Electromagnetic imaging method and device for formation while drilling

Abstract

The present invention discloses a method and device for electromagnetic imaging of formations while drilling, the method comprising: determining an inversion formation model for electromagnetic imaging of formations while drilling; constructing an objective function with a regularization term for the inversion formation model, and calculating the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term; using a pre-gradient matrix obtained by pre-training the gradient function, iteratively updating the formation parameters of the inversion formation model according to the pre-gradient matrix according to the number of inversion iterations, and determining the formation parameters for constructing an inversion map of the current formation based on the objective function to obtain the inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations. By combining inversion with a formation model obtained by machine learning, the inversion accuracy and global convergence ability are improved. The pre-gradient matrix obtained by pre-training is directly applied in the inversion, which significantly improves the inversion speed and real-time performance, and improves the inversion accuracy of complex formation parameters.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of geophysical exploration, and particularly to a method and device for electromagnetic imaging of formation while drilling. Background Art

[0002] With the increasing scarcity of oil and natural gas resources, the importance of accurately exploring the characteristics of underground formations has become more prominent. Traditional geophysical measurement methods have problems such as instability and insufficient accuracy under complex formation conditions. Especially during the process of logging while drilling, due to the anisotropic and non-linear characteristics of the formation, the analysis of the inductive logging response becomes more complex. Using existing inversion algorithms often depends on the initial model and is prone to falling into local optimal solutions, resulting in uncertainty in the electromagnetic imaging inversion results. Therefore, a more accurate method for electromagnetic imaging of formation while drilling is needed. Summary of the Invention

[0003] In view of the above problems, the embodiments of the present invention are proposed to provide a method and device for electromagnetic imaging of formation while drilling that overcome the above problems or at least partially solve the above problems.

[0004] According to one aspect of the embodiments of the present invention, a method for electromagnetic imaging of formation while drilling is provided, and the method includes:

[0005] Determine the inversion formation model for electromagnetic imaging of formation while drilling;

[0006] Construct an objective function with a regularization term for the inversion formation model, and calculate the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term;

[0007] Using the pre-gradient matrix pre-trained by the gradient function, according to the number of inversion iterations, iteratively update the formation parameters of the inversion formation model based on the pre-gradient matrix, and determine the formation parameters for constructing the inversion map of the current formation based on the objective function to obtain the inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations.

[0008] According to another aspect of the embodiments of the present invention, a device for electromagnetic imaging of formation while drilling is provided, and it includes:

[0009] A formation model module, adapted to determine the inversion formation model for electromagnetic imaging of formation while drilling;

[0010] A gradient module, adapted to construct an objective function with a regularization term for the inversion formation model, and calculate the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term;

[0011] An iterative inversion module, which is adapted to use a pre-gradient matrix obtained by pre-training with a gradient function, update the formation parameters of an inversion formation model iteratively according to the pre-gradient matrix according to the number of inversion iterations, and determine the formation parameters for constructing an inversion map of the current formation based on an objective function, so as to obtain an inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations.

[0012] According to another aspect of the embodiments of the present invention, there is provided a computing device, including: a processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0013] The memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned electromagnetic imaging method for formation while drilling.

[0014] According to still another aspect of the embodiments of the present invention, there is provided a computer storage medium, in which at least one executable instruction is stored, and the executable instruction causes a processor to execute the operations corresponding to the above-mentioned electromagnetic imaging method for formation while drilling.

[0015] According to yet another aspect of the embodiments of the present invention, there is provided a computer program product, including at least one executable instruction, and the executable instruction causes a processor to execute the operations corresponding to the above-mentioned electromagnetic imaging method for formation while drilling.

[0016] According to the electromagnetic imaging method and device for formation while drilling provided by the embodiments of the present invention, by combining inversion with an inversion formation model of machine learning, the inversion accuracy and global convergence ability are improved. Dynamic regularization can adaptively adjust the regularization parameter during the iteration process, enhancing the stability and accuracy of the inversion result. Directly applying the pre-gradient matrix obtained by pre-training in the inversion significantly improves the inversion speed and real-time performance, and improves the inversion accuracy of complex formation parameters.

