Method, device and electronic equipment for determining cleanliness of cuttings bed destroyer

Abstract

This specification provides a method, device and electronic device for determining the cleanliness of a cuttings bed breaker, the method comprising: obtaining a value combination of each drilling parameter and the corresponding cleanliness as a data to obtain multiple sample data; dividing the multiple sample data into a training sample set and a verification sample set, and using the training sample set to train a network model; during training, each drilling parameter in each sample data is used as the input of the network model, and the cleanliness is used as the output of the network model; the network model is verified based on the verification set; when the verification result meets the requirements, each drilling parameter under the target working condition in actual production is input into the network model to obtain the cleanliness of the cuttings bed breaker under the target working condition. This solution calculates the cleanliness through a network model, which has the advantages of accurate calculation of other methods on the one hand, and greatly reduces the calculation efficiency on the other hand, and the network model is more suitable for the rapid optimization design of on-site drilling parameters.

Description

Technical Field

[0001] The present application relates to the field of seismic exploration technology, and in particular to a method, device and electronic equipment for determining the cleanliness of a cuttings bed breaker. Background Art

[0002] During the drilling of extended-reach wells and long horizontal wells, the drill bit breaks the rock to form fine cuttings particles, which are easily accumulated at the bottom of the wellbore annulus under the action of gravity and easily accumulate to form a cuttings bed. If the cuttings bed cannot be removed in time, it will cause problems such as increased friction torque and drill sticking. As an auxiliary wellbore cleaning tool, the cuttings bed breaker can remove the cuttings bed under the dual action of mechanical and hydraulic forces, thereby cleaning the wellbore.

[0003] In order to facilitate the evaluation of the actual use effect of the cuttings bed breaker and provide a basis for the optimization of its structure, it is necessary to establish scientific and reasonable rock cleaning effect evaluation indicators. There are many existing evaluation methods for wellbore cleaning degree, including the following evaluation methods:

[0004] The first evaluation method is the content of cuttings in the annulus. The thickness of the cuttings bed and the volume fraction of the cuttings are used as evaluation indicators. When the thickness or volume fraction is lower than the critical value, it can be considered that the wellbore is in good cleanliness. However, the distribution of cuttings at the bottom of the annulus is uneven. This evaluation method is difficult to fully consider the impact of bottom cuttings deposition, and there is a certain deviation in the evaluation.

[0005] The second type of evaluation method is the critical conditions for the formation of cuttings beds, which use critical migration velocity, suspended migration velocity, etc. as evaluation indicators, but there is no specific evaluation standard. It is generally believed that when the annular drilling fluid flow rate exceeds the critical velocity, the wellbore cleaning effect is better.

[0006] The third evaluation method is the geometric properties of the cuttings. It is believed that when the returned cuttings are of different sizes, the drilling fluid has a better rock-carrying effect. This method relies too much on field experience and the evaluation criteria are unclear.

[0007] Existing wellbore cleaning evaluation methods have their limitations and cannot obtain data such as the actual cuttings bed thickness and cuttings particle shape downhole. Therefore, they cannot be used directly to evaluate the rock cleaning effect of the cuttings bed breaker.

[0008] Considering the rock cleaning principle of the cuttings bed breaker, the inventor proposed to construct a drilling simulation model through numerical simulation method to calculate the rock cleaning effect evaluation index: cleanliness. However, this cleanliness calculation method requires re-modeling each time, and the turbulence model is relatively complex, the calculation efficiency is low, which is not conducive to promotion and application in actual construction. Summary of the invention

[0009] The purpose of the present application is to provide a method, device and electronic equipment for determining the cleanliness of a cuttings bed breaker, so as to improve the accuracy of determining the cleanliness of the cuttings bed breaker.

[0010] To solve the above technical problems, the first aspect of the present specification provides a method for determining the cleanliness of a cuttings bed breaker, comprising: obtaining the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter, and taking the value combination of each drilling parameter and the corresponding cleanliness as one data to obtain multiple sample data; dividing the multiple sample data into a training sample set and a verification sample set, and using the training sample set to train a network model; during training, each drilling parameter in each sample data is used as the input of the network model, and the cleanliness is used as the output of the network model; the network model is verified based on the verification set; when the verification result meets the requirements, each drilling parameter under the target working condition in actual production is input into the network model to obtain the cleanliness of the cuttings bed breaker under the target working condition.

