Mechanical drilling speed prediction method, device, electronic equipment, and storage medium based on seismic frequency division attributes.

Abstract

This invention relates to the field of drilling mechanical rate of penetration (MRLP) analysis technology, specifically a method, device, electronic device, and storage medium for predicting MRLP based on seismic frequency division attributes. The method includes acquiring the MRLP of a completed well along its well trajectory after seismic frequency division attributes and well-seismic matching calibration; inputting the MRLP of the completed well along its well trajectory into a MRLP prediction model to obtain a three-dimensional spatial MRLP prediction value; and extracting the MRLP prediction value of the design well from the three-dimensional spatial MRLP prediction value using the design well trajectory. The model of this invention can fully learn the nonlinear relationship between the MRLP of a completed well and the seismic frequency division attributes, improving the accuracy of the three-dimensional spatial MRLP prediction, thereby improving the prediction accuracy of the MRLP prediction value of the design well. The MRLP prediction value can then be used for drilling parameter optimization, tool selection, improving drilling efficiency, and reducing operating costs.

Description

Technical Field

[0001] This invention relates to the field of drilling mechanical drilling speed analysis technology, and is a method, device, electronic device and storage medium for predicting mechanical drilling speed based on seismic frequency division attributes. Background Technology

[0002] In oil drilling, the mechanical rate of penetration (MRP) is one of the key factors affecting drilling efficiency and an important economic indicator for oilfield drilling operations. MRP represents the footage advanced per unit time, providing the most direct and concrete reflection of drilling operation time and cost. Accurate prediction of MRP is crucial for optimizing drilling parameters, selecting the best tools, improving drilling efficiency, and reducing operating costs.

[0003] Currently, mechanical drilling speed prediction is often combined with deep learning, for example:

[0004] Existing patent document 1, publication number CN113689055B, discloses a method for predicting and optimizing mechanical drilling rate (MRDR) in oil and gas drilling based on Bayesian optimization. It collects raw drilling data according to a preset sampling period and constructs an initial sample dataset based on the raw drilling data. A MRDR prediction model is built using the obtained initial sample data, and the MRDR prediction model is combined with Gaussian process regression to predict the MRDR at the next sampling point. This method enables rapid analysis of historical drilling data, predicts the MRDR range of sampling points within the feasible region, and performs Bayesian optimization to obtain the optimized MRDR and its optimized engineering parameters. It imposes few restrictions on drilling engineering parameters and original formation parameters, has high prediction accuracy, and can find the engineering parameters at the optimal MRDR without the problem of ambiguous parameter value boundaries; its optimized values ​​are clear and accurate.

[0005] Existing patent document two, publication number CN114065603B, discloses a method and apparatus for predicting mechanical drilling rate, comprising: obtaining mechanical drilling data; classifying the mechanical drilling data according to rock breaking mechanism information; inputting mechanical drilling data of type linear relationship parameters into a pre-trained single-input linear neural network model and outputting linear correlation parameters; inputting mechanical drilling data of type nonlinear relationship parameters into a pre-trained deep neural network model and outputting nonlinear fitting parameters; inputting mechanical drilling data of type fuzzy linear relationship parameters into a pre-trained hybrid-input linear neural network model and outputting linear fitting parameters; and inputting the linear correlation parameters, nonlinear fitting parameters, and linear fitting parameters into a pre-trained cross-neural network model and outputting the mechanical drilling rate prediction result. This invention can predict mechanical drilling rate and improve prediction accuracy by exploring the underlying physical mechanisms of the data.

[0006] The third existing publicly available paper, "Machine Learning for Predicting Mechanical Drilling Rate and Its Engineering Applications," uses 14 engineering parameters, including drill bit type, drilling pressure, and rotation speed, as inputs and mechanical drilling rate as the output. It builds a model using over 17,000 data points from 10 wells in Block X and attempts various machine learning algorithms, including linear regression, random forest, K-nearest neighbor algorithm, and gradient boosting tree. Comparisons show that the gradient boosting tree algorithm performs best. The algorithm's parameters are then optimized to further improve model performance, increasing the correlation coefficient from 0.806 to 0.866.