[0017] The above description is only an overview of the technical solutions of the embodiments of the present invention. In order to be able to understand the technical means of the embodiments of the present invention more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of the embodiments of the present invention more obvious and understandable, the following specifically describes the specific embodiments of the present invention. Description of the Drawings

[0018] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the embodiments of the present invention. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0019] Figure 1Shows a flowchart of a formation electromagnetic imaging method while drilling according to an embodiment of the present invention;

[0020] Figure 2 Shows a schematic diagram of a seven - layer complex formation model;

[0021] Figure 3 Shows a schematic diagram of the parameters of the seven - layer complex formation model;

[0022] Figure 4 Shows a schematic diagram of the inversion map of the current formation obtained from the seven - layer complex formation model;

[0023] Figure 5 Shows a schematic structural diagram of a formation electromagnetic imaging device while drilling according to an embodiment of the present invention;

[0024] Figure 6 Shows a schematic structural diagram of a computing device according to an embodiment of the present invention. Detailed implementation manners

[0025] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present invention can be more thoroughly understood and the scope of the present invention can be fully conveyed to those skilled in the art.

[0026] Figure 1 Shows a flowchart of a formation electromagnetic imaging method while drilling according to an embodiment of the present invention, as Figure 1 shown, the method includes the following steps:

[0027] Step S101, determine the inversion formation model of the formation electromagnetic imaging while drilling.

[0028] To improve the inversion accuracy, an inversion method based on machine learning and deep learning can be used. Although machine learning and the like have strong global optimization capabilities, they lack physical interpretation and the generalizability of the model, resulting in limitations in practical applications. Based on the above problems, in this embodiment, machine learning is combined with electromagnetic wave measurement technology and gradient optimization during inversion, which can quickly and accurately invert the electrical parameters of each underground layer, provide high - resolution formation imaging results, thereby effectively overcoming the defects in traditional inversion technologies, improving the reconstruction ability of the inversion formation model, and providing more reliable data support for oil and gas exploration and geological engineering.

[0029] When performing logging-while-drilling inversion, constructing an accurate inversion formation model is the basis of inversion. According to formation parameters, such as the anisotropy and complexity of formation resistivity, different inversion formation models can be used to describe the electrical properties of the formation. For example, the isotropic model is applicable to the case where the relative changes in electrical parameters (such as resistivity and conductivity) of the formation in all directions are relatively small, which can simplify the calculation process. The anisotropic model is applicable to complex formation conditions, where the resistivity and conductivity in the formation vary with direction, and the anisotropic model can be closer to the actual formation situation. Specifically, it is not limited here according to the implementation situation.

[0030] Step S102: Construct an objective function with a regularization term for the inversion formation model, and calculate the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term.

[0031] The objective function is used to measure the gap between the logging data and the model prediction data. The smaller the value of the objective function, the higher the matching degree between the model and the actual situation. Construct the objective function of the inversion formation model based on the logging data and the model prediction data as follows:

[0032]

[0033] where θ is the formation parameter to be inverted, y i is the i-th actual logging data, and f(θ) is the predicted value of the inversion formation model.

[0034] Furthermore, when constructing the objective function of the inversion formation model, based on logging instruments, such as logging-while-drilling electromagnetic logging equipment, etc., collect the logging data of the well to be inverted. The logging data includes electrical parameters such as resistivity and conductivity. The logging data can be preprocessed first, and the objective function can be constructed based on the preprocessed logging data, which can provide more accurate formation parameters. The preprocessing includes data denoising and data compensation. During the data acquisition process, the original logging data may be affected by environmental noise and equipment errors, resulting in the appearance of random noise. Data denoising uses denoising algorithms (such as wavelet denoising, average correction algorithm, etc.) to process the data to eliminate these interferences, improve the signal-to-noise ratio of the data, and ensure the smoothness and reliability of the data. Data compensation aims to correct the logging data so that the response values at different depths are relatively uniform. Since the electromagnetic wave response characteristics of different logging instruments may vary at different depths, through compensation processing, the non-uniformity can be eliminated. For example, interpolation algorithms and statistical correction methods are used to smooth the logging data to ensure that the measured values at each layer can accurately reflect the electrical properties of the underground formation, etc. The above is for illustrative purposes and is specifically set according to the implementation situation and is not limited here.