[0011] In some embodiments, the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter is obtained, including: for different value combinations of each drilling parameter, numerically simulating the annular two-phase flow when the cuttings bed breaker is working through computational fluid dynamics software; and calculating the cleanliness based on the flow field parameters obtained from the simulation.

[0012] In some embodiments, the cleanliness is calculated as:

[0013]

[0014] Among them, Q is the cleanliness, A is the annular cross-sectional area, C is the volume fraction of cuttings, v1 is the axial velocity component after the fluid and cuttings are mixed, v2 is the velocity component perpendicular to the axis after the annular fluid and cuttings are mixed, t1 is the moment when the simulation process tends to steady state, t2 is the moment when the numerical simulation ends, and D P D is the drill pipe diameter, h is the wellbore diameter.

[0015] In some embodiments, the drilling parameters include: mechanical parameters, wellbore geometry parameters, and fluid parameters.

[0016] In some embodiments, a validation set is used to validate the network model, including: for each sample data, calculating the error value between the output value of the network model and the cleanliness in the sample data; dividing the error value interval into multiple sub-intervals, and counting the number of error values ​​in each sub-interval; judging whether the quantity distribution of the error values ​​conforms to the normal distribution; if so, calculating the correlation coefficient between the output value of the network model and the cleanliness in the sample data; if the correlation coefficient is greater than a predetermined threshold, determining that the network model has been validated.

[0017] According to a second aspect of the present specification, there is provided a device for determining the cleanliness of a cuttings bed breaker, comprising: an acquisition unit, for acquiring the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter, and taking the value combination of each drilling parameter and the corresponding cleanliness as one data to obtain a plurality of sample data; a training unit, for dividing the plurality of sample data into a training sample set and a verification sample set, and using the training sample set to train a network model; during training, each drilling parameter in each sample data is used as an input of the network model, and the cleanliness is used as an output of the network model; a verification unit, for verifying the network model based on the verification set; and a processing unit, for inputting each drilling parameter under the target working condition in actual production into the network model when the verification result meets the requirements, so as to obtain the cleanliness of the cuttings bed breaker under the target working condition.

[0018] In some embodiments, the acquisition unit includes: a simulation subunit, which is used to perform numerical simulation of the annular two-phase flow when the cuttings bed breaker is working by computational fluid dynamics software for different value combinations of each drilling parameter; and a first calculation subunit, which is used to calculate the cleanliness based on the flow field parameters obtained by simulation.

[0019] In some embodiments, the verification unit includes: a second calculation subunit, used to calculate the error value between the output value of the network model and the cleanliness in the sample data for each sample data; a statistical subunit, used to divide the value interval of the error value into multiple sub-intervals, and count the number of error values ​​in each sub-interval; a judgment subunit, used to judge whether the quantitative distribution of the error values ​​conforms to the normal distribution; a third calculation subunit, used to calculate the correlation coefficient between the output value of the network model and the cleanliness in the sample data if yes; a determination subunit, used to determine that the network model verification has passed if the correlation coefficient is greater than a predetermined threshold.

[0020] The third aspect of this specification provides an electronic device, comprising: a memory and a processor, the processor and the memory are communicatively connected to each other, the memory stores computer instructions, and the processor implements the steps of any one of the methods described in the first aspect by executing the computer instructions.

[0021] A fourth aspect of the present specification provides a computer storage medium, wherein the computer storage medium stores computer program instructions, and when the computer program instructions are executed, the steps of any one of the methods described in the first aspect are implemented.