[0007] However, the aforementioned existing technologies all rely on model training based on engineering parameters or mechanical drilling data, and use the model to obtain the mechanical drilling rate prediction value through engineering parameters or mechanical drilling data. They lack consideration for the three-dimensional seismic body of the block where the design well is located, which will affect the accuracy of the mechanical drilling rate prediction. Summary of the Invention

[0008] This invention provides a mechanical drilling rate prediction method, device, electronic device, and storage medium based on seismic frequency division attributes, which overcomes the shortcomings of the prior art and can effectively solve the problem of the lack of consideration for the three-dimensional seismic body of the block where the design well is located in the existing mechanical drilling rate prediction methods.

[0009] One of the technical solutions of this invention is achieved through the following measures: a mechanical drilling rate prediction method based on seismic frequency division attributes, comprising:

[0010] The drilling data of a completed well in the same block as the design well and the three-dimensional seismic volume of the block where the design well is located are obtained. The well-seismic matching calibration function and the frequency division attribute function are introduced to process the two to obtain the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time and density.

[0011] Input the seismic frequency division attribute and the mechanical drilling rate along the well trajectory of the completed well after well-seismic matching calibration to the mechanical drilling rate prediction model to obtain the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples.

[0012] Using the design well trajectory, the mechanical drilling rate prediction value of the design well is extracted from the mechanical drilling rate prediction value in three-dimensional space.

[0013] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0014] The aforementioned mechanical drilling rates along the well trajectory of completed wells after obtaining seismic frequency division attributes and well-seismic matching calibration include:

[0015] Synthetic seismic data is generated based on the sonic transit time and density of completed wells. Well-seismic matching calibration is performed by combining the mechanical drilling rate of completed wells with the three-dimensional seismic body of the block where the design well is located, and the mechanical drilling rate of completed wells along the well trajectory is obtained after well-seismic matching calibration.

[0016]

[0017] in, The mechanical drilling rate along the well trajectory after drilling is completed. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic volume of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration;

[0018] Extract seismic frequency division attributes from the 3D seismic volume of the block where the design well is located;

[0019]

[0020] in, To design the 3D seismic volume of the block where the well is located, It is a frequency division attribute function. It is the extracted seismic frequency division attribute.

[0021] The construction process of the above-mentioned mechanical drilling rate prediction model includes:

[0022] Multiple samples were acquired and divided into training sample set and test sample set according to the proportion. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data.

[0023] The support vector machine model is trained using the training sample set. The training ends when the loss function is stable or the number of iterations exceeds the limit, thus obtaining the mechanical drilling speed prediction model.

[0024] The mechanical drilling rate prediction model is tested using a test training set, the model parameters are optimized, and a mechanical drilling rate prediction model that meets the test evaluation requirements is output.

[0025] The above method utilizes the design well trajectory to extract the predicted mechanical drilling rate (MRR) value of the design well from the three-dimensional spatial MRR prediction value, as detailed below:

[0026]

[0027] in, This is a three-dimensional spatial mechanical drilling rate prediction value. The design well trajectory for the design well. This is the predicted mechanical drilling rate for the designed well.

[0028] The second technical solution of the present invention is achieved through the following measures: a mechanical drilling speed prediction device based on seismic frequency division attributes, comprising:

[0029] The data acquisition unit acquires drilling data from a completed well in the same block as the design well, as well as the three-dimensional seismic body of the block where the design well is located. It then uses a well-seismic matching calibration function and a frequency-division attribute function to process the data, obtaining the seismic frequency-division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0030] The first prediction unit takes the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration as input to the mechanical drilling rate prediction model, and obtains the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples.

[0031] The second prediction unit uses the design well trajectory of the design well to extract the predicted mechanical drilling rate of the design well from the predicted mechanical drilling rate in three-dimensional space.