[0035] To prevent the inversion formation model from overfitting, this embodiment introduces a regularization term λ||θ|| 2, the objective function with a regularization term is obtained , as follows:

[0036]

[0037] λ||θ|| 2 is the regularization term, which can suppress excessive fluctuations in formation parameters, including the regularization coefficient λ and the formation parameters θ. The regularization coefficient λ can control the weight of the regularization term.

[0038] To better adapt to different inversion stages, the regularization parameter λ is dynamically adjusted. It decays sequentially according to the inversion iteration times based on the preset initial regularization parameter, and the decay is determined by the decay factor. The adaptive adjustment is as follows:

[0039]

[0040] where λ 0 is the preset initial regularization parameter, β is the decay factor, and k is the inversion iteration times. A larger value of the regularization parameter λ is adopted at the initial stage of inversion to suppress overfitting of the inversion formation model. As the inversion iteration progresses, the regularization parameter λ is decreased, making the inversion formation model gradually fit the complex characteristics of the actual formation.

[0041] For the objective function with a regularization term, its gradient is calculated to obtain the gradient function with a regularization term, as follows:

[0042]

[0043]

[0044] where ▽J(θ) is the error gradient of the objective function. For the regularization term, L2 regularization can be adopted, and λ||θ|| 2 = 2θ. Substituting the error gradient and the regularization term into the gradient function, we get the following:

[0045]

[0046] Step S103: Using the pre-gradient matrix pre-trained by the gradient function, according to the inversion iteration times, the formation parameters of the inversion formation model are iteratively updated based on the pre-gradient matrix, and the formation parameters for constructing the inversion map of the current formation are determined based on the objective function, and the inversion map of the current formation is obtained.

[0047] According to the gradient function, through adaptive gradient optimization, the formation parameters are updated using the gradient descent method, as follows:

[0048]

[0049] Among them, k is the number of inversion iterations, and α is a preset learning rate that can adjust the update step size and is set according to the actual situation. The change direction of the formation parameters can be controlled according to the gradient function so as to adjust the formation parameters in the next inversion iteration.

[0050] Substituting the gradient function into the above gradient descent method, the formula for updating the formation parameters is as follows:

[0051] In the corresponding gradient function, the first term is the error gradient, which adjusts the inversion formation model to converge to the actual observed data; the second term is the regularization gradient, which smooths the change of the formation parameters of the inversion formation model.

[0052] To facilitate the rapid implementation of online real-time inversion, a pre-gradient matrix can be pre-trained using the gradient function in advance, that is, the pre-gradient matrix is obtained through offline training in advance. During online real-time inversion, the pre-gradient matrix can be directly used to quickly update the formation parameters in the iterative inversion process, significantly improving the inversion speed and real-time performance.

[0053] The specific process of pre-training to obtain the pre-gradient matrix is as follows: Construct a training sample set based on prior data, determine the initial inversion parameters from it, train the inversion formation model according to the training sample set, and obtain the pre-training matrices for each inversion iteration during the training process. The pre-training matrices correspond to the number of inversion iterations and are determined according to the gradient function. For example, if the number of inversion iterations is 7 times, 7 pre-training matrices are obtained, each corresponding to each inversion iteration. After obtaining the pre-gradient matrix, for any number of inversion iterations, based on the difference between the formation parameters of the current inversion iteration and the pre-gradient matrix corresponding to the current inversion iteration, the formation parameters of the next inversion iteration can be obtained and substituted into the inversion formation model for inversion. Here, the prior data includes, for example, being determined according to the well logging data of adjacent wells obtained by acquisition. The well logging data of adjacent wells can determine the initial value range of the formation parameters of the inversion formation model. If the well logging data of adjacent wells shows that the formation resistivity range is 1 - 10 ohm-meters, the initial value range of the formation parameters can be selected to start the inversion within the range of 1 - 10. As the inversion progresses, the range can also be adjusted according to the inversion results of each iteration to make the inversion results more accurate.