[0022] The method, device and electronic device for determining the cleanliness of a cuttings bed breaker provided in this specification first obtain the cleanliness of the cuttings bed breaker under different value combinations of various drilling parameters through other methods to form multiple sample data to provide data support for the training of the network model; train the network model, and when the fitting performance of the network model is good, use the network model instead of other methods to directly calculate the cleanliness. Since the data set of the network model comes from other methods and the network model has good fitting performance, the calculation of cleanliness through the network model has the advantages of accurate calculation of other methods on the one hand, and greatly reduces the calculation efficiency on the other hand (for example, the calculation complexity and amount of the network model are far less than the calculation of cleanliness by constructing a drilling model using a numerical model), and the network model is more suitable for the rapid optimization design of on-site drilling parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the implementation methods of the present application or the technical solutions in the prior art, the drawings required for use in the implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some implementation methods recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0024] Figure 1 A flow chart showing a method for determining the cleanliness of a cuttings bed breaker provided in this specification is shown;

[0025] Figure 2 A schematic diagram of the operation of a spiral streamline blade cuttings bed breaker is shown;

[0026] Figure 3 A schematic diagram of the geometric model dimensions of a cuttings bed breaker is shown;

[0027] Figure 4 A schematic diagram of the structure of a neural network model is shown;

[0028] Figures 5 to 7 A schematic diagram showing that the quantitative distribution of error values ​​conforms to a normal distribution;

[0029] Figures 8 to 10 The regression graph of network model training and validation results is shown;

[0030] Figures 11 to 13 Regression plot of generalization ability test results;

[0031] Fig.14 A principle block diagram of a device for determining the cleanliness of a cuttings bed destroyer provided in this specification is shown;

[0032] Fig.15A schematic diagram of the structure of an electronic device provided in this specification is shown. DETAILED DESCRIPTION

[0033] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0034] This specification provides a method for determining the cleanliness of a cuttings bed breaker, such as Figure 1 As shown, the following steps are included:

[0035] S10: obtaining the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter, and taking the value combination of each drilling parameter and the corresponding cleanliness as one data to obtain a plurality of sample data.

[0036] The drilling parameters may include mechanical parameters, wellbore geometry parameters, fluid parameters, etc. Among them, the mechanical parameters may include mechanical drilling speed, drill pipe drilling speed, etc., the wellbore geometry parameters may include wellbore size, eccentricity, etc., and the fluid parameters may include displacement, rheological parameters, etc.

[0037] The values ​​of each drilling parameter can be manually selected to form different value combinations, and then each value combination is input into the drilling simulation model to obtain the corresponding cleanliness.

[0038] It should be noted that cleanliness is not a data that can be obtained through simple observation, but requires the construction of a complex drilling simulation model and calculation through numerical simulation.

[0039] Specifically, due to the complex structure of the cuttings bed breaker, the existing theoretical model cannot directly obtain the annular flow field distribution. Therefore, it is necessary to first carry out numerical simulation analysis of the flow field of the cuttings bed breaker, on this basis, the cleanliness can be calculated to evaluate the rock cleaning effect of the cuttings bed breaker.

[0040] The velocity inlet and pressure outlet boundary conditions are set at the inlet and outlet of the simulated well section respectively. The Realizable k-ε turbulence model and the Euler-Euler two-fluid model are used. The solid-liquid two-phase flow consisting of drilling fluid and uniformly dispersed cuttings with consistent density enters the calculation domain vertically from the annulus inlet (such as Figure 2The CFD commercial software Ansys Fluent 18.0 was used for simulation calculation, and the residual was set to 10-3, the transient calculation time step was 1×10-3s, the number of iterations was set to 10000 times, the actual flow simulation time was 10s, and the result data was automatically saved every 0.1s during the simulation process.

[0041] There are many factors that affect the cleanliness of the cuttings bed breaker, including drilling fluid displacement, drill pipe speed, mechanical penetration rate, wellbore size, drilling fluid rheological properties, eccentricity, etc. The drilling fluid rheological properties and eccentricity can be assumed to be constants. Different combinations of rotation speed, drilling fluid displacement, mechanical penetration rate and wellbore size are considered, totaling 325 groups. The established value combination scheme is shown in Table 1. The structure of the cuttings bed breaker used in the drilling model can be as follows: Figure 3 As shown, the structural geometric parameters can be as shown in Table 2, and other drilling parameters can be as shown in Table 3.