[0032] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0033] The aforementioned data acquisition unit includes:

[0034] The basic data acquisition module acquires drilling data from a completed well in the same block as the design well, as well as the 3D seismic body of the block where the design well is located. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0035] The first data analysis module generates synthetic seismic data based on the sonic transit time and density of completed wells. It then performs well-seismic matching calibration by combining the mechanical drilling rate of completed wells with the 3D seismic volume of the block where the designed well is located, obtaining the mechanical drilling rate of the completed wells along the well trajectory after well-seismic matching calibration.

[0036]

[0037] in, The mechanical drilling rate along the well trajectory after drilling is completed. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic volume of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration;

[0038] The second data analysis module extracts seismic frequency division attributes from the 3D seismic volume of the block where the design well is located:

[0039]

[0040] in, To design the 3D seismic volume of the block where the well is located, It is a frequency division attribute function. It is the extracted seismic frequency division attribute.

[0041] The above also includes a model building unit for building a mechanical drilling rate prediction model, including:

[0042] Multiple samples were acquired and divided into training sample set and test sample set according to the proportion. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data.

[0043] The support vector machine model is trained using the training sample set. The training ends when the loss function is stable or the number of iterations exceeds the limit, thus obtaining the mechanical drilling speed prediction model.

[0044] The mechanical drilling rate prediction model is tested using a test training set, the model parameters are optimized, and a mechanical drilling rate prediction model that meets the test evaluation requirements is output.

[0045] The third technical solution of the present invention is achieved through the following measures: an electronic device, including a processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement the steps in the mechanical drilling speed prediction method based on seismic frequency division attributes.

[0046] The fourth technical solution of the present invention is achieved by the following measures: a storage medium storing a computer program that can be read by a computer, the computer program being configured to execute the steps in the mechanical drilling speed prediction method based on seismic frequency division attributes when running.

[0047] This invention employs Support Vector Machine (SVM) supervised learning to construct a mechanical drilling rate prediction model. SVM has advantages such as high-dimensional data processing capabilities and the ability to solve nonlinear problems by using different kernel functions. This allows the model to fully learn the nonlinear relationship between the mechanical drilling rate of completed wells and seismic frequency distribution attributes, thereby improving the accuracy of the model's three-dimensional spatial mechanical drilling rate prediction. This, in turn, improves the prediction accuracy of the mechanical drilling rate of the designed well. The predicted mechanical drilling rate is then used for drilling parameter optimization, tool selection, improving drilling efficiency, and reducing operating costs. Attached Figure Description

[0048] Appendix Figure 1This is a schematic diagram of an implementation environment provided for an embodiment of the present invention.

[0049] Appendix Figure 2 This is a schematic diagram of the mechanical drilling speed prediction method provided in an embodiment of the present invention.

[0050] Appendix Figure 3 This is a schematic diagram of the process for obtaining synthetic seismic record calibration data and seismic frequency division attributes provided in an embodiment of the present invention.

[0051] Appendix Figure 4 This is a schematic diagram of the process for constructing a mechanical drilling rate prediction model according to an embodiment of the present invention.

[0052] Appendix Figure 5 This is a mechanical drilling rate curve diagram of two completed wells provided in an embodiment of the present invention.

[0053] Appendix Figure 6 This is a three-dimensional seismic data map of the block where the design well is located, provided as an embodiment of the present invention.

[0054] Appendix Figure 7 This is a synthetic seismic record calibration map provided in an embodiment of the present invention.

[0055] Appendix Figure 8 This is an extraction diagram of earthquake frequency division attribute 1 provided in an embodiment of the present invention.

[0056] Appendix Figure 9 This is a diagram showing the extraction of earthquake frequency division attributes 2 provided in an embodiment of the present invention.

[0057] Appendix Figure 10 This is a diagram showing the extraction of earthquake frequency division attributes 3 provided in an embodiment of the present invention.