[0054] According to the formation parameters corresponding to each inversion iteration number, the values of the objective function for each inversion iteration number can be calculated respectively. By comparing the values of the objective function for each inversion iteration number, the formation parameters corresponding to the inversion iteration number with the minimum value of the objective function are determined as the formation parameters for constructing the inversion map of the current formation, that is, according to the objective function, the formation parameters with the smallest error from the actual logging data are determined as the formation parameters for constructing the inversion map of the current formation. After determining the formation parameters, the inversion map of the current formation can be constructed according to the formation parameters, such as constructing the inversion map of the current formation based on the formation parameters at different depths, etc., which will not be elaborated here.

[0055] Furthermore, during the inversion iteration process, the range of formation parameters can be adjusted according to the results of the inversion iteration. For example, the initial resistivity parameter range is set to [1, 10] ohm-meters according to the logging data. After performing inversion iteration for several times, such as k times, the mean value of formation parameters and the standard deviation of formation parameters are calculated based on the formation parameters for multiple inversion iteration numbers.

[0056]

[0057] Among them, is the mean value of formation parameters, and σ is the standard deviation of formation parameters. According to the mean value of formation parameters, the central tendency of formation parameters can be found. For example, if the calculated mean value of formation parameters is 6 ohm-meters, it indicates that the value range of formation parameters is closer to the range of 6 ohm-meters. The smaller the standard deviation of formation parameters, the lower the dispersion of formation parameters.

[0058] According to the mean value of formation parameters and the standard deviation of formation parameters, the value range of formation parameters can be updated, and the parameter interval [θ min , θ max is further updated as follows:

[0059]

[0060] Among them, Δ is the range offset determined according to the standard deviation of formation parameters, which is specifically set according to the actual situation and will not be limited here. For example, if Δ = 2, the new value range is obtained, θ min = 6 - 2 = 4, θ max = 6 + 2 = 8, and the updated value range of formation parameters is [4, 8]. When continuing the inversion, the updated value range of formation parameters can be used, such as taking values from the updated value range [4, 8] of formation parameters, and calculating according to the pre-trained matrix, and continuing to invert the inversion formation model. By adjusting the range of formation parameters, the inversion accuracy can be effectively improved, the calculation amount can be reduced, and the convergence can be accelerated, making the inversion result closer to the actual formation parameters.

[0061] Furthermore, for the pre-gradient matrix, the dimension of the pre-gradient matrix pre-trained according to prior data is relatively large, such as including dimensions of 1-10, 1-100, etc. If it is directly used for the inversion result of the current formation while drilling logging, the result will not be accurate enough. However, by using the pre-gradient matrix obtained through pre-training and based on the logging data of the while-drilling logging, the range of the pre-gradient matrix can be determined, and the inversion can be further refined based on the range of the pre-gradient matrix, so as to obtain a more accurate inversion map. Moreover, by using the pre-trained pre-gradient matrix, the inversion time can be greatly saved and the inversion efficiency can be improved.

