption capacity, and an apparatus.

[0008] This embodiment provides a method for constructing a water injection profile inversion model. The method includes: obtaining historical initial parameters; using a pre-constructed temperature profile model, wherein parameters in the historical initial parameters that are highly correlated with wellbore temperature are used as constraint parameters; determining the output results obtained after inputting the constraint parameters into the temperature profile model as a training sample dataset, wherein the training sample dataset includes historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate; training an initial machine learning model using the training sample dataset, and using the trained machine learning model as the water injection profile inversion model, wherein the output of the water injection profile inversion model is the flow rate distribution of the wellbore.

[0009] According to one aspect of the embodiments herein, a temperature profile model is used to determine parameters among the historical initial parameters that are highly correlated with wellbore temperature, which are then used as constraint parameters. This includes: inputting each historical initial parameter into the temperature profile model and quantitatively calculating the temperature change at a certain location in the wellbore corresponding to each historical initial parameter; and determining the initial parameters whose temperature change is greater than a preset value as constraint parameters. The constraint parameters include one or more of the following: perforation section thickness, perforation section permeability, injection volume, injection temperature, and perforation section interval distance.

[0010] According to one aspect of the embodiments herein, determining the output results obtained after inputting the constraint parameters into the temperature profile model as a training sample dataset includes: setting different values ​​for each constraint parameter to form multiple constraint parameter groups; inputting the constraint parameter groups into the temperature profile model to calculate multiple sets of historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate that vary with well depth; and using the historical wellbore temperature, historical wellbore pressure, historical wellbore flow rate, and well depth data as a training sample dataset.

[0011] According to one aspect of the embodiments herein, the temperature profile model is determined by the following steps: obtaining initial parameters at a certain time in the past, the initial parameters including reservoir parameters, wellbore parameters, fluid parameters and operating regime data; establishing a wellbore model and a reservoir model respectively based on the initial parameters, the principle of conservation of mass, the principle of conservation of momentum and the principle of conservation of energy, the wellbore model and the reservoir model are coupled to form the temperature profile model.

[0012] According to one aspect of the embodiments herein, training an initial machine learning model using a training sample dataset includes: inputting the historical wellbore temperature and historical wellbore pressure into the initial machine learning model to obtain a predicted wellbore flow rate; constructing a loss function based on the predicted wellbore flow rate and the historical wellbore flow rate; and training the initial machine learning model using the loss function to obtain a water injection profile inversion model.

[0013] This embodiment provides a method for determining the water absorption ratio. The method uses the water injection profile inversion model to determine the water absorption capacity of each perforation, including: real-time acquisition of field wellbore temperature distribution data; inputting the temperature distribution data into the water injection profile inversion model, the water injection profile inversion model outputting the wellbore flow distribution; and determining the water absorption capacity of each perforation section based on the wellbore flow distribution.

[0014] According to one aspect of the embodiments herein, the method determines the water absorption capacity of each perforation section according to the following formula:

[0015] Where Q(z) represents the flow rate at a wellbore depth of z, Q t Q represents the total injection volume at the wellhead. i This represents the flow rate entering the i-th perforation section, where i ranges from the first perforation section to the N(z)-th perforation section, and N(z) is the last perforation section at a wellbore depth of z.

[0016] This embodiment provides a device for constructing a water injection profile inversion model. The device includes: a historical initial parameter acquisition unit for acquiring historical initial parameters; a constraint parameter determination unit for determining parameters that are highly correlated with wellbore temperature from the historical initial parameters using a pre-constructed temperature profile model, as constraint parameters; a training sample dataset determination unit for determining the output results obtained after inputting the constraint parameters into the temperature profile model as a training sample dataset, the training sample dataset including historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate; and a water injection profile inversion model determination unit for training an initial machine learning model using the training sample dataset, and using the trained machine learning model as the water injection profile inversion model, the output of which is the flow rate distribution of the wellbore.

[0017] This embodiment also provides a water absorption capacity determination device, which includes: a data acquisition unit for real-time acquisition of wellbore temperature distribution data; a wellbore flow distribution determination unit for inputting the temperature distribution data into the water injection profile inversion model, wherein the water injection profile inversion model outputs the wellbore flow distribution; and a water absorption capacity determination unit for determining the water absorption capacity of each perforation section based on the wellbore flow distribution.

[0018] This embodiment also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the water injection profile inversion model construction method and the water absorption capacity determination method.

[0019] This scheme quantitatively explains the flow rate along the wellbore and the water absorption of each perforation section in the injection well, quantifies the degree of utilization of the target layer by the perforation in heterogeneous reservoirs, and provides data support for subsequent measures such as well profile adjustment, pressure reduction and injection enhancement. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments or prior art described herein, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this article. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 The diagram shown is a flowchart of a water injection profile inversion model construction method according to an embodiment of this paper;

[0022] Figure 2 The diagram shown is a flowchart of a method for determining constraint parameters according to an embodiment of this paper.

