Fracture size prediction method, device, and storage medium

Abstract

The present application relates to oil and gas geological exploration technology field, disclose a kind of fracture size prediction method, equipment and storage medium, the method comprises: determining wellbore flow model, wellhead well bottom boundary calculation model, fracture size calculation model, well bottom boundary calculation model is equivalent to the circuit series structure containing resistance, capacitance, inductance for fracture dynamic response.Use fracture size prediction model and the above multiple calculation models, the value range of multiple control parameters, respectively determine the simulated water hammer pressure value of wellhead at different time under multiple groups of control parameter values, to invert the target group control parameter value corresponding to the simulated water hammer pressure value of wellhead at different time and the actual water hammer pressure value match, determine the reference size of fracture according to target group control parameter value and relevant calculation model, to realize the accurate determination of fracture size.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of oil and gas geological exploration technology, and in particular to a method, equipment and storage medium for predicting fracture size. Background Technology

[0002] In the petroleum industry, fracturing refers to a method of creating fractures in oil and gas reservoirs using hydraulic force during oil or gas production; it is also known as hydraulic fracturing. Fracturing artificially creates fractures in the formation, improving the underground flow environment for oil and increasing oil well production. It plays an important role in improving bottomhole flow conditions, mitigating inter-layer flow, and improving reservoir activation.

[0003] In the field of hydraulic fracturing, real-time and accurate acquisition of fracture morphology (length, height, and width) is crucial for optimizing construction. Current mainstream technologies such as microseismic monitoring and distributed fiber optic sensing suffer from drawbacks such as high cost, large indirect estimation errors, and difficulty in capturing transient responses during pump shutdown. Summary of the Invention

[0004] The purpose of this invention is to provide at least one method, device and storage medium for predicting crack size, which can predict crack size in a timely and accurate manner after the pump is shut down in water hammer fracturing.

[0005] To address the aforementioned technical problems, at least one embodiment of this application provides a crack size prediction method, comprising: The wellbore flow rate model, wellhead boundary calculation model, bottom boundary calculation model, and fracture size calculation model are determined. The bottom boundary calculation model equates the fracture dynamic response to a series circuit structure containing resistance, capacitance, and inductance. Resistance represents the ratio of pressure difference to flow rate that maintains fracture opening, capacitance represents the rate of change of fracture volume with pressure, and inductance represents the inertial effect of fluid acceleration or deceleration within the fracture. Collect the actual water hammer pressure values ​​at the wellhead of the target well at different times after the pump is stopped; The actual water hammer pressure values ​​at the wellhead at different times, the value ranges of various control parameters of the target well, the wellbore flow rate model, the wellhead boundary calculation model, the bottom well boundary calculation model, and the fracture size calculation model are input into the fracture size prediction model, so that the fracture size prediction model performs the following process: Based on the range of values ​​of the various control parameters, multiple sets of control parameter values ​​are determined, wherein the various control parameters include resistance, capacitance and inductance; Based on the wellbore flow model, the wellhead boundary calculation model, and the well bottom boundary calculation model, the simulated water hammer pressure values ​​at the wellhead at different times under the multiple sets of control parameter values ​​are determined respectively. From the multiple sets of control parameter values, determine the target set of control parameter values ​​that match the simulated water hammer pressure value at the wellhead at different times with the actual water hammer pressure value; Based on the target group control parameter values, the well bottom boundary calculation model, the wellbore flow rate model, and the fracture size calculation model, the reference size of the fracture under the target group control parameter values ​​is determined.

[0006] In some optional embodiments, the bottom boundary calculation model is represented as:

[0007] in, The instantaneous driving pressure difference at the bottom of the well is defined as the bottom-hole fluid pressure time history. With minimum ground stress Difference: , This represents the flow rate at the bottom of the well.

[0008] In some optional embodiments, determining the reference size of the fracture under the target group control parameter values ​​based on the target group control parameter values, the bottom hole boundary calculation model, the wellbore flow rate model, and the fracture size calculation model includes: Based on the target group circuit parameter values, the bottom hole boundary calculation model, and the wellbore flow model, the bottom hole fluid pressure time history at different times is determined under the target group circuit parameter values. The target group control parameter values ​​include the target group circuit parameter values, which include target resistance, target capacitance, and target inductance values. The average value of the bottom hole fluid pressure time history at different times is determined as the average bottom hole pressure; The average net pressure within the fracture is determined based on the average bottom hole pressure and the minimum ground stress. The reference size of the fracture is determined based on the average net pressure within the fracture, the rock mechanics parameters of the target well, and the fracture size calculation model.