[0042] Table 1 Cleanliness calculation scheme (a) Wellbore size: 8-1 / 2"

[0043]

[0044] (b) Wellbore size: 12-1 / 4"

[0045]

[0046] (c) Wellbore size: 17-1 / 2"

[0047]

[0048] Table 2 Geometric parameters of streamlined blade cuttings bed breaker

[0049]

[0050] Table 3 Geometric parameters and operating conditions in numerical simulation

[0051]

[0052] Cleanliness is an important indicator for evaluating cleaning effect. The calculation formula is:

[0053]

[0054] Where, Q is cleanliness, dimensionless; A is annular cross-sectional area, m 2 ; C—cuttings volume fraction, dimensionless; v1—axial velocity component after the fluid and cuttings are mixed, m / s; v2—the velocity component perpendicular to the axis after the annular fluid and cuttings are mixed, that is, the lateral component, m / s; t1—the moment when the simulation process tends to a steady state, t2—the moment when the numerical simulation ends. Figure 2 As shown, D P D is the drill pipe diameter, mm h is the borehole diameter, mm.

[0055] The constructed drilling model and the calculation formula for cleanliness are not the focus of this solution. For details, please refer to the patent text that the inventor has authorized.

[0056] S20: Divide the plurality of sample data into a training sample set and a validation sample set, and use the training sample set to train the network model; during training, use each drilling parameter in each sample data as an input of the network model, and use the cleanliness as an output of the network model.

[0057] In theory, the network model can adopt any type of network model.

[0058] BP (Back Propagation) neural network is a widely used artificial neural network and a variant of feedforward neural network. The principle of BP neural network is based on the error back propagation algorithm, which transmits error information from the output layer to the input layer layer by layer, calculates the weight error of each layer, and updates the weight of each layer in the network, so that the network gradually improves the prediction of the real output in multiple iterations.

[0059] In BP neural network, the number of hidden layers has an important impact on the performance of neural network: when the number of hidden layers is 0, that is, there are only input layer and output layer, the network can only represent linearly separable functions or decisions; when the number of hidden layers is 1, it can fit the continuous mapping function from one finite space to another finite space, but the representation ability is still limited; if the BP network has two hidden layers, the representation ability is significantly enhanced, and under appropriate circumstances, it can represent complex decision boundaries of arbitrary precision; when the number of hidden layers is more than 2, the extra hidden layers can learn more complex descriptions. Since the data set involved in this scheme is not very complex, 1 hidden layer has met the requirements. Increasing the number of hidden layers will enhance the nonlinear mapping ability of the model and have higher accuracy on the training set, but it will also increase the computational complexity, which may cause overfitting and affect the generalization ability.

[0060] When training a neural network, different input variables often have different orders of magnitude, which will affect the training of the network, making the network's learning speed slower, and may even fall into a local optimal solution and fail to obtain better results. To avoid this, we use normalization. You can use the mapminmax function in MATLAB software, which can scale the input data to the interval of [-1,1] so that all input data have the same order of magnitude, avoiding the impact of data that is too large or too small on the network.

[0061] When building a neural network, the empirical formula can generally be used to determine the number of nodes in the hidden layer:

[0062]

[0063] Among them, k is the number of hidden layer nodes, m is the number of input layer nodes, n is the number of output layer nodes, a is an adjustment constant ranging from 1 to 10. The number of hidden layer nodes can be set to 8.

[0064] Neural network models used to predict cleanliness such as Figure 4 As shown, there is 1 hidden layer with 8 nodes. In order to ensure the generalization ability of the neural network, the data set division method is adopted, in which 70% of the numerical simulation results are used as training sample sets, 15% are used as validation sample sets, and the remaining 15% are used as test sample sets (i.e., test sets) to evaluate the accuracy of the model. In the training of the neural network, the "Levenberg-Marquardt" algorithm is used. This algorithm does not need to directly calculate the Hessian matrix, which can effectively avoid the overfitting phenomenon caused by the algorithm falling into local extreme values.

[0065] S30: Validating the network model based on the validation set.

[0066] The "verification" here may be the verification of the fitting performance of the network model. Specifically, step S30 may include the following steps:

[0067] S31: For each sample data, calculate the error value between the output value of the network model and the cleanliness in the sample data.

[0068] S32: Divide the error value interval into multiple sub-intervals, and count the number of error values ​​in each sub-interval.

[0069] S33: Determine whether the quantity distribution of the error values ​​conforms to the normal distribution.