[0023] Figure 3 The diagram shown is a flowchart of a method for determining a training sample dataset according to an embodiment of this paper;

[0024] Figure 4 The diagram shown is a flowchart of a method for determining a temperature profile model according to an embodiment of this paper.

[0025] Figure 5 The diagram shown is a flowchart of a method for training an initial machine learning model according to an embodiment of this paper.

[0026] Figure 6 The diagram shown is a flowchart of a method for determining water absorption capacity according to an embodiment of this article;

[0027] Figure 7 The diagram shown is a structural schematic of a water injection profile inversion model construction device according to an embodiment of this paper.

[0028] Figure 8 The diagram shown is a structural schematic of a water absorption capacity determination device according to an embodiment of this article;

[0029] Figure 9 The diagram shown illustrates the change in wellbore temperature caused by an initial parameter in one of the embodiments described in this paper.

[0030] Figure 10 The diagram shown is a structural schematic of a computer device according to an embodiment of this article.

[0031] Explanation of symbols in the attached drawings:

[0032] 701. Historical Initial Parameter Acquisition Unit;

[0033] 702. Constraint Parameter Determination Unit;

[0034] 703. Training sample dataset determination unit;

[0035] 704. Determination of the unit for the water injection profile inversion model;

[0036] 801. Acquisition Unit;

[0037] 802. Wellbore flow distribution determination unit;

[0038] 803. Water absorption capacity determination unit;

[0039] 1002. Computer equipment;

[0040] 1004, Processor;

[0041] 1006. Memory;

[0042] 1008. Drive mechanism;

[0043] 1010. Input / Output Module;

[0044] 1012. Input devices;

[0045] 1014. Output devices;

[0046] 1016. Presentation device;

[0047] 1018. Graphical User Interface;

[0048] 1020. Network interface;

[0049] 1022. Communication link;

[0050] 1024. Communication bus. Detailed Implementation

[0051] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments herein will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments herein, and not all of the embodiments. Based on the embodiments herein, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this document.

[0052] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings herein are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0053] This specification provides the operational steps of the methods described in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or device products, the methods shown in the embodiments or drawings can be executed sequentially or in parallel.

[0054] It should be noted that the water injection profile inversion model construction method in this paper can be used in the field of petroleum exploration and oil and gas field development. This paper does not limit the application fields of the water injection profile inversion model construction method, water absorption capacity determination method and equipment.

[0055] Figure 1 The diagram shows a flowchart of a water injection profile inversion model construction method according to an embodiment of this paper, which specifically includes the following steps:

[0056] Step 101: Obtain historical initial parameters. In some embodiments of this specification, historical initial parameters are data related to the reservoir, wellbore, and fluids obtained from water injection wells over a past period using logging instruments and other tools.

[0057] In some embodiments of this specification, historical initial parameters include historical reservoir parameters, historical wellbore parameters, historical fluid parameters, and historical operating regime data. Historical reservoir parameters include: top and bottom depths, permeability, porosity, overall compressibility, average specific heat capacity of rock / formation fluid, and thermal conductivity of rock / formation fluid for each injection layer; historical fluid parameters mainly include the temperature of historically injected fluids at the wellhead, injected fluid density, viscosity, specific heat capacity, and thermal conductivity; historical wellbore parameters mainly include historical completion string structures, including but not limited to the length, inner diameter, and outer diameter of tubing and casing at each layer, the inner and outer diameters, specific heat capacity, and thermal conductivity of the cement sheath, the specific heat capacity and thermal conductivity of tubing and casing, and perforation locations; historical operating regime information mainly includes historical well opening and shut-in times and durations, and injection rates.

[0058] In some embodiments of this specification, a series of temperature change data along the wellbore can be obtained based on historical initial parameters and a pre-built temperature profile model of the injection well.

[0059] Step 102: Use the pre-built temperature profile model to determine the parameters among the historical initial parameters that are highly correlated with wellbore temperature, and use them as constraint parameters.

[0060] In some embodiments of this specification, temperature profile models are used to predict wellbore temperature distribution and are constructed based on the principles of conservation of momentum, mass, and energy. Specifically, the temperature profile model includes a wellbore flow model, a wellbore heat transfer model, and a reservoir heat transfer model. Inputting different historical initial parameters into the temperature profile model yields different wellbore temperature distribution results. Methods for constructing temperature profile models can be found [link to relevant documentation]. Figure 4 Detailed description.

[0061] In this step, each of the multiple historical initial parameters is quantitatively controlled and input into the temperature profile model to obtain the temperature results output by the model. Based on the rate of change of the temperature results, the parameters with the greatest impact on temperature among the multiple historical initial parameters are identified, that is, the parameters that are highly correlated with wellbore temperature.