[0058] Appendix Figure 11 This is a diagram showing the extraction of earthquake frequency division attributes 4 provided in an embodiment of the present invention.

[0059] Appendix Figure 12 This is the earthquake frequency division attribute extraction diagram provided in the embodiment of the present invention.

[0060] Appendix Figure 13 This is a schematic diagram of the model training results provided in an embodiment of the present invention.

[0061] Appendix Figure 14 A three-dimensional spatial mechanical drilling rate prediction diagram provided for an embodiment of the present invention.

[0062] Appendix Figure 15 This is a mechanical drilling rate prediction diagram for a design well according to an embodiment of the present invention.

[0063] Appendix Figure 16 This is a schematic diagram of a mechanical drilling speed prediction device provided in an embodiment of the present invention.

[0064] Appendix Figure 17 This is a schematic diagram of another mechanical drilling speed prediction device provided in an embodiment of the present invention. Detailed Implementation

[0065] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.

[0066] Those skilled in the art will understand that, unless specifically stated otherwise, in the embodiments of the present invention, a "module" or "unit" refers to a computer program or part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0067] In addition, in the embodiments of the present invention, "multiple" refers to two or more, and "first" and "second" are used to distinguish descriptions and should not be construed as implying relative importance.

[0068] This invention provides a method, apparatus, electronic device, and storage medium for predicting mechanical drilling rate based on seismic frequency division attributes. The method acquires the mechanical drilling rate along the well trajectory of a completed well after seismic frequency division attributes and well-seismic matching calibration. It inputs the mechanical drilling rate along the well trajectory of the completed well after seismic frequency division attributes and well-seismic matching calibration into a mechanical drilling rate prediction model to obtain a three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained through support vector machine supervised learning from multiple samples. Using the design well trajectory of the design well, the mechanical drilling rate prediction value of the design well is extracted from the three-dimensional spatial mechanical drilling rate prediction value.

[0069] The method provided in this embodiment of the invention may involve artificial intelligence (AI) technology and may be implemented based on artificial intelligence technology, such as using deep learning to train a corresponding model using samples.

[0070] Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory, among others. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence.

[0071] Deep learning (DL) specifically refers to machine learning based on deep neural network models and methods. It has developed from statistical machine learning, artificial neural network algorithms, and other algorithms, combined with the advancements in big data and computing power. The most important technical feature of deep learning is its ability to automatically extract features.

[0072] The aforementioned machine learning and deep learning typically include techniques such as neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.

[0073] In deep learning, the loss function is used to predict the target value by comparing the predicted value with the target value. This is done by updating the weight vector of each layer of the neural network based on the difference between the two values ​​(usually with an initialization process before the first update, where parameters are pre-configured for each layer) until the network can predict the target value or a value very close to it. Therefore, deep learning requires pre-defining "how to compare the difference between the predicted value and the target value," which is the loss function.

[0074] As attached Figure 1 The diagram illustrates an implementation environment provided by an embodiment of the present invention. This implementation environment may include: training equipment and usage equipment.

[0075] Both the training equipment and the equipment used are computer devices; optionally, the computer device is a terminal device, such as a mobile phone, tablet computer, PC (Personal Computer) or other electronic devices; or, the computer device is a server, which can be a single server, a server cluster composed of multiple servers, or a cloud computing service center. This embodiment of the invention does not limit this.

[0076] Training equipment refers to computer equipment capable of training and learning support vector machines (SVMs). Optionally, the training equipment has the ability to acquire SVMs and train and learn them according to application requirements. For example, the training equipment acquires an SVM from another device via a network and then trains it using training samples according to application requirements, so that the SVM has the ability to obtain three-dimensional spatial mechanical drilling rate prediction values. Optionally, the training equipment has the ability to construct SVMs. It can construct an SVM itself according to application requirements and then train and learn it. For example, in order to obtain three-dimensional spatial mechanical drilling rate prediction values ​​based on the mechanical drilling rate of completed wells along the well trajectory after seismic frequency division attributes and well-seismic matching calibration, the training equipment constructs an SVM itself and then trains and learns it using samples according to application requirements.