[0009] In some optional embodiments, determining the average of the bottom hole fluid pressure time histories at different times as the average bottom hole pressure includes: The average bottom hole pressure is determined using the following formula:

[0010] in, This represents the initial moment when the pressure stabilizes after the water hammer wave decays. The set time window length.

[0011] In some optional embodiments, determining the reference size of the fracture based on the average net pressure within the fracture, the set rock mechanics parameters of the target well, and the fracture size calculation model includes: The length, height, and width of the cracks are determined using the following formulas:

[0012]

[0013] in, The crack half-length is expressed in units of 1 / 2000. ; The crack height is expressed in units of... ; The effective thickness of the reservoir in the target well, in units of ; The crack width is expressed in units of 1000 mm. ; The Young's modulus of the rock in the target well, in units of ; The Poisson's ratio of the rock in the target well is dimensionless. The average net pressure within the crack, in units of ; This is a near-wellbore friction correction function related to the target resistance value in the target group circuit parameter values.

[0014] In some optional embodiments, the wellhead boundary calculation model is as follows:

[0015] in, Initial displacement, unit: ; This refers to the time taken to stop the pump.

[0016] In some optional embodiments, the wellbore model uses the method of characteristics to discretize and solve the water hammer equation. The recursive formula for the pressure head H and flow rate Q at different nodes inside the wellbore is as follows:

[0017]

[0018] in, The pressure head is the unit of pressure head. ; Volumetric flow rate, unit: ; Pipe impedance coefficient, unit: ; This is the pipe resistance coefficient, in units of... ; superscript , Representing the current time and the next calculation time respectively; subscript , , This indicates the spatial location node number after discretization along the wellbore pipe.

[0019] In some optional embodiments, the crack size prediction model includes a first crack size prediction sub-model, a second crack size prediction sub-model, and the fusion sub-model; The working process of the crack size prediction model includes: The actual water hammer pressure at the wellhead at different times, the range of values ​​of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the bottom boundary calculation model, and the fracture size calculation model are respectively input into the first fracture size prediction sub-model and the second fracture size prediction sub-model, so that the first fracture size prediction sub-model and the second fracture size prediction sub-model respectively output the fracture size vector under the target group control parameter values; The crack size vectors of the first crack size prediction sub-model and the second crack size prediction sub-model are fused by a fusion sub-model, and the reference size is determined based on the fusion result.

[0020] An embodiment of this application provides a crack size prediction device, the device comprising: The determination module is used to determine the wellbore flow rate model, wellhead boundary calculation model, bottom boundary calculation model, and fracture size calculation model. The bottom boundary calculation model equates the fracture dynamic response to a series circuit structure containing resistance, capacitance, and inductance. Resistance represents the ratio of pressure difference to flow rate that maintains fracture opening, capacitance represents the rate of change of fracture volume with pressure, and inductance represents the inertial effect of fluid acceleration or deceleration within the fracture. The data acquisition module is used to collect the actual water hammer pressure values ​​at the wellhead of the target well at different times after the pump is stopped; The processing module is used to input the actual water hammer pressure values ​​at the wellhead at different times, the value ranges of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the well bottom boundary calculation model, and the fracture size calculation model into the fracture size prediction model, so that the fracture size prediction model performs the following process: determining multiple sets of control parameter values ​​based on the value ranges of the various control parameters; determining the simulated water hammer pressure values ​​at the wellhead at different times under the multiple sets of control parameter values ​​based on the wellbore flow model, the wellhead boundary calculation model, and the well bottom boundary calculation model; determining the target set of control parameter values ​​from the multiple sets of control parameter values ​​that match the simulated water hammer pressure values ​​at the wellhead at different times with the actual water hammer pressure values; and determining the reference size of the fracture under the target set of control parameter values ​​based on the target set of control parameter values, the well bottom boundary calculation model, the wellbore flow model, and the fracture size calculation model; wherein, the various control parameters include resistance, capacitance, and inductance.