[0062] Specifically, based on the pre-constructed temperature profile model, the changes in wellbore temperature and reservoir temperature under different well opening and closing times and geological conditions are simulated. The effects of historical initial parameters such as injection rate, injection water temperature, water injection time, shut-in time, injection layer permeability, and total heat transfer coefficient on the temperature profile are analyzed, and the correlation between each parameter and wellbore profile temperature is determined.

[0063] In some embodiments of this specification, wellbore temperature refers to the temperature of the injected fluid after it flows into the wellbore and mixes with the wellbore fluid. Wellbore temperature variation refers to the temperature change along the wellbore trajectory at regular intervals, for example, the temperature change every 0.5 meters to 1 meter along the wellbore trajectory. In some embodiments of this specification, reservoir temperature variation refers to the temperature change along the radial direction of the reservoir with the wellbore as the center, under different sounding conditions. For example, when the sounding depth is 500 meters, the temperature at locations in the reservoir 0.1 meters, 0.3 meters, 0.6 meters...10 meters, 50 meters, 100 meters from the wellbore.

[0064] Step 103 involves inputting the constraint parameters into the temperature profile model and using the resulting output as the training sample dataset. This training sample dataset includes historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate. In this step 103, based on the constraint parameters determined in step 102 that are highly correlated with wellbore temperature, the constraint parameters are input into the temperature profile model to obtain its output, which includes historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate. The wellbore flow rate refers to the flow distribution along the wellbore from the wellhead to the bottom. Further data preprocessing of the historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate is then used as training samples for the water injection profile inversion model to be trained.

[0065] Step 104: Train an initial machine learning model using the training sample dataset. Use the trained machine learning model as the water injection profile inversion model, where the output of the water injection profile inversion model is the flow distribution of the wellbore. In some embodiments of this specification, one or more initial machine learning models can be selected for model training, including but not limited to: Random Forest, XGBoost, LightGBM, etc. Further calculate the prediction error of each trained model and evaluate the inversion accuracy of each model to determine the water injection profile inversion model.

[0066] Figure 2 The diagram shown is a flowchart of a method for determining constraint parameters according to an embodiment of this paper, which specifically includes the following steps:

[0067] Step 201 involves inputting each historical initial parameter into the temperature profile model to quantitatively calculate the temperature change at a specific location in the wellbore corresponding to each historical initial parameter. In this step, historical reservoir parameters, historical wellbore parameters, historical fluid parameters, and historical operating regime data are input into the temperature profile model, and only one historical initial parameter changes value at each input. The temperature profile model outputs the change in wellbore temperature value based on this input historical initial parameter value.

[0068] Specifically, base values ​​are set for all historical parameters, and each time the base value of each parameter is increased or decreased by 50%. The sensitivity of the parameters is determined by quantitatively controlling the values ​​of each historical parameter and the average change in wellbore temperature at the perforation location in the middle of the wellbore at the end of water injection.

[0069] In some embodiments of this specification, based on the quantitative simulation results of step 201, it is determined that the changes in wellbore temperature caused by the six parameters of quantitatively controlling perforation section thickness, perforation section permeability, perforation section porosity, wellbore diameter, injection volume, and injection temperature are significantly higher than those caused by other parameters.

[0070] Table 1 shows six historical initial parameters determined as constraint parameters. Each historical initial parameter has a pre-set base value, and each base value is reduced by 50% and increased by 50%, resulting in two new values. Correspondingly, the base value, the smaller 50% value, and the increased 50% value of each historical initial parameter, along with other historical initial parameters that remain unchanged, are input into the temperature profile model to obtain different temperatures.

[0071] Table 1

[0072]

[0073] Step 202: Initial parameters whose temperature changes exceed preset values ​​are identified as constraint parameters. These constraint parameters include one or more of the following: perforation section thickness, perforation section permeability, injection volume, injection temperature, and perforation section spacing. This step determines initial parameters whose temperature changes exceed preset values ​​based on a comparison of the wellbore temperature changes caused by variations in multiple historical initial parameter baseline values.

[0074] like Figure 9 As shown, when the preset value is 0.001, the temperature changes of the perforation section thickness, perforation section permeability, injection volume, injection temperature, and perforation section interval distance are all greater than the preset value of 0.001. Therefore, the perforation section thickness, perforation section permeability, injection volume, injection temperature, and perforation section interval distance are all determined to be constraint parameters.

[0075] In some embodiments of this specification, the preset values ​​can be pre-set or adjusted in real time according to actual conditions. Based on the magnitude of the preset value, different parameters can be determined from historical initial parameters as constraint parameters. Different types of constraint parameters can form different constraint parameter groups.