[0070] S34: In the case of yes, calculate the correlation coefficient between the output value of the network model and the cleanliness in the sample data.

[0071] S35: When the correlation coefficient is greater than a predetermined threshold, it is determined that the network model verification has passed.

[0072] Figures 5 to 7 A schematic diagram showing that the number distribution of error values ​​conforms to the normal distribution. When the training effect of the network model is good, the number distribution of error values ​​is close to the normal distribution, and the standard deviation is small. This scheme uses the normal distribution as a verification condition for the training effect of the network model.

[0073] Figures 8 to 10The regression diagram of the network model training and verification results is shown, where Figures 8 to 10 In the four sub-graphs in , the labels Training, Validation, Test, and All above respectively indicate that the data comes from the training set, validation set, test set, and all data (all data includes the data of the training set, validation set, and test set); the horizontal axis represents the result value of the numerical simulation, the vertical axis represents the result of the network model, and the R value represents the correlation coefficient between the output value of the network model and the cleanliness in the training data. The correlation coefficient is used to evaluate the predictive ability and degree of fit of the model. The closer the R value is to 1, the closer the relationship between the output value of the network model and the cleanliness in the training data is, and the stronger the predictive ability of the network model is; the closer the R value is to 0, the greater the randomness of the relationship between the two, and the weaker the predictive ability of the network model. The results show that for the three wellbore sizes, the R values ​​are close to 1 and the regression lines are close to 45°, which proves that the network model has a good fitting effect on the training set and the validation set. On this basis, the generalization ability of the network model can be further tested. For example, about 20 groups of untrained data can be selected for three different wellbore sizes, and the generalization ability of the network model can be tested. The test results are as follows Figures 11 to 13 As shown, the angle of the regression line is close to 45°, which indicates that the network model has good generalization ability.

[0074] The normal distribution in the verification method provided in this specification requires the support of a large amount of data. Usually, the amount of data in the verification sample set is insufficient. In this case, all the data in the training sample set and the verification sample set can be used to verify the normal distribution.

[0075] S40: When the verification result meets the requirements, various drilling parameters under the target working conditions in actual production are input into the network model to obtain the cleanliness of the cuttings bed breaker under the target working conditions.

[0076] If the verification result meets the requirements, it means that the prediction result of the network model is very close to the cleanliness data obtained by the structural drilling model through numerical simulation calculation. Therefore, the network model can be used to obtain the cleanliness instead of the numerical simulation calculation in the future.

[0077] This scheme also carried out numerical simulation and network model prediction for the cleanliness of three different wellbore sizes of 8-1 / 2", 12-1 / 4", and 17-1 / 4". The cleanliness results obtained by the two methods were compared, and the specific results are shown in Table 4. The results show that the accuracy of the network model is greater than 95%. Therefore, it can be considered that the network model is reliable and can be used to calculate the cleanliness of the cuttings bed breaker.

[0078] Table 4 Comparison of cleanliness results

[0079]

[0080] Note: The calculation of relative error is based on the numerical simulation results.

[0081] The method for determining the cleanliness of a cuttings bed breaker provided in this specification first obtains the cleanliness of a cuttings bed breaker under different combinations of values ​​of various drilling parameters through other methods to form multiple sample data to provide data support for the training of a network model; the network model is trained, and when the fitting performance of the network model is good, the cleanliness is directly calculated using the network model instead of other methods. Since the data set of the network model comes from other methods and the network model has good fitting performance, the calculation of cleanliness through the network model has the advantages of accurate calculation of other methods on the one hand, and greatly reduces the calculation efficiency on the other hand (for example, the calculation complexity and amount of the network model are far less than the calculation of cleanliness by constructing a drilling model using a numerical model), and the network model is more suitable for the rapid optimization design of on-site drilling parameters.

[0082] This specification provides a device for determining the cleanliness of a cuttings bed breaker, which can be used to implement the above-mentioned method for determining the cleanliness of a cuttings bed breaker. Fig.14 As shown, the device includes an acquisition unit 10, a training unit 20, a verification unit 30 and a processing unit 40.

[0083] The acquisition unit 10 is used to acquire the cleanliness of the cuttings bed breaker under different value combinations of various drilling parameters, and to use the value combinations of various drilling parameters and the corresponding cleanliness as one data to obtain a plurality of sample data.