[0021] At least one embodiment of this application also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the crack size prediction method described above.

[0022] At least one embodiment of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the crack size prediction method described above.

[0023] At least one embodiment of this application also provides a computer program product, including a computer program that, when executed by a processor, implements the crack size prediction method described above.

[0024] The fracture size prediction method provided in the above-described embodiments of this application establishes a wellbore flow rate model, a wellhead boundary calculation model, a bottom-of-well boundary calculation model, and a fracture size calculation model. The bottom-of-well boundary calculation model equates the dynamic response of the fracture to a series circuit structure containing resistance, capacitance, and inductance. Through a machine learning model—the fracture size prediction model—preset value ranges for various parameters determining the fracture size, along with the aforementioned calculation models, are input into the fracture size prediction model. This allows the fracture size prediction model to simulate and calculate the wellhead water hammer pressure at different times for each combination of parameter values. The simulation calculation is then compared with the water hammer pressure curve formed by the actual collected wellhead water hammer pressure values. This determines the target set of control parameter values ​​that match the simulated water hammer pressure values ​​at different times with the actual water hammer pressure values ​​at the corresponding times. These values ​​are used as the inversion result of the control parameters. Finally, based on the target control set parameter values ​​and the aforementioned calculation models, the fracture size is calculated.

[0025] This scheme uses an equivalent circuit model of resistor R, inductor L, and capacitor C to accurately characterize the dynamic mechanical properties of the crack, and integrates a machine learning model to achieve high-precision real-time prediction of crack morphology and crack size during water hammer fracturing. Attached Figure Description

[0026] One or more embodiments are illustrated by way of example with reference to the accompanying drawings, and these illustrative descriptions do not constitute a limitation on the embodiments.

[0027] Figure 1 This is a flowchart of a crack size prediction method provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the water hammer pressure curve matching effect provided in an embodiment of this application; Figure 3 This is a schematic diagram of a crack size prediction device provided in an embodiment of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this application to help readers better understand this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments. The division of the various embodiments below is for the convenience of description and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0029] Although water hammer pressure wave signals contain dynamic information about fractures, traditional methods face two major bottlenecks: (1) the mechanism model relies on simplified assumptions and cannot accurately analyze the nonlinear propagation of water hammer waves in the wellbore-fracture system; (2) the strong noise and time-varying nature of high-frequency pressure data in the field make it difficult for conventional algorithms to establish a mapping relationship between the signal and the fracture.

[0030] To address this, this application proposes a solution that integrates high-precision numerical simulation with hybrid machine learning. By establishing a wellbore flow-fracture coupling model, water hammer curve matching is achieved. Furthermore, based on the LSTM-random forest fusion algorithm, the quantitative correlation between pressure wave characteristics and fracture parameters is mined, and fracture morphology is predicted, thereby overcoming the limitations of accuracy and timeliness in existing technologies.

[0031] The following is a detailed description of the implementation details of the crack size prediction method in this embodiment. The following content is only for the convenience of understanding the implementation details and is not necessary for implementing this solution.

[0032] Example 1: The crack size prediction method provided in this application can be applied to electronic devices with communication, computing, and data storage capabilities. Figure 1 This is a flowchart of the crack size prediction method provided in the embodiments of this application, and its specific process can be as follows: Figure 1 As shown, it includes: Step 101: Determine the wellbore flow rate model, wellhead boundary calculation model, bottom hole boundary calculation model, and fracture size calculation model. The bottom hole boundary calculation model equates the fracture dynamic response to a series circuit structure containing resistance, capacitance, and inductance. Resistance represents the ratio of pressure difference to flow rate to maintain fracture opening, capacitance represents the rate of change of fracture volume with pressure, and inductance represents the inertial effect of fluid acceleration or deceleration within the fracture.

[0033] Step 102: Collect the actual water hammer pressure values ​​at the wellhead of the target well at different times after the pump is stopped.

[0034] Step 103: Input the actual water hammer pressure values ​​at the wellhead at different times, the value ranges of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the bottom-of-well boundary calculation model, and the fracture size calculation model into the fracture size prediction model, so that the fracture size prediction model performs the following process: Step 1041: Determine multiple sets of control parameter values ​​based on the range of values ​​for various control parameters, including resistance, capacitance, and inductance.