[0076] Figure 3 The diagram shown is a flowchart of a method for determining a training sample dataset according to an embodiment of this paper, which specifically includes the following steps:

[0077] Step 301 involves setting different values ​​for each constraint parameter to form multiple constraint parameter groups. In this step, different parameter values ​​are set for each constraint parameter, and by changing the parameter values ​​of each constraint parameter, multiple different constraint parameter groups are formed. The constraint parameters mentioned in step 202 include: perforation section thickness, perforation section permeability, injection volume, injection temperature, and perforation section spacing. These constraint parameters form constraint parameter groups. By setting different parameter values ​​for each parameter in a constraint parameter group, multiple different constraint parameter groups can be formed.

[0078] For example, the parameter values ​​for the perforation section thickness can be [2, 4, 6, 8, 10], the parameter values ​​for the perforation section permeability can be [0.15, 0.5, 1, 3, 5], and the parameter values ​​for the injection volume can be any one of [20, 30, 40, 50, 60]; the parameter values ​​for the injection temperature can be any one of [7, 9, 11, 13, 15]. Assume there are 3 perforation sections in the wellbore, with 2 perforation intervals. The perforation interval distance can be any one of [(8, 3), (2, 3)]. Here, (8, 3) indicates that the interval between the first and second perforation sections is 8 meters, and (2, 3) indicates that the interval between the second and third perforation sections is 3 meters. Parameter values ​​are selected for each constraint parameter to form multiple constraint parameter groups. In the embodiments of this specification, it is assumed that a water injection well has three perforation sections, and each constraint parameter has multiple different parameter values ​​to choose from. Based on different combinations, 5000 sets of constraint parameters can be formed.

[0079] Step 302 involves inputting the constraint parameter set into the temperature profile model to calculate multiple sets of historical wellbore temperature, historical wellbore pressure, and historical wellbore flow rate data that vary with well depth. This step inputs 5000 sets of constraint parameter sets into the temperature profile model, yielding 5000 sets of historical wellbore temperature, historical wellbore pressure, and corresponding historical wellbore flow rate data for different constraint parameter sets.

[0080] Step 303: The historical wellbore temperature, historical wellbore pressure, historical wellbore flow rate, and well depth data are used as a training sample dataset. The historical wellbore temperature, historical wellbore pressure, historical wellbore flow rate, and well depth corresponding to different combinations of constraint parameters are used as training sample datasets for constructing the water injection profile inversion model.

[0081] Figure 4The diagram shown is a flowchart of a method for determining a temperature profile model according to an embodiment of this paper. In some embodiments of this specification, a micro-element segment at any location in the wellbore is taken, and a transient single-phase wellbore model considering micro-thermal effects is established based on the equations of mass conservation, momentum conservation, and energy conservation. The transient single-phase wellbore model includes two parts: a wellbore flow model and a wellbore heat transfer model. The specific steps for determining the temperature profile model are as follows:

[0082] Step 401: Obtain initial parameters from a past moment. These initial parameters include reservoir parameters, wellbore parameters, fluid parameters, and operating regime data. This step is similar to step 101 and will not be described in detail here. This step uses the reservoir parameters, wellbore parameters, fluid parameters, and operating regime as inputs to the temperature profile model to initialize the reservoir and wellbore temperatures and pressures.

[0083] Step 402: Based on the initial parameters, the principle of mass conservation, the principle of momentum conservation, and the principle of energy conservation, establish a wellbore model and a reservoir model respectively. The wellbore model and the reservoir model are coupled to form the temperature profile model.

[0084] Taking a small segment at any location within the wellbore, and by solving for the mass and momentum conservation principles at time t, a single-phase fluid flow model of the wellbore is established, namely, the wellbore flow model. The mass conservation principle can be described as follows:

[0085]

[0086] Where z is the well depth, ΔZ is the well depth micro-segment; R is the wellbore radius; ρ is the fluid density; r is the radial distance from the well axis; v is the fluid velocity in the wellbore; v0 is the fluid velocity flowing from the wellbore to the reservoir; ρ0 is the fluid density flowing from the wellbore to the reservoir; and γ is the degree of wellbore opening.

[0087] By analyzing the forces of gravity, friction, pressure difference, and reaction force from the pipe wall acting on the control volume, the momentum balance equation (2) of the control volume is obtained:

[0088]

[0089] Where p is the wellbore pressure, v is the wellbore velocity, R is the wellbore radius, z is the well depth, t is time, θ is the well inclination angle, cosθ is the cosine of the well inclination angle, g is the acceleration due to gravity, and f is the friction force.

[0090] Wellbore heat transfer models and reservoir heat transfer models were established during water injection and well shut-in periods, respectively. Based on the wellbore and reservoir heat transfer models, the wellbore temperature distribution and reservoir temperature distribution can be further calculated.