[0077] The device used refers to a computer device that has the requirement to use a support vector machine. Optionally, the device uses a support vector machine from other devices through a network according to the application requirements. For example, if the device has the requirement to predict the three-dimensional mechanical drilling rate, it can obtain a support vector machine from other devices through a network to complete the training and learning of predicting the three-dimensional mechanical drilling rate, and use the support vector machine to predict the three-dimensional mechanical drilling rate.

[0078] Based on this, the technical solution of the present invention will be described and explained below with reference to several examples.

[0079] Example 1: As shown in the attached document Figure 2 As shown in the figure, this invention discloses a method for predicting mechanical drilling speed based on seismic frequency division attributes, including:

[0080] Step S110: Obtain drilling data of a completed well in the same block as the design well, and the three-dimensional seismic body of the block where the design well is located. Introduce well-seismic matching calibration function and frequency division attribute function to process the two, and obtain seismic frequency division attribute and mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0081] Step S120: Input the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration to the mechanical drilling rate prediction model to obtain the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained through support vector machine supervised learning of multiple samples.

[0082] Step S130: Using the design well trajectory of the design well, extract the predicted mechanical drilling rate of the design well from the predicted mechanical drilling rate in three-dimensional space.

[0083] This invention discloses a method for predicting mechanical drilling rate based on seismic frequency division attributes. A mechanical drilling rate prediction model is constructed using support vector machine (SVM) supervised learning. SVM has advantages such as high-dimensional data processing capabilities and the ability to solve nonlinear problems by using different kernel functions. This allows the model to fully learn the nonlinear relationship between the mechanical drilling rate of completed wells and the seismic frequency division attributes, thereby improving the accuracy of the model's three-dimensional spatial mechanical drilling rate prediction. This, in turn, improves the accuracy of the mechanical drilling rate prediction for designed wells. The predicted mechanical drilling rate is then used for drilling parameter optimization, tool selection, improving drilling efficiency, and reducing operating costs.

[0084] Example 2: As shown in the attached document Figure 3As shown, the present invention embodiment is a further optimization of the above embodiment, wherein the drilling data of a completed well in the same block as the design well and the three-dimensional seismic body of the block where the design well is located are obtained. A well-seismic matching calibration function and a frequency-division attribute function are introduced to process the two data points, obtaining the seismic frequency-division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, including:

[0085] Step S210: Obtain drilling data of a completed well in the same block as the design well, and the three-dimensional seismic body of the block where the design well is located. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0086] Step S220: Based on the sonic transit time and density of the completed well, generate a synthetic seismic event, and combine the mechanical drilling rate of the completed well with the three-dimensional seismic body of the block where the design well is located to perform well-seismic matching calibration, so as to obtain the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration.

[0087]

[0088] in, The mechanical drilling rate along the well trajectory after drilling is completed. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic volume of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration;

[0089] Step S230: Extract seismic frequency division attributes from the three-dimensional seismic volume of the block where the design well is located;

[0090]

[0091] in, To design the 3D seismic volume of the block where the well is located, It is a frequency division attribute function. It is the extracted seismic frequency division attribute.

[0092] Example 3: As shown in the attached document Figure 4 As shown, the embodiments of the present invention are further optimizations of the above embodiments, wherein the construction process of the mechanical drilling rate prediction model includes:

[0093] Step S310: Obtain multiple samples and divide them into training sample set and test sample set according to the ratio. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data.

[0094] Each of the above samples represents a completed well in the same block as the design well. The process of obtaining the mechanical drilling rate along the well trajectory of the completed well after well-seismic matching calibration is the same as in Example 2 and will not be repeated here.

[0095] Step S320: Train the support vector machine model using the training sample set. When the loss function is stable or the number of iterations exceeds the limit, end the training to obtain the mechanical drilling speed prediction model.