[0062] Furthermore, the data volume of while-drilling logging is huge and complex, and the inversion process requires efficient and real-time processing. To improve the inversion speed, this embodiment uses parallel computing technology for inversion. Parallel computing includes data block processing, GPU acceleration, distributed computing, and a hybrid parallel architecture integrating various technologies. Data block processing can divide large-scale logging data and the inversion formation model into multiple sub-tasks according to depth or attributes and allocate them to different computing nodes or threads for parallel processing. Each data block is calculated as an independent task, and finally the results are summarized, so as to significantly shorten the calculation time while ensuring the inversion accuracy. By means of a fine-grained block method, the pressure of single calculation can be effectively reduced and the overall efficiency can be improved. On this basis, GPU acceleration further optimizes the calculation speed through its high parallel processing ability. Running the task on the GPU can greatly shorten the processing time. The GPU can use its multi-threaded architecture to parallel process different data blocks and improve the calculation efficiency through technologies such as data parallelism and shared memory, which is especially suitable for the high-density data processing requirements in deep formation inversion. For larger-scale inversion tasks, distributed computing can be adopted, and the inversion task is distributed to multiple computers for parallel processing. Each node completes the data part it is responsible for, and then the results are summarized to form a complete inversion model. Distributed computing adopts a high-speed connection and dynamic allocation mechanism in data transmission and task scheduling to ensure load balancing and full utilization of computing resources, which can not only significantly improve the inversion speed but also enhance the stability of the inversion. Combining the advantages of the above data block processing, GPU acceleration, and distributed computing, the hybrid parallel architecture can further improve the inversion efficiency, optimize multi-level task allocation, asynchronous task management, and fault tolerance mechanisms, etc., and can simultaneously utilize multi-node and GPU parallel operations to accelerate the inversion process, complete it quickly, and has high fault tolerance, making large-scale inversion under complex formation conditions more efficient and reliable.

[0063] Optionally, this embodiment may further include the following steps:

[0064] Step S104: For the curve data at any coordinate in the inversion map of the current formation, perform smoothing processing on the curve data according to the second-order differences in the horizontal and vertical directions of the coordinate and a preset smoothing factor, to obtain the inversion map of the current formation after smoothing processing.

[0065] Different formation parameters can be obtained through inversion iteration. For example, for a seven-layer complex inversion formation model, the resistivity values from top to bottom are respectively assigned as: 8 ohm-meters, 3 ohm-meters, 35 ohm-meters, 15 ohm-meters, 25 ohm-meters, 2 ohm-meters, and 60 ohm-meters. Assume that the thickness of each layer is different, and randomly set multiple formation dips to establish an inversion formation model with varying undulations, as Figure 2 shown. Among them, the red line is the trajectory designed for logging while drilling, which can be adjusted according to the real-time detection situation during actual logging. The parameters used in the inversion formation model are as Figure 3 shown, including data such as dip angle, azimuth angle, vertical depth, X-axis offset, Y-axis offset, fold angle, build-up rate, etc. corresponding to different depths. Through forward calculation, the relevant logging curve responses are obtained, the curve data is inverted through adaptive gradient optimization, and the inversion effect is optimized to obtain the final inversion map of the current formation.

[0066] Furthermore, the smoothness of the inversion map of the current formation has a great impact on the accuracy of formation feature recognition. However, the inversion map of the current formation generated during the inversion process may be affected by logging curve data, resulting in problems such as jagged color blocks and uneven distribution. This problem is mainly caused by the discontinuity of logging data in space, or by noise caused by the inhomogeneity and anisotropy of the formation.

[0067] To eliminate problems such as jagged color blocks in the inversion map of the current formation, in this embodiment, based on the inversion result, smoothing processing is performed on it, such as using high-order finite difference smoothing to reduce the discreteness and unnecessary fluctuations of the image, making the inversion map of the current formation smoother and more coherent while retaining important boundary information.

[0068] Specifically, for the curve data at any coordinate in the inversion map of the current formation, perform smoothing processing on the curve data according to the second-order differences in the horizontal and vertical directions of the coordinate and a preset smoothing factor, such as finite difference smoothing processing, as shown below:

[0069]

[0070] Among them, is the curve data at the coordinate (x, y) in the inversion map, is the curve data at the coordinate (x, y) after smoothing processing of , and λ is the preset smoothing factor used to control the intensity of smoothing, which is set according to the actual situation and is not limited here. They are the second-order differences in the horizontal and vertical directions respectively. During the smoothing process, the above finite-difference smoothing process can be repeated several times to gradually reduce the discreteness in the inversion map of the current formation, making the inversion result more tend to be smooth. As Figure 4 shown, where the red line is the trajectory designed for measurement while drilling, which can be adjusted according to the real-time detection situation during actual logging. The smoothing process can mainly act on local areas and will not have an obvious impact on important formation boundaries, thus ensuring the practicality of the inversion result. The finite-difference smoothing process is simple in calculation, can quickly complete the smoothing operation, and can be seamlessly integrated with the previous inversion process. Through the finite-difference smoothing process, the smoothness of the inversion map of the current formation is improved, further enhancing the coherence and readability of formation features.