[0091] The wellbore heat transfer model during water injection is as follows:

[0092]

[0093] The wellbore heat transfer model during shut-in is as follows:

[0094]

[0095] Where v is the fluid velocity in the wellbore; T is the wellbore temperature; t is time; T r U represents the reservoir temperature. T γ is the reservoir temperature; γ is the wellbore opening degree; R is the wellbore radius; ρ is the fluid density. U is the heat capacity of the fluid; T ρ is the overall heat transfer coefficient; g is the gravitational acceleration; z is the wellbore depth; r is the radial distance from the well axis. w K is the radius of the wellbore. T is the thermal conductivity.

[0096] Taking full account of the micro-thermal effect, reservoir heat transfer models during water injection and during well shut-in were established based on the principle of energy conservation.

[0097] The reservoir heat transfer model during water injection is as follows:

[0098]

[0099] The reservoir heat transfer model during well shut-in is as follows:

[0100]

[0101] in, The average properties of formation fluids and rocks; Water volume injected per meter; K T denoted as λ, where λ is the thermal conductivity; r is the radial distance from the well axis; and z is the wellbore depth.

[0102] In the embodiments of this specification, since there is fluid flow and heat conduction between the wellbore and the reservoir model, and heat exchange occurs between the wellbore and the reservoir at all times, it is necessary to couple the wellbore heat transfer model and the reservoir heat transfer model to solve for the transient temperature profile of the reservoir and the transient temperature profile inside the wellbore.

[0103] The iterations are performed sequentially according to formulas (1) and (2) to ensure that the temperature inside the wellbore converges. The wellbore pressure and flow velocity at each location in the wellbore at time t can be determined. Furthermore, the wellbore flow rate at each location is determined by multiplying the flow velocity at each location in the wellbore by the area of ​​the wellbore cross-section.

[0104] By solving the wellbore energy balance equation and the reservoir energy balance equation, the wellbore flow rate, wellbore pressure, reservoir temperature distribution, and wellbore temperature distribution are calculated. The same method is then used to iteratively calculate the wellbore pressure, flow rate, and temperature for the next time step. For each time step, the calculations are updated based on the pressure and temperature from the previous time step. This process ultimately determines the temperature profile model.

[0105] Figure 5 The diagram shown is a flowchart of a method for training a water injection profile inversion model according to an embodiment of this paper, which specifically includes the following steps:

[0106] Step 501: Input the historical wellbore temperature, historical wellbore pressure, and well depth into the initial machine learning model to obtain the predicted wellbore flow rate.

[0107] Specifically, historical data on well depth, wellbore pressure, and wellbore temperature during water injection and shut-in phases are preprocessed. Preprocessing steps include, but are not limited to: using a moving average to remove parameter noise, using the KNN algorithm to impute missing values, and data normalization. The preprocessed data is used as the training dataset, and three machine learning models—Random Forest, XGBoost, and LightGBM—are selected as the initial models for this step. The training dataset is then input into each of the three initial models for training, yielding predicted wellbore flow rates for each model.

[0108] Step 502: Construct a loss function based on the predicted wellbore flow rate and the historical wellbore flow rate. As described in step 101, the training sample dataset also includes the historical wellbore flow rate, which is used to evaluate the error of the predicted values ​​output by the three initial machine learning models. Specifically, based on the root mean square error between the predicted flow rate and the wellbore flow rate corresponding to the sample set, a loss function is constructed for each initial machine learning model, and the hyperparameters of each initial machine learning model are continuously adjusted.

[0109] Step 503: Train the initial machine learning model using the loss function to obtain the water-filled profile inversion model. Adjust the parameters of each initial machine learning model until the loss function converges to a certain range, confirming training completion and obtaining the water-filled profile inversion model. In this step, compare the prediction errors of models trained using three initial machine learning algorithms: Random Forest, XGBoost, and LightGBM, and select the model with the smallest prediction error as the water-filled profile inversion model. Specifically, the root mean square error (RMSE) and coefficient of determination (R²) are used. 2 Two metrics were used to evaluate the performance of the three models, and the model trained using XGBoost was ultimately determined to be the water injection profile inversion model.

[0110]

[0111]

[0112] Where RMSE is the root mean square error, R 2 The coefficient of determination is N, where N is the number of data points, and y is the coefficient of determination. i For the actual historical wellbore flow rate, To predict flow for the model, This represents the average value of the actual flow rate in the wellbore.

[0113] In some embodiments of this specification, the XGBoost water injection profile prediction and inversion model runs quickly, has strong noise resistance and regression calculation capabilities, and avoids the time-consuming drawbacks of conventional inversion methods.