[0096] Step S330: Test the mechanical drilling rate prediction model using the test training set, optimize the model parameters of the mechanical drilling rate prediction model, and output a mechanical drilling rate prediction model that meets the test evaluation requirements.

[0097] It should also be noted that different blocks require retraining of the mechanical drilling rate prediction model. As the number of completed wells in the same block increases, an optimization interval can be set, and the samples can be continuously updated according to the optimization interval to retrain the mechanical drilling rate prediction model, thereby updating the mechanical drilling rate prediction model and making the mechanical drilling rate prediction results more accurate.

[0098] Example 4: The mechanical drilling rate prediction method based on seismic frequency division attributes disclosed in this invention is verified through an embodiment of this invention, as detailed below:

[0099] (i) Obtain drilling data from two completed wells in the same block as the design well, and three-dimensional seismic bodies of the block where the design well is located. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0100] The mechanical drilling rates of the two completed wells are shown in the attached figure. Figure 5 As shown in the attached figure, the horizontal axis represents the mechanical drilling rate (unit: meters per hour); the vertical axis represents the depth (unit: meters); and the 3D seismic body of the block where the designed well is located is shown in the attached figure. Figure 6 As shown in the attached figure, the horizontal axis represents the track number (unit: dimensionless); the vertical axis represents time (unit: milliseconds).

[0101] (ii) Based on the acoustic time difference and density, a synthetic earthquake is generated, and well-seismic matching calibration is performed by combining the mechanical drilling rate of the completed well and the three-dimensional seismic body of the block where the design well is located, so as to obtain the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration.

[0102] The well-vibration matching calibration is attached here. Figure 7 As shown in the attached figure, the horizontal axis represents, in order, the acoustic transit time (unit: microseconds / foot), density (unit: grams / cubic centimeter), reflection coefficient (unit: dimensionless), synthetic seismic record (unit: dimensionless), and 3D seismic and synthetic seismic records; the vertical axis represents the depth (unit: meters) and time (unit: milliseconds).

[0103] (iii) Extracting seismic frequency division attributes from the three-dimensional seismic body of the block where the design well is located;

[0104] The extracted seismic frequency division attributes are shown in the attached figure. Figures 8 to 12 As shown, the horizontal axis represents the track number (unit: dimensionless); the vertical axis represents time (unit: milliseconds).

[0105] (iv) Using the seismic frequency division attributes of the two completed wells and the mechanical drilling rate along the well trajectory after well-seismic matching calibration, a support vector machine supervised learning was performed to obtain a mechanical drilling rate prediction model. The training results are attached. Figure 13 As shown in the attached figure, the horizontal axis represents the mechanical drilling speed (unit: meters per hour); the vertical axis represents time (unit: milliseconds).

[0106] (v) Input the seismic frequency division attributes and the mechanical drilling rate along the well trajectory of the completed well after well-seismic matching calibration to the mechanical drilling rate prediction model to obtain the three-dimensional spatial mechanical drilling rate prediction value, as shown in the appendix. Figure 14 As shown in the attached figure, the horizontal axis represents the track number (unit: dimensionless); the vertical axis represents time (unit: milliseconds).

[0107] (vi) Using the design well trajectory, extract the predicted mechanical drilling rate (MRR) value of the design well from the three-dimensional spatial MRR prediction value, as shown in the appendix. Figure 15 As shown in the attached figure, the horizontal axis represents the mechanical drilling speed (unit: meters per hour); the vertical axis represents the depth (unit: meters).

[0108] Example 5: As shown in the attached document Figure 16 As shown, this embodiment of the invention discloses a mechanical drilling rate prediction device based on seismic frequency division attributes, comprising:

[0109] The data acquisition unit acquires drilling data from a completed well in the same block as the design well, as well as the three-dimensional seismic body of the block where the design well is located. It then uses a well-seismic matching calibration function and a frequency-division attribute function to process the data, obtaining the seismic frequency-division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0110] The first prediction unit takes the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration as input to the mechanical drilling rate prediction model, and obtains the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples.