[0035] Step 1042: Based on the wellbore flow rate model, wellhead boundary calculation model, and bottom boundary calculation model, determine the simulated water hammer pressure values ​​at the wellhead at different times under multiple sets of control parameter values.

[0036] Step 1043: From multiple sets of control parameter values, determine the target set of control parameter values ​​that match the simulated water hammer pressure value and the actual water hammer pressure value at the wellhead at different times.

[0037] Step 1044: Determine the reference size of the fracture under the target group control parameter values, based on the target group control parameter values, the bottom hole boundary calculation model, the wellbore flow model, and the fracture size calculation model.

[0038] When it is necessary to predict the size of fractures formed during water hammer fracturing in a target well, the solution provided in this application embodiment can be used. In this case, the fracture size prediction model in this application embodiment is a machine learning model that has already been trained. Since the training process of this model is similar to the actual use, i.e., the inference process, this embodiment will use the model inference process as an example for explanation.

[0039] In this embodiment, the water hammer pressure wave signal is the oscillation decay curve of the wellhead pressure over time, collected by a high-frequency pressure gauge at the wellhead location after the pump is stopped.

[0040] To accurately predict the size of fractures formed in a target well during water hammer fracturing, a wellbore flow model must first be established, and wellhead and bottom boundary calculation models must be constructed based on the method of characteristics. The bottom boundary calculation model equates the dynamic response of the fracture to a series circuit structure containing resistance R, capacitance C, and inductance I, with the following governing equations:

[0041] in, The instantaneous driving pressure difference at the bottom of the well is defined as the bottom-hole fluid pressure time history. With minimum ground stress Difference: , This represents the flow rate at the bottom of the well. Resistance. Characterized by the ratio of pressure differential to flow rate maintaining fracture opening, capacitance Inductance is a characterizing factor representing the rate of change of crack volume with pressure. Characterizes the inertial effect of fluid acceleration or deceleration within a crack.

[0042] The wellhead boundary calculation model adopts the following linear decay model of pump stoppage flow:

[0043] in, Initial displacement, in units of ; This represents the pump shutdown time. Where t=0 indicates the start time of triggering the pump shutdown.

[0044] The wellbore model (i.e., the wellbore-fracture coupled transient flow model) uses the method of characteristics to discretize and solve the water hammer equations for any node inside the wellbore pipe. exist The pressure of time With traffic The pressure head H and flow rate Q at different locations inside the wellbore can be explicitly obtained from the parameters of the adjacent nodes at the previous time step. The recursive formula for these parameters is as follows:

[0045]

[0046] in, The pressure head is the unit. ; Volumetric flow rate, unit: ; Pipe impedance coefficient, unit: ; This is the pipe resistance coefficient, in units of... ; superscript , Representing the current time and the next calculation time respectively; subscript , , This indicates the spatial location node number after discretization along the wellbore pipe.

[0047] It should be noted that in the above wellhead boundary calculation model Corresponding to the above C -- In the equation .

[0048] The crack size calculation model is used to calculate the length, height, and width of cracks, using the following formula:

[0049]

[0050] in, The crack half-length is expressed in units of 1 / 2. ; The crack height is expressed in units of... ; The effective thickness of the reservoir in the target well, in units of ; The crack width is expressed in units of 1000 mm. ; The Young's modulus of the rock in the target well, in units of ; Let be the Poisson's ratio of the rock in the target well, which is dimensionless; The average net pressure within the crack, in units of ; To match the target resistance value in the target group circuit parameter values ​​below. Related near-wellbore friction correction function.

[0051] In addition to constructing the aforementioned calculation model, it is also necessary to collect the actual water hammer pressure values ​​at the wellhead at different times after the pump is stopped. Based on these collected values, an actual water hammer pressure wave curve at the wellhead can be generated. Figure 2 As shown.

[0052] In addition, it is necessary to determine the main controlling factors used to determine the crack size prediction results—the value range of various control parameters.

[0053] Optionally, these control factors may include: Young's modulus, Poisson's ratio, minimum ground stress, capacitance, resistance, and inductance, with the following value ranges: Young's modulus: 10–40 GPa; Poisson's ratio: 0.2–0.3; Minimum ground stress: 30–80 MPa; Capacitance: 0.2 × 10 -5 ~2.1×10 -5 m³ / Pa; Inductance: 0.3~2.9 Pa·s² / m³; Resistance: 1×10⁹~1×10¹¹.