[0114] Figure 6 The diagram shown is a flowchart of a method for determining water absorption capacity according to an embodiment of this paper, which specifically includes the following steps:

[0115] Step 601: Real-time acquisition of wellbore temperature distribution data. This step can utilize distributed fiber optic temperature sensors to obtain wellbore temperature data distributed along the wellbore trajectory of a sample well.

[0116] Step 602: The temperature distribution data is input into the water injection profile inversion model, and the water injection profile inversion model outputs the wellbore flow distribution. This step inputs the real-time collected wellbore temperature distribution data into the water injection profile inversion model trained in the previous steps to obtain the wellbore flow distribution. The wellbore flow changes with well depth. As the well depth increases, the wellbore flow gradually decreases.

[0117] Step 603: Determine the water absorption capacity of each perforation section based on the wellbore flow distribution. In this step, the water absorption capacity of each perforation section is determined based on the wellbore flow distribution. Typically, multiple injection layers exist in the same injection well, with each perforation section corresponding to one injection layer. The fluid in the wellbore is injected from the wellhead and enters multiple injection layers through the perforations. Therefore, the flow rate is highest at the wellhead and decreases sequentially through each perforation, theoretically dropping to zero at the bottom of the well. However, in reality, a small amount of injected water usually accumulates at the bottom of the well. Therefore, the wellbore flow distribution changes according to the different water absorption capacities of each injection layer. When passing through layers with strong water absorption capacity, the flow rate decreases significantly, while when passing through layers with weak water absorption capacity, the flow rate decreases more slowly. Therefore, the water absorption capacity at different locations within the wellbore depth is determined based on the wellbore flow rate at those locations.

[0118] The relative water absorption capacity of each perforation section is determined using the following formula:

[0119]

[0120] Where Q(z) represents the wellbore flow rate at wellbore depth z, Q t Q represents the total injection volume at the wellhead. i This represents the flow rate entering the i-th perforation section, where i ranges from the first perforation section to the N(z)-th perforation section, and N(z) is the last perforation section at a wellbore depth of z.

[0121] For example, a water injection well has a depth range of 0-2000 meters. The first perforation section is at a depth of 1500-1550 meters, the second at 1650-1700 meters, and the third at 1800-1820 meters. If we calculate Q(1750m), then: The total injection volume at the wellhead is reduced by the wellbore flow rates corresponding to the first and second perforation sections. Here, Q1 and Q2 represent the wellbore flow rates of the first and second perforation sections, respectively, which are obtained from the wellbore flow rates predicted by the water injection profile inversion model in the aforementioned steps.

[0122] If we calculate Q(1680m), assuming the flow rate inside the perforated section decreases uniformly along the well depth, we can calculate it according to the depth ratio:

[0123]

[0124] The wellbore flow rate at a depth of 1680 meters can be calculated.

[0125] By observing the changes in wellbore flow rate at different locations within the well depth, the water absorption capacity at those locations can be determined. If the wellbore flow rate decreases significantly at a certain location, it indicates that the water absorption capacity of that wellbore layer is strong; if the wellbore flow rate decreases slowly at a certain location, it indicates that the water absorption capacity of that wellbore layer is weak.

[0126] like Figure 7 The diagram shown is a structural schematic of a water injection profile inversion model construction device according to an embodiment of this paper. The basic structure of the device is illustrated in the diagram. The functional units and modules can be implemented using software, general-purpose chips, or specific chips to construct the water injection profile inversion model. The device specifically includes:

[0127] Historical initial parameter acquisition unit 701 is used to acquire historical initial parameters;

[0128] The constraint parameter determination unit 702 is used to determine parameters that are highly correlated with wellbore temperature from the historical initial parameters using a pre-built temperature profile model, and use these parameters as constraint parameters.

[0129] The training sample dataset determination unit 703 is used to determine the output result obtained after inputting the constraint parameters into the temperature profile model as a sample dataset. The training sample dataset includes historical wellbore temperature, historical wellbore pressure and historical wellbore flow rate.

[0130] The water injection profile inversion model determination unit 704 is used to train an initial machine learning model using the training sample dataset, and to use the trained machine learning model as the water injection profile inversion model. The output of the water injection profile inversion model is the flow distribution of the wellbore.

[0131] This scheme quantitatively explains the flow rate along the wellbore and the water absorption of each perforation section in the injection well, quantifying the degree of utilization of the target layer by perforations in heterogeneous reservoirs. Data-driven machine learning models are trained using detailed datasets, ensuring the accuracy of the injection profile inversion model and compensating for the shortcomings of machine learning methods in oil and gas field development injection wells. It provides data support for subsequent injection well profile modification, pressure reduction, and injection enhancement measures.

[0132] As one embodiment of this article, reference may also be made to, for example, Figure 8 The diagram shown is a schematic representation of the water absorption capacity determination device in this embodiment. The device specifically includes:

[0133] The data acquisition unit 801 is used to acquire real-time data on the temperature distribution of the wellbore.