[0071] According to the formation electromagnetic imaging method while drilling provided by the embodiments of the present invention, by combining the inversion with the formation model of machine learning, the inversion accuracy and global convergence ability are improved. Dynamic regularization can adaptively adjust the regularization parameter during the iteration process, enhancing the stability and accuracy of the inversion result. Directly applying the pre-trained pre-gradient matrix in the inversion significantly improves the inversion speed and real-time performance, and improves the inversion accuracy of complex formation parameters.

[0072] Figure 5 shows a schematic structural diagram of the formation electromagnetic imaging device while drilling provided by the embodiments of the present invention. As Figure 5 shown, the device includes:

[0073] A formation model module 510, adapted to determine an inversion formation model for formation electromagnetic imaging while drilling;

[0074] A gradient module 520, adapted to construct an objective function with a regularization term for the inversion formation model and calculate the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term;

[0075] An iterative inversion module 530, adapted to use the pre-gradient matrix pre-trained by the gradient function, according to the number of inversion iterations, iteratively update the formation parameters of the inversion formation model based on the pre-gradient matrix, and determine the formation parameters for constructing the inversion map of the current formation based on the objective function to obtain the inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations.

[0076] Optionally, the gradient module 520 is further adapted to:

[0077] Obtain the logging data of the well to be inverted and preprocess the logging data of the well to be inverted; the preprocessing includes data noise reduction and data compensation;

[0078] Construct an objective function for the inversion formation model based on the preprocessed logging data and model prediction data;

[0079] Add a regularization term to the objective function to obtain an objective function with a regularization term; wherein, the regularization term includes a regularization coefficient and formation parameters; the regularization coefficient decays sequentially according to the inversion iteration times based on a preset initial regularization parameter, and the decay is determined according to a decay factor.

[0080] Calculate the gradient of the objective function with a regularization term to obtain a gradient function with a regularization term.

[0081] Optionally, the iterative inversion module 530 is further adapted to:

[0082] Construct a training sample set according to prior data;

[0083] Determine the initial inversion parameters, and train the inversion formation model according to the training sample set to obtain a pre-training matrix for each inversion iteration time during the training process; wherein, the pre-training matrix corresponds to the inversion iteration times and is determined according to the gradient function.

[0084] Optionally, the iterative inversion module 530 is further adapted to:

[0085] Collect the logging data of adjacent wells;

[0086] Determine the initial value range of the formation parameters of the inversion formation model according to the logging data of adjacent wells; the preprocessing includes data noise reduction and data compensation;

[0087] For any inversion iteration time, obtain the formation parameters of the next inversion iteration time according to the difference between the formation parameters of the current inversion iteration time and the pre-gradient matrix corresponding to the current inversion iteration time, so as to substitute them into the inversion formation model for inversion;

[0088] According to the formation parameters corresponding to each inversion iteration time, calculate the value of the objective function for each inversion iteration time, and determine the formation parameters of the inversion iteration time corresponding to the minimum value of the objective function as the formation parameters for constructing the inversion map of the current formation.

[0089] Optionally, the device further includes: a range update module 540, adapted to calculate the formation parameter mean value and the formation parameter standard deviation according to the formation parameters of multiple inversion iteration times; update the value range of the formation parameters according to the formation parameter mean value and the formation parameter standard deviation; and perform inversion on the inversion formation model according to the updated value range of the formation parameters.