[0054] It should be noted that Young's modulus, Poisson's ratio, and minimum in-situ stress vary for different wells, and are values ​​determined by prior knowledge. Capacitance, resistance, and inductance, however, can share the same range of values ​​for all wells.

[0055] The various calculation models mentioned above, the actual water hammer pressure values ​​at the wellhead at different times (i.e., the actual water hammer pressure wave curve at the wellhead), and the value ranges of various control parameters of the target well are provided to the fracture size prediction model. This allows the fracture size prediction model to combine this information and predict the size of the fractures in the target well. The core idea of ​​the prediction process is as follows: First, numerical simulation of the water hammer pressure curve is performed based on the wellbore flow model and boundary calculation model. By adjusting the values ​​of resistor R, capacitor C, and inductor I, the simulated water hammer pressure wave curve at the wellhead is matched with the actual water hammer pressure wave curve collected on-site, thereby inverting the parameter values ​​of the equivalent circuit at the bottom of the well to obtain the equivalent circuit at the bottom of the well. In other words, the target resistance, target capacitance, and target inductance values ​​are determined: The target set of circuit parameter values, under these target circuit parameter values, the simulated water hammer pressure wave curve at the wellhead calculated based on the above-mentioned multiple calculation models matches the actual water hammer pressure wave curve at the wellhead. Secondly, [the following is a separate, unrelated sentence:] ... Substitute the data into the bottom-hole boundary calculation model and simultaneously establish the wellbore flow model to calculate the bottom-hole fluid pressure time history after pump shutdown. Then, the crack size is calculated based on the crack size calculation model.

[0056] Specifically, the crack size prediction model performs the following process: 1. Determine multiple sets of control parameter values ​​based on the range of values ​​for various control parameters, including resistance, capacitance, and inductance.

[0057] 2. Based on the wellbore flow model, wellhead boundary calculation model, and bottom boundary calculation model, determine the simulated water hammer pressure values ​​at the wellhead at different times under multiple sets of control parameter values.

[0058] 3. From multiple sets of control parameter values, determine the target set of control parameter values ​​that match the simulated water hammer pressure value and the actual water hammer pressure value at the wellhead at different times.

[0059] 4. Based on the target group control parameter values, the well bottom boundary calculation model, the wellbore flow rate model, and the fracture size calculation model, determine the reference size of the fracture under the target group control parameter values.

[0060] Each set of control parameter values ​​is composed of a parameter value taken from the range of values ​​for each control parameter.

[0061] For each set of control parameter values: the characteristic line equations of the wellhead boundary calculation model and the wellhead nodes can be used. The flow rate and head at the wellhead at different times are solved simultaneously. Based on the head calculation results at the wellhead at different times, the corresponding pressure is calculated to obtain the simulated water hammer pressure value at the wellhead at different times, thus forming the simulated water hammer pressure curve at the wellhead.

[0062] Optionally, a least-squares optimization algorithm can then be used to match the simulated water hammer pressure curve with the actual water hammer pressure curve collected in the field. The objective function is to minimize the root mean square error between the simulated and measured pressure values ​​at the wellhead. The R, C, and I parameters are iteratively adjusted until convergence. In other words, the optimal algorithm is used to find the one that best matches the simulated and actual water hammer pressure curves. Matching conditions are as follows Figure 2 As shown.

[0063] Using the bottom boundary calculation model and the characteristic line equation C of the bottom node + It can solve for the head and flow rate at the bottom of the well at different times, and then determine the water hammer pressure at the bottom of the well at different times. Combined with the minimum ground stress, it can obtain... .

[0064] It should be noted that the wellhead boundary solution and the wellbottom boundary solution are not two independent calculation processes, but rather two boundary conditions processed sequentially within each time step. They are tightly coupled with the nodal flow field inside the wellbore through characteristic line equations: the wellhead boundary provides the system excitation, and the wellbottom boundary provides the fracture impedance reflection; together, they determine the transient response of the entire wellbore-fracture system.