[0111] The second prediction unit uses the design well trajectory of the design well to extract the predicted mechanical drilling rate of the design well from the predicted mechanical drilling rate in three-dimensional space.

[0112] The data acquisition unit includes:

[0113] The basic data acquisition module acquires drilling data from at least one completed well in the same block as the design well, as well as the three-dimensional seismic body of the block where the design well is located. The drilling data includes mechanical drilling rate, sonic transit time, and density.

[0114] The first data analysis module generates synthetic seismic data based on acoustic transit time and density, and performs well-seismic matching calibration by combining the mechanical drilling rate of completed wells and the 3D seismic body of the block where the designed well is located. This yields the mechanical drilling rate of the completed wells along the well trajectory after well-seismic matching calibration.

[0115]

[0116] in, The mechanical drilling rate along the well trajectory after drilling is completed. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic volume of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration;

[0117] The second data analysis module extracts seismic frequency division attributes from the 3D seismic volume of the block where the design well is located:

[0118]

[0119] in, To design the 3D seismic volume of the block where the well is located, It is a frequency division attribute function. It is the extracted seismic frequency division attribute.

[0120] Example 6: As shown in the appendix Figure 17 As shown, this embodiment of the invention discloses a mechanical drilling rate prediction device based on seismic frequency division attributes, comprising:

[0121] The model building unit constructs a mechanical drilling rate prediction model, including:

[0122] Multiple samples were acquired and divided into training sample set and test sample set according to the proportion. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data.

[0123] The support vector machine model is trained using the training sample set. The training ends when the loss function is stable or the number of iterations exceeds the limit, thus obtaining the mechanical drilling speed prediction model.

[0124] The mechanical drilling rate prediction model is tested using a test training set, the model parameters of the mechanical drilling rate prediction model are optimized, and a mechanical drilling rate prediction model that meets the test evaluation requirements is output.

[0125] The data acquisition unit acquires seismic frequency division attributes and the mechanical drilling rate of completed wells along the well trajectory after well-seismic matching calibration.

[0126] The first prediction unit takes the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration as input to the mechanical drilling rate prediction model, and obtains the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples.

[0127] The second prediction unit uses the design well trajectory of the design well to extract the predicted mechanical drilling rate of the design well from the predicted mechanical drilling rate in three-dimensional space.

[0128] Example 7: This embodiment of the invention discloses a storage medium storing a computer program that can be read by a computer. The computer program is configured to execute a mechanical drilling speed prediction method based on seismic frequency division attributes when running.

[0129] The aforementioned storage media may include, but are not limited to, USB flash drives, read-only memory, portable hard drives, magnetic disks, optical disks, and other media capable of storing computer programs.

[0130] Example 8: This embodiment of the invention discloses an electronic device, including a processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement a mechanical drilling rate prediction method based on seismic frequency division attributes.

[0131] The processor described above can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. It can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The memory can include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, portable hard drives, magnetic disks, or optical disks.