[0090] Optionally, the device further includes: a smoothing module 550, adapted to perform smoothing processing on the curve data of any coordinate in the inversion map of the current formation according to the second-order differences in the horizontal and vertical directions of the coordinate and a preset smoothing factor, to obtain the inversion map of the current formation after smoothing processing.

[0091] The descriptions of the above modules refer to the corresponding descriptions in the method embodiments and will not be elaborated here.

[0092] An embodiment of the present invention further provides a non-volatile computer storage medium. The computer storage medium stores at least one executable instruction, and the executable instruction can execute the operations corresponding to the electromagnetic imaging method while drilling in any of the above method embodiments.

[0093] An embodiment of the present application provides a computer program product. The computer program product includes at least one executable instruction or computer program, and the executable instruction or computer program can enable a processor to execute the operations corresponding to the electromagnetic imaging method while drilling in any of the above method embodiments.

[0094] Figure 6 A schematic structural diagram of a computing device according to an embodiment of the present invention is shown. Specific embodiments of the present invention do not limit the specific implementation of the computing device.

[0095] As Figure 6 shown, the computing device may include: a processor 602, a communication interface 604, a memory 606, and a communication bus 608.

[0096] Among them:

[0097] The processor 602, the communication interface 604, and the memory 606 communicate with each other through the communication bus 608.

[0098] The communication interface 604 is used to communicate with network elements of other devices such as clients or other servers.

[0099] The processor 602 is used to execute the program 610, and specifically can execute the relevant steps in the above embodiments of the electromagnetic imaging method while drilling.

[0100] Specifically, the program 610 may include program codes, and the program codes include computer operation instructions.

[0101] The processor 602 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the computing device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0102] A memory 606 for storing a program 610. The memory 606 may include high-speed RAM memory and may also include non-volatile memory, such as at least one magnetic disk memory.

[0103] The program 610 can specifically be used to cause the processor 602 to execute the electromagnetic imaging method while drilling in any of the above method embodiments. For the specific implementation of each step in the program 610, reference can be made to the corresponding steps and units in the above electromagnetic imaging embodiments while drilling, which will not be elaborated here. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and modules can refer to the corresponding process descriptions in the foregoing method embodiments, which will not be repeated here.

[0104] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. Based on the above description, the structure required to construct such a system is obvious. In addition, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the embodiments of the present invention described herein can be implemented using various programming languages, and the descriptions of specific languages above are for disclosing the preferred embodiments of the embodiments of the present invention.

[0105] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0106] Similarly, it should be understood that in order to streamline the embodiments of the present invention and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed embodiments of the present invention require more features than are expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate embodiment of the present invention.

[0107] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all the features disclosed in this specification (including the accompanying claims, abstract and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract and drawings) can be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0108] In addition, those skilled in the art can understand that although some of the embodiments herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present invention and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0109] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The embodiments of the present invention can also be implemented as a device or apparatus program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the embodiments of the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0110] It should be noted that the above embodiments illustrate the embodiments of the present invention rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Embodiments of the present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In the unit claims listing several devices, several of these devices may be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for electromagnetic imaging of formations while drilling, characterized in that: Methods include: Determine the inversion formation model for electromagnetic imaging of formations while drilling; An objective function with a regularization term is constructed for the inverted formation model, and the gradient of the objective function with the regularization term is calculated to obtain a gradient function with the regularization term; wherein the regularization term includes a regularization coefficient and a formation parameter; the regularization coefficient decays in sequence according to the preset initial regularization parameter and the number of inversion iterations, and the attenuation is determined according to an attenuation factor; The pre-gradient matrix obtained by pre-training with the gradient function is used to iteratively update the formation parameters of the inversion formation model according to the pre-gradient matrix according to the number of inversion iterations, and the formation parameters for constructing the inversion map of the current formation are determined based on the objective function to obtain the inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations; the pre-gradient matrix obtained by pre-training with the gradient function is specifically: a training sample set is constructed according to prior data; initial inversion parameters are determined, and the inversion formation model is trained according to the training sample set to obtain the pre-gradient matrix of each inversion iteration number in the training process; wherein the pre-gradient matrix corresponds to the number of inversion iterations and is determined according to the gradient function.