[0065] In obtaining inclusion After determining the target group control parameter values, the reference fracture size under the target group control parameter values ​​is determined based on the target group control parameter values, the bottom hole boundary calculation model, the wellbore flow rate model, and the fracture size calculation model, including: Based on the target group circuit parameter values ​​( The bottom boundary calculation model and wellbore flow model are used to determine the bottom fluid pressure time history at different times under the target group circuit parameter values. The average value of the bottom hole fluid pressure time history at different times is determined as the average bottom hole pressure; The average net pressure within the fracture is determined based on the average bottom hole pressure and the minimum ground stress. The reference size of the fracture is determined based on the average net pressure within the fracture, the rock mechanics parameters of the target well, and the fracture size calculation model.

[0066] Specifically, as mentioned above, will Substituting the values ​​into the bottom-hole boundary calculation model, the time history of the bottom-hole fluid pressure can be calculated after the pump is stopped. The average pressure at the bottom of the well is taken as the average pressure during the stable period after the fluctuations have subsided. ; Calculate the average net pressure within the crack The length, height, and width of the cracks are calculated based on the average net pressure and rock mechanics parameters.

[0067] Among them, determining the average bottom hole pressure as the average bottom hole pressure by the time history of the bottom hole fluid pressure at different times includes: The average bottom hole pressure is determined using the following formula:

[0068] in, This represents the initial moment when the pressure stabilizes after the water hammer wave decays. The set time window length.

[0069] Based on the average net pressure within the fracture, the set rock mechanics parameters of the target well, and the fracture size calculation model, the reference size of the fracture is determined, including: The length, height, and width of the cracks are determined using the following formulas:

[0070]

[0071] in, The crack half-length is expressed in units of 1 / 2. ; The crack height is expressed in units of... ; The effective thickness of the reservoir in the target well, in units of ; The crack width is expressed in units of 1000 mm. ; The Young's modulus of the rock in the target well, in units of ; The Poisson's ratio of the rock in the target well is dimensionless. The average net pressure within the crack, in units of ; This is a near-wellbore friction correction function related to the target resistance value in the target group circuit parameter values.

[0072] In one optional embodiment, the crack size prediction model includes a first crack size prediction sub-model, a second crack size prediction sub-model, and a fusion sub-model. Based on this, the operation of the crack size prediction model includes: The actual water hammer pressure at the wellhead at different times, the range of values ​​of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the bottom boundary calculation model, and the fracture size calculation model are respectively input into the first fracture size prediction sub-model and the second fracture size prediction sub-model, so that the first fracture size prediction sub-model and the second fracture size prediction sub-model respectively output the fracture size vector under the target group control parameter values. The crack size vectors of the first crack size prediction sub-model and the second crack size prediction sub-model are fused by the fusion sub-model, and the reference size is determined based on the fusion result.

[0073] In other words, the first fracture size prediction sub-model and the second fracture size prediction sub-model respectively execute the steps in the aforementioned embodiments to predict the fracture size of the target well, and finally fuse them through the fusion sub-model to obtain the final result.

[0074] Understandably, due to The determination criteria are fixed: they need to match the actual water hammer pressure curve at the wellhead. Therefore, the two fracture size prediction sub-models determine the... They are similar, meaning the difference is small. The above fusion strategy can be a single-choice fusion or a weighted fusion strategy, etc., without specific limitations.

[0075] Optionally, the first crack size prediction sub-model and the second crack size prediction sub-model can be, for example, an LSTM model and a random forest model, and are not limited thereto.

[0076] During the training process of the above-mentioned fracture size prediction model, data from different wells (actual water hammer pressure curves, rock mechanics parameters such as Young's modulus included in the control parameters) can be collected as training data. The monitoring signal corresponding to each well is the actual fracture size information of that well (which can be calculated in other ways).

[0077] During training, the dataset is divided into a training set and a validation set. An LSTM model and a random forest model are trained independently on the training set, and their hyperparameters are tuned. Then, the trained LSTM model and random forest model are used to make predictions on the validation set, resulting in two columns of predicted feature vectors. A linear fusion model is then used to weight and fuse these two columns of predicted feature vectors, outputting the final crack morphology prediction result.

[0078] In summary, the solution provided in this application uses an equivalent circuit model of resistor R, inductor L, and capacitor C to accurately characterize the dynamic mechanical properties of cracks. It also integrates the advantages of LSTM time-series processing and random forest feature analysis to achieve high-precision real-time prediction of crack morphology during fracturing. This effectively breaks through the comprehensive bottlenecks of existing technologies in terms of dynamic response analysis, computational efficiency, and engineering applicability.