[0132] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0133] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0134] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0135] The above content is only a specific embodiment of the present invention, which has strong adaptability and implementation effect. However, the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be covered within the protection scope of the present invention. Therefore, equivalent changes made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for predicting mechanical drilling speed based on seismic frequency division attributes, characterized in that, include: The drilling data of a completed well in the same block as the design well and the three-dimensional seismic volume of the block where the design well is located are obtained. The well-seismic matching calibration function and the frequency division attribute function are introduced to process the two to obtain the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time and density. Input the seismic frequency division attribute and the mechanical drilling rate along the well trajectory of the completed well after well-seismic matching calibration to the mechanical drilling rate prediction model to obtain the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples. Using the design well trajectory, the mechanical drilling rate prediction value of the design well is extracted from the mechanical drilling rate prediction value in three-dimensional space. Among them, the mechanical drilling rate along the well trajectory of the completed well after obtaining the seismic frequency division attributes and well-seismic matching calibration includes: Synthetic seismic data is generated based on the sonic transit time and density of completed wells. Well-seismic matching calibration is performed by combining the mechanical drilling rate of completed wells with the three-dimensional seismic body of the block where the design well is located, and the mechanical drilling rate of completed wells along the well trajectory is obtained after well-seismic matching calibration. in, The mechanical drilling rate along the well trajectory after drilling is completed. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic body of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration; Extract seismic frequency division attributes from the 3D seismic volume of the block where the design well is located; in, To design the 3D seismic body of the block where the well is located, For frequency division attribute functions, For the extracted seismic frequency division attributes; Specifically, the mechanical drilling rate prediction value of the design well is extracted from the mechanical drilling rate prediction value in three-dimensional space using the design well trajectory, as detailed below: in, This is a predicted value for the mechanical drilling rate in three-dimensional space. The design well trajectory for the design well. This is the predicted mechanical drilling rate for the designed well.

2. The mechanical drilling rate prediction method based on seismic frequency division attributes according to claim 1, characterized in that, The process of constructing a mechanical drilling rate prediction model includes: Multiple samples were acquired and divided into training sample set and test sample set according to the proportion. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data. The support vector machine model is trained using the training sample set. The training ends when the loss function is stable or the number of iterations exceeds the limit, thus obtaining the mechanical drilling speed prediction model. The mechanical drilling rate prediction model is tested using a test training set, the model parameters are optimized, and a mechanical drilling rate prediction model that meets the test evaluation requirements is output.

3. A mechanical drilling rate prediction device based on seismic frequency division attributes, applying the method described in any one of claims 1 to 2, characterized in that, include: The data acquisition unit acquires drilling data from a completed well in the same block as the design well, as well as the three-dimensional seismic body of the block where the design well is located. It then uses a well-seismic matching calibration function and a frequency-division attribute function to process the data, obtaining the seismic frequency-division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration. The drilling data includes mechanical drilling rate, sonic transit time, and density. The first prediction unit takes the seismic frequency division attribute and the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration as input to the mechanical drilling rate prediction model, and obtains the three-dimensional spatial mechanical drilling rate prediction value. The mechanical drilling rate prediction model is obtained by support vector machine supervised learning through multiple samples. The second prediction unit uses the design well trajectory of the design well to extract the predicted mechanical drilling rate of the design well from the predicted mechanical drilling rate in three-dimensional space.

4. The mechanical drilling rate prediction device based on seismic frequency division attributes according to claim 3, characterized in that, The data acquisition unit includes: The basic data acquisition module acquires drilling data from a completed well in the same block as the design well, as well as the 3D seismic body of the block where the design well is located. The drilling data includes mechanical drilling rate, sonic transit time, and density. The first data analysis module generates synthetic seismic data based on the sonic transit time and density of completed wells. It then performs well-seismic matching calibration by combining the mechanical drilling rate of completed wells with the 3D seismic volume of the block where the designed well is located, obtaining the mechanical drilling rate of the completed wells along the well trajectory after well-seismic matching calibration. in, The mechanical drilling rate along the well trajectory of the completed well. The acoustic transit time for completed drilling. The density of completed drilling. To design the 3D seismic body of the block where the well is located, For well-seismic matching calibration function, The mechanical drilling rate along the well trajectory of the completed well after well vibration matching calibration; The second data analysis module extracts seismic frequency division attributes from the 3D seismic volume of the block where the design well is located: in, To design the 3D seismic body of the block where the well is located, For frequency division attribute functions, The extracted seismic frequency division attributes.

5. The mechanical drilling rate prediction device based on seismic frequency division attributes according to claim 3 or 4, characterized in that, It also includes a model building unit for building a mechanical drilling rate prediction model, including: Multiple samples were acquired and divided into training sample set and test sample set according to the proportion. Each sample includes the seismic frequency division attribute of the block, the mechanical drilling rate