[0084] The training unit 20 is used to divide the multiple sample data into a training sample set and a verification sample set, and use the training sample set to train the network model; during training, each drilling parameter in each sample data is used as the input of the network model, and the cleanliness is used as the output of the network model.

[0085] The verification unit 30 is used to verify the network model based on the verification set.

[0086] The processing unit 40 is used to input various drilling parameters under the target working conditions in actual production into the network model to obtain the cleanliness of the cuttings bed breaker under the target working conditions when the verification result meets the requirements.

[0087] In some embodiments, the acquisition unit 10 includes a simulation subunit and a first calculation subunit.

[0088] The simulation subunit is used to numerically simulate the two-phase flow in the annulus when the cuttings bed breaker is working according to different combinations of drilling parameters using computational fluid dynamics software.

[0089] The first calculation subunit is used to calculate the cleanliness based on the flow field parameters obtained by simulation.

[0090] In some embodiments, the verification unit 30 includes a second calculation subunit, a statistics subunit, a judgment subunit, a third calculation subunit and a determination subunit.

[0091] The second calculation subunit is used to calculate, for each sample data, an error value between an output value of the network model and the cleanliness in the sample data.

[0092] The statistical subunit is used to divide the value interval of the error value into multiple sub-intervals, and count the number of error values ​​in each sub-interval.

[0093] The judging subunit is used to judge whether the quantity distribution of the error values ​​conforms to the normal distribution.

[0094] The third calculation subunit is used to calculate the correlation coefficient between the output value of the network model and the cleanliness in the sample data when yes.

[0095] The determination subunit is used to determine whether the network model has passed the verification when the correlation coefficient is greater than a predetermined threshold.

[0096] For the detailed description and beneficial effects of the above-mentioned device, please refer to the corresponding method embodiment, which will not be repeated here.

[0097] The embodiment of the present invention further provides an electronic device, such as Fig.15 As shown, the electronic device may include a processor 1501 and a memory 1502, wherein the processor 1501 and the memory 1502 may be connected via a bus or other means. Fig.15 The example of connecting through bus is taken in the following.

[0098] The processor 1501 may be a central processing unit (CPU). The processor 1501 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0099] The memory 1502 is a non-transitory computer-readable storage medium that can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the method for determining the cleanliness of a cuttings bed breaker in an embodiment of the present invention (for example, Fig.14 The processor 1501 executes various functional applications and data classification of the processor by running the non-transitory software programs, instructions and modules stored in the memory 1502, that is, the method for determining the cleanliness of the cuttings bed breaker in the above method embodiment is implemented.

[0100] The memory 1502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created by the processor 1501, etc. In addition, the memory 1502 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 1502 may optionally include a memory remotely arranged relative to the processor 1501, and these remote memories may be connected to the processor 1501 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0101] The one or more modules are stored in the memory 1502, and when executed by the processor 1501, the above-mentioned method for determining the cleanliness of the cuttings bed breaker is performed.

[0102] The specific details of the above electronic device can be understood by referring to the relevant descriptions and effects in the corresponding embodiments, which will not be repeated here.

[0103] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0104] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0105] Although the present specification embodiment provides the method operation steps as described in the embodiment or flow chart, more or less operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiment is only one way in the order of execution of many steps, and does not represent a unique execution order. When the device or terminal product in practice is executed, it can be executed in sequence or in parallel (such as a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment) according to the method shown in the embodiment or the accompanying drawings. The term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or equipment including a series of elements not only includes those elements, but also includes other elements not clearly listed, or also includes elements inherent to such process, method, product or equipment. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or equipment including the elements.

[0106] The present specification embodiments may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present specification embodiments may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0107] The above is only an example of the embodiment of the present specification and is not intended to limit the embodiment of the present specification. For those skilled in the art, the embodiment of the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiment of the present specification shall be included in the scope of the claims of the embodiment of the present specification.