[0079] Example 2: Another embodiment of this application relates to a crack size prediction device. The implementation details of the crack size prediction device in this embodiment are described below. The following implementation details are provided for ease of understanding and are not essential for implementing this solution. A schematic diagram of the crack size prediction device in this embodiment can be seen as follows: Figure 3 As shown, it includes: a determination module 31, a data acquisition module 32, and a processing module 33.

[0080] The determination module 31 is used to determine the wellbore flow rate model, wellhead boundary calculation model, bottom well boundary calculation model and fracture size calculation model. The bottom well boundary calculation model equates the fracture dynamic response to a series circuit structure containing resistance, capacitance and inductance. Resistance represents the ratio of pressure difference to flow rate to maintain fracture opening, capacitance represents the rate of change of fracture volume with pressure, and inductance represents the inertial effect of fluid acceleration or deceleration within the fracture. The acquisition module 32 is used to acquire the actual water hammer pressure value at the wellhead of the target well at different times after the pump is stopped; Processing module 33 is used to input the actual water hammer pressure values ​​at the wellhead at different times, the value ranges of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the well bottom boundary calculation model, and the fracture size calculation model into the fracture size prediction model, so that the fracture size prediction model performs the following process: determining multiple sets of control parameter values ​​according to the value ranges of the various control parameters; determining the simulated water hammer pressure values ​​at the wellhead at different times under the multiple sets of control parameter values ​​according to the wellbore flow model, the wellhead boundary calculation model, and the well bottom boundary calculation model; determining the target set of control parameter values ​​from the multiple sets of control parameter values ​​that match the simulated water hammer pressure values ​​at the wellhead at different times with the actual water hammer pressure values; determining the reference size of the fracture under the target set of control parameter values ​​according to the target set of control parameter values, the well bottom boundary calculation model, the wellbore flow model, and the fracture size calculation model; wherein, the various control parameters include resistance, capacitance, and inductance.

[0081] It is worth mentioning that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed in this application; however, this does not mean that other units are absent in this embodiment.

[0082] Example 3: Another embodiment of this application relates to an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the crack size prediction method in the above embodiments.

[0083] The memory and processor are connected via a bus, which can include any number of interconnecting buses and bridges, connecting various circuits of one or more processors and memories. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and will not be described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over the wireless medium via an antenna, which further receives data and transmits it to the processor.

[0084] The processor manages the bus and general processing, and also provides various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory is used to store data used by the processor during operation.

[0085] Example 4: Another embodiment of this application relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the above-described crack size prediction method.

[0086] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0087] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing this application, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of this application.

[0088] Example 5: Another embodiment of this application relates to a computer program product, including a computer program that, when executed by a processor, implements the crack size prediction method described above.

[0089] According to embodiments of this disclosure, the process described above with reference to the flowchart can be implemented as a computer program product, which includes a computer program that, when executed by a processor, implements the above-described geophysical lithology interpretation and determination method.

[0090] In some possible implementations, various aspects of this disclosure may also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the “Exemplary Methods” section of this specification according to various exemplary embodiments of this disclosure.

[0091] In practical implementation, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

Claims

1. A method for predicting crack size, characterized in that, include: The wellbore flow rate model, wellhead boundary calculation model, bottom boundary calculation model, and fracture size calculation model are determined. The bottom boundary calculation model equates the fracture dynamic response to a series circuit structure containing resistance, capacitance, and inductance. Resistance represents the ratio of pressure difference to flow rate that maintains fracture opening, capacitance represents the rate of change of fracture volume with pressure, and inductance represents the inertial effect of fluid acceleration or deceleration within the fracture. Collect the actual water hammer pressure values ​​at the wellhead of the target well at different times after the pump is stopped; The actual water hammer pressure values ​​at the wellhead at different times, the value ranges of various control parameters of the target well, the wellbore flow rate model, the wellhead boundary calculation model, the bottom well boundary calculation model, and the fracture size calculation model are input into the fracture size prediction model, so that the fracture size prediction model performs the following process: Based on the range of values ​​of the various control parameters, multiple sets of control parameter values ​​are determined, wherein the various control parameters include resistance, capacitance and inductance; Based on the wellbore flow model, the wellhead boundary calculation model, and the well bottom boundary calculation model, the simulated water hammer pressure values ​​at the wellhead at different times under the multiple sets of control parameter values ​​are determined respectively. From the multiple sets of control parameter values, determine the target set of control parameter values ​​that match the simulated water hammer pressure value at the wellhead at different times with the actual water hammer pressure value; Based on the target group control parameter values, the well bottom boundary calculation model, the wellbore flow rate model, and the fracture size calculation model, the reference size of the fracture under the target group control parameter values ​​is determined.