[0134] The wellbore flow distribution determination unit 802 is used to input the temperature distribution data into the water injection profile inversion model, and the water injection profile inversion model outputs the wellbore flow distribution.

[0135] The water absorption capacity determination unit 803 is used to determine the water absorption capacity of each perforation section based on the flow distribution of the wellbore.

[0136] Figure 9The diagram illustrates the temperature change in the wellbore caused by initial parameters in an embodiment of this paper. The baseline value for the perforation section thickness is 8 meters. When the baseline value is reduced by 50%, the temperature change in the wellbore caused by the perforation section thickness is -0.063; when the baseline value is increased by 50%, the temperature change is 0.082. The baseline value for the perforation section permeability is 10 mD. When the baseline value is reduced by 50%, the temperature change in the wellbore caused by the permeability is -0.028; when the baseline value is increased by 50%, the temperature change is 0.044. The baseline value for the perforation section porosity is 0.2. When the baseline value is reduced by 50%, the temperature change in the wellbore caused by the porosity is 0.001, and the temperature change is 0.002. The baseline value for the wellbore diameter is 0.1m. When the baseline value decreases by 50%, the temperature change in the wellbore caused by the wellbore diameter is -0.021; when the baseline value increases by 50%, the temperature change in the wellbore caused by the wellbore diameter is 0.034. The baseline value for the injection rate is 40m³ / d. When the baseline value decreases by 50%, the temperature change in the wellbore caused by the injection rate is -2.459; when the baseline value increases by 50%, the temperature change in the wellbore caused by the injection rate is 3.757. The baseline value for the injection temperature is 12℃. When the baseline value decreases by 50%, the temperature change in the wellbore caused by the injection temperature is -0.995; when the baseline value increases by 50%, the temperature change in the wellbore caused by the injection temperature is 0.993.

[0137] like Figure 10 As shown, a computer device provided in this embodiment is used in the method for constructing a water injection profile inversion model and determining water absorption capacity. The computer device 1002 may include one or more processors 1004, such as one or more central processing units (CPUs), each of which can implement one or more hardware threads. The computer device 1002 may also include any memory 1006 for storing information of any kind, such as code, settings, data, etc. Non-limitingly, for example, the memory 1006 may include any type of RAM, any type of ROM, flash memory, hard disk, optical disk, etc. More generally, any memory can use any technology to store information. Further, any memory can provide volatile or non-volatile retention of information. Further, any memory can represent a fixed or removable component of the computer device 1002. In one case, when the processor 1004 executes associated instructions stored in any memory or combination of memories, the computer device 1002 can perform any operation of the associated instructions. The computer device 1002 also includes one or more drive mechanisms 1008 for interacting with any memory, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0138] Computer device 1002 may further include an input / output module 1010 (I / O) for receiving various inputs (via input device 1012) and providing various outputs (via output device 1014). A specific output mechanism may include a presentation device 1016 and an associated graphical user interface (GUI) 1018. In other embodiments, the input / output module 1010 (I / O), input device 1012, and output device 1014 may be omitted, and the device may function solely as a computer device within a network. Computer device 802 may also include one or more network interfaces 1020 for exchanging data with other devices via one or more communication links 1022. One or more communication buses 1024 couple the components described above together.

[0139] The communication link 1022 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. The communication link 1022 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0140] Corresponding to Figures 1 to 6 In addition to the methods described above, this embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the above-described methods.

[0141] This embodiment also provides a computer-readable instruction, wherein when a processor executes the instruction, the program therein causes the processor to perform the following: Figures 1 to 6 The method shown.

[0142] It should be understood that in the various embodiments of this document, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0143] It should also be understood that, in the embodiments herein, the term "and / or" is merely a description of the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following associated objects have an "or" relationship.

[0144] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this document.

[0145] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0146] In the embodiments provided herein, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or units, or they may be electrical, mechanical, or other forms of connection.

[0147] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments described herein, depending on actual needs.

[0148] Furthermore, the functional units in the various embodiments of this document can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0149] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this paper, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this paper. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] This document uses specific embodiments to illustrate the principles and implementation methods of this document. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this document. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this document. Therefore, the content of this specification should not be construed as a limitation of this document.