Claims

1. A method for determining the cleanliness of a cuttings bed breaker, characterized in that: include: Obtaining the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter, and taking the value combination of each drilling parameter and the corresponding cleanliness as one data to obtain a plurality of sample data; The plurality of sample data are divided into a training sample set and a verification sample set, and the network model is trained using the training sample set; during the training, each drilling parameter in each sample data is used as an input of the network model, and the cleanliness is used as an output of the network model; Verifying the network model based on a verification set; When the verification result meets the requirements, various drilling parameters under the target working conditions in actual production are input into the network model to obtain the cleanliness of the cuttings bed breaker under the target working conditions; Obtain the cleanliness of the cuttings bed breaker under different combinations of drilling parameters, including: According to different combinations of drilling parameters, the two-phase flow in the annulus when the cuttings bed breaker is working is numerically simulated by computational fluid dynamics software. Calculate the cleanliness based on the flow field parameters obtained by simulation; The calculation formula for cleanliness is: Among them, Q is the cleanliness, A is the annulus cross-sectional area, c is the volume fraction of cuttings, v1 is the axial velocity component after the fluid and cuttings are mixed, v2 is the velocity component perpendicular to the axis after the annulus fluid and cuttings are mixed, t1 is the moment when the simulation process tends to a steady state, and t2 is the moment when the numerical simulation ends.

2. The method according to claim 1, characterized in that Drilling parameters include: mechanical parameters, wellbore geometry parameters, and fluid parameters.

3. The method according to claim 1, characterized in that The network model is verified using a verification set, including: For each sample data, the error value between the output value of the network model and the cleanliness in the sample data is calculated; Divide the error value interval into multiple sub-intervals, and count the number of error values ​​in each sub-interval; Determine whether the quantity distribution of error values ​​conforms to the normal distribution; If yes, calculate the correlation coefficient between the output value of the network model and the cleanliness in the sample data; When the correlation coefficient is greater than a predetermined threshold, it is determined that the network model has passed the verification.

4. A device for determining the cleanliness of a cuttings bed breaker, characterized in that: include: An acquisition unit, used to acquire the cleanliness of the cuttings bed breaker under different value combinations of each drilling parameter, and to use the value combination of each drilling parameter and the corresponding cleanliness as one data to obtain a plurality of sample data; A training unit, used to divide the plurality of sample data into a training sample set and a validation sample set, and train the network model using the training sample set; During training, each drilling parameter in each sample data is used as the input of the network model, and the cleanliness is used as the output of the network model; A verification unit, used to verify the network model based on a verification set; A processing unit, for inputting various drilling parameters under target working conditions in actual production into the network model to obtain the cleanliness of the cuttings bed breaker under the target working conditions when the verification result meets the requirements; The acquisition unit comprises: The simulation subunit is used to numerically simulate the two-phase flow in the annulus when the cuttings bed breaker is working by using computational fluid dynamics software for different combinations of drilling parameters; A first calculation subunit is used to calculate the cleanliness based on the flow field parameters obtained by simulation; The calculation formula for cleanliness is: Among them, Q is the cleanliness, A is the annulus cross-sectional area, c is the volume fraction of cuttings, v1 is the axial velocity component after the fluid and cuttings are mixed, v2 is the velocity component perpendicular to the axis after the annulus fluid and cuttings are mixed, t1 is the moment when the simulation process tends to a steady state, and t2 is the moment when the numerical simulation ends.

5. The device according to claim 4, characterized in that The verification unit comprises: A second calculation subunit is used to calculate, for each sample data, an error value between an output value of the network model and a cleanliness in the sample data; A statistical subunit, used to divide the error value interval into multiple sub-intervals, and count the number of error values ​​in each sub-interval; A judgment subunit, used to judge whether the quantity distribution of the error value conforms to the normal distribution; A third calculation subunit is used to calculate the correlation coefficient between the output value of the network model and the cleanliness in the sample data if yes; The determination subunit is used to determine whether the network model verification has passed when the correlation coefficient is greater than a predetermined threshold.

6. An electronic device, characterized in that: include: A memory and a processor, wherein the processor and the memory are communicatively connected to each other, the memory stores computer instructions, and the processor implements the steps of the method according to any one of claims 1 to 3 by executing the computer instructions.

7. A computer storage medium, characterized in that: The computer storage medium stores computer program instructions, and when the computer program instructions are executed, the steps of the method according to any one of claims 1 to 3 are implemented.

Images