2. The method according to claim 1, characterized in that, The calculation model for the bottom boundary of the well is represented as follows: in, The instantaneous driving pressure difference at the bottom of the well is defined as the bottom-hole fluid pressure time history. With minimum ground stress Difference: , This represents the flow rate at the bottom of the well.

3. The method according to claim 2, characterized in that, The step of determining the reference size of the fracture under the target group control parameter values ​​based on the target group control parameter values, the bottom hole boundary calculation model, the wellbore flow rate model, and the fracture size calculation model includes: Based on the target group circuit parameter values, the bottom hole boundary calculation model, and the wellbore flow model, the bottom hole fluid pressure time history at different times is determined under the target group circuit parameter values. The target group control parameter values ​​include the target group circuit parameter values, which include target resistance, target capacitance, and target inductance values. The average value of the bottom hole fluid pressure time history at different times is determined as the average bottom hole pressure; The average net pressure within the fracture is determined based on the average bottom hole pressure and the minimum ground stress. The reference size of the fracture is determined based on the average net pressure within the fracture, the rock mechanics parameters of the target well, and the fracture size calculation model.

4. The method according to claim 3, characterized in that, Determining the average bottom hole fluid pressure time history at different times as the average bottom hole pressure includes: The average bottom hole pressure is determined using the following formula: in, This represents the initial moment when the pressure stabilizes after the water hammer wave decays. The set time window length.

5. The method according to claim 3, characterized in that, The step of determining the reference size of the fracture based on the average net pressure within the fracture, the set rock mechanics parameters of the target well, and the fracture size calculation model includes: The length, height, and width of the cracks are determined using the following formulas: in, The crack half-length is expressed in units of 1 / 2. ; The crack height is expressed in units of... ; The effective thickness of the reservoir in the target well, in units of ; The crack width is expressed in units of 1000 mm. ; The Young's modulus of the rock in the target well, in units of ; The Poisson's ratio of the rock in the target well is dimensionless. The average net pressure within the crack, in units of ; This is a near-wellbore friction correction function related to the target resistance value in the target group circuit parameter values.

6. The method according to claim 1, characterized in that, The wellhead boundary calculation model is as follows: in, Initial displacement, in units of ; This refers to the time taken to stop the pump.

7. The method according to claim 1, characterized in that, The wellbore model uses the method of characteristics to discretize and solve the water hammer equation. The recursive formula for the pressure head H and flow rate Q at different nodes inside the wellbore is as follows: in, The pressure head is the unit. ; Volumetric flow rate, unit: ; Pipe impedance coefficient, unit: ; This is the pipe resistance coefficient, in units of... ; superscript , Representing the current time and the next calculation time respectively; subscript , , This indicates the spatial location node number after discretization along the wellbore pipe.

8. The method according to any one of claims 1-7, characterized in that, The crack size prediction model includes a first crack size prediction sub-model, a second crack size prediction sub-model, and a fusion sub-model. The working process of the crack size prediction model includes: The actual water hammer pressure at the wellhead at different times, the range of values ​​of various control parameters of the target well, the wellbore flow model, the wellhead boundary calculation model, the bottom boundary calculation model, and the fracture size calculation model are respectively input into the first fracture size prediction sub-model and the second fracture size prediction sub-model, so that the first fracture size prediction sub-model and the second fracture size prediction sub-model respectively output the fracture size vector under the target group control parameter values; The crack size vectors of the first crack size prediction sub-model and the second crack size prediction sub-model are fused by a fusion sub-model, and the reference size is determined based on the fusion result.

9. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the crack size prediction method as described in any one of claims 1-8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the crack size prediction method according to any one of claims 1-8.

Images