Claims

1. A method for constructing a water injection profile inversion model, characterized in that: The method comprises: Acquiring historical initial parameters, wherein the historical initial parameters include: historical reservoir parameters, historical wellbore parameters, historical fluid parameters and historical working system data; The parameters with the greatest correlation with the wellbore temperature in the historical initial parameters are determined by using a pre-constructed temperature profile model as constraint parameters, wherein the temperature profile model is determined by the following steps: obtaining initial parameters at a certain time in the past, wherein the initial parameters include reservoir parameters, wellbore parameters, fluid parameters and working system data; establishing a wellbore model and a reservoir model respectively according to the initial parameters, the principle of conservation of mass, the principle of conservation of momentum and the principle of conservation of energy, wherein the wellbore model and the reservoir model are coupled to form the temperature profile model; The output result obtained after inputting the constraint parameters into the temperature profile model is determined as a training sample data set, wherein the training sample data set includes historical wellbore temperature, historical wellbore pressure and historical wellbore flow rate; The training sample data set is used to train an initial machine learning model, and the trained machine learning model is used as a water injection profile inversion model, and the output of the water injection profile inversion model is the flow distribution of the wellbore.

2. The method for constructing a water injection profile inversion model according to claim 1, characterized in that: The temperature profile model is used to determine the parameters with the greatest correlation with the wellbore temperature in the historical initial parameters, which are used as constraint parameters, including: Inputting each historical initial parameter into the temperature profile model, and quantitatively calculating the temperature change at a certain position of the wellbore corresponding to each historical initial parameter; An initial parameter whose temperature variation is greater than a preset value is determined as a constraint parameter, wherein the constraint parameter includes one or more of perforation section thickness, perforation section permeability, injection volume, injection temperature and perforation section spacing.

3. The method for constructing a water injection profile inversion model according to claim 1, characterized in that: The output result obtained after inputting the constraint parameters into the temperature profile model is determined as the training sample data set including: Setting different values ​​for each constraint parameter to form multiple constraint parameter groups; Inputting the constraint parameter group into the temperature profile model, and calculating multiple groups of historical wellbore temperatures, historical wellbore pressures, and historical wellbore flow rates that vary with well depth; The historical wellbore temperature, the historical wellbore pressure, the historical wellbore flow rate and well depth data are used as a set of training sample data sets.

4. The method for constructing a water injection profile inversion model according to claim 1, characterized in that: Training the initial machine learning model using the training sample dataset includes: Inputting the historical wellbore temperature, historical wellbore pressure, and well depth into an initial machine learning model to obtain a wellbore flow prediction value; Constructing a loss function according to the predicted wellbore flow rate and the historical wellbore flow rate; The initial machine learning model is trained using the loss function to obtain a water injection profile inversion model.

5. A method for determining water absorption capacity, characterized in that: The method uses the water injection profile inversion model trained by the water injection profile inversion model construction method according to any one of claims 1 to 4, and comprises: Real-time collection of on-site wellbore temperature distribution data; Inputting the temperature distribution data into the water injection profile inversion model, the water injection profile inversion model outputs the wellbore flow distribution; The water absorption capacity of each perforation section is determined according to the wellbore flow distribution.

6. The method for determining water absorption capacity according to claim 5, characterized in that: The method determines the water absorption capacity of each perforation section according to the following formula: Where Q(z) represents the flow rate at the wellbore depth z, Q t is the total injection volume at the wellhead, Q i represents the flow rate entering the i-th perforation section, where i ranges from the 1st perforation section to the N(z)th perforation section, and N(z) is the last perforation section at the wellbore depth z.

7. A water injection profile inversion model construction device, characterized in that: The device comprises: A historical initial parameter acquisition unit, used to acquire historical initial parameters, the historical initial parameters including: historical reservoir parameters, historical wellbore parameters, historical fluid parameters and historical working system data; A constraint parameter determination unit is used to use a pre-constructed temperature profile model to determine a parameter with a large correlation with the wellbore temperature from the historical initial parameters as a constraint parameter, wherein the temperature profile model is determined by the following steps: obtaining initial parameters at a certain time in the past, the initial parameters including reservoir parameters, wellbore parameters, fluid parameters and working system data; establishing a wellbore model and a reservoir model respectively according to the initial parameters, the principle of conservation of mass, the principle of conservation of momentum and the principle of conservation of energy, wherein the wellbore model and the reservoir model are coupled to form the temperature profile model; A training sample data set determination unit, used for determining the output result obtained after the constraint parameters are input into the temperature profile model as a training sample data set, wherein the training sample data set includes historical wellbore temperature, historical wellbore pressure and historical wellbore flow rate; The water injection profile inversion model determination unit is used to train an initial machine learning model using the training sample data set, and use the trained machine learning model as a water injection profile inversion model, wherein the output of the water injection profile inversion model is the flow distribution of the wellbore.

8. A device for determining water absorption capacity, characterized in that: The device comprises: A collection unit, used to collect on-site wellbore temperature distribution data in real time; A wellbore flow rate distribution determination unit, used for inputting the temperature distribution data into a water injection profile inversion model trained according to the water injection profile inversion model construction method according to any one of claims 1 to 4, and the water injection profile inversion model outputs the wellbore flow rate distribution; The water absorption capacity determination unit is used to determine the water absorption capacity of each perforation section according to the wellbore flow distribution.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images