2. The method according to claim 1, characterized in that The step of constructing an objective function with a regularization term for the inversion formation model, and calculating the gradient of the objective function with the regularization term to obtain the gradient function with the regularization term further comprises: Acquire the logging data of the well to be inverted, and pre-process the logging data of the well to be inverted; the pre-processing includes data noise reduction and data compensation; Constructing the objective function of the inversion formation model based on the preprocessed logging data and the model prediction data; Adding a regularization term to the objective function to obtain an objective function with a regularization term; The gradient of the objective function with the regularization term is calculated to obtain a gradient function with the regularization term.

3. The method according to claim 1, characterized in that The method further comprises: iteratively updating the formation parameters of the inversion formation model according to the pre-gradient matrix obtained by pre-training the gradient function according to the number of inversion iterations, and determining the formation parameters of the inversion map of the current formation based on the objective function. Collect logging data of adjacent wells; Determining the initial value range of the formation parameters of the inversion formation model according to the logging data of the adjacent wells; For any inversion iteration number, the formation parameters of the next inversion iteration number are obtained according to the difference between the formation parameters of the current inversion iteration number and the pre-gradient matrix corresponding to the current inversion iteration number, so as to substitute them into the inversion formation model for inversion; According to the formation parameters corresponding to each inversion iteration number, the value of the objective function for each inversion iteration number is calculated, and the formation parameters for the inversion iteration number corresponding to the minimum value of the objective function are determined as the formation parameters for constructing the inversion map of the current formation.

4. The method according to claim 3, characterized in that The method further comprises: According to the formation parameters of multiple inversion iterations, the formation parameter mean and formation parameter standard deviation are calculated; Updating the value range of the formation parameter according to the formation parameter mean and the formation parameter standard deviation; The inversion formation model is inverted according to the updated value range of the formation parameter.

5. The method according to claim 1, characterized in that The method further comprises: For the curve data of any coordinate in the inversion map of the current formation, the curve data is smoothed according to the second-order differences in the horizontal and vertical directions of the coordinates and a preset smoothing factor to obtain the inversion map of the current formation after smoothing.

6. A formation electromagnetic imaging device while drilling, characterized in that: The device includes: A formation model module suitable for determining the inversion formation model of electromagnetic imaging of formations while drilling; A gradient module, adapted to construct an objective function with a regularization term for the inverted formation model, and calculate the gradient of the objective function with the regularization term to obtain a gradient function with the regularization term; wherein the regularization term includes a regularization coefficient and a formation parameter; the regularization coefficient decays in sequence according to a preset initial regularization parameter and the number of inversion iterations, and the decay is determined according to an attenuation factor; The iterative inversion module is suitable for using the pre-gradient matrix obtained by pre-training the gradient function, iteratively updating the formation parameters of the inversion formation model according to the pre-gradient matrix according to the number of inversion iterations, and determining the formation parameters for constructing the inversion map of the current formation based on the objective function to obtain the inversion map of the current formation; the pre-gradient matrix corresponds to the number of inversion iterations; the pre-gradient matrix obtained by pre-training with the gradient function is specifically: constructing a training sample set according to prior data; determining initial inversion parameters, training the inversion formation model according to the training sample set, and obtaining the pre-gradient matrix of each inversion iteration number in the training process; wherein the pre-gradient matrix corresponds to the number of inversion iterations and is determined according to the gradient function.

7. A computing device, characterized in that include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the method for electromagnetic imaging of formations while drilling as described in any one of claims 1-5.

8. A computer storage medium, characterized in that The storage medium stores at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the method for electromagnetic imaging of formations while drilling as described in any one of claims 1-5.

9. A computer program product, characterized in that The method comprises at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the method for electromagnetic imaging of formations while drilling as described in any one of claims 1 to 5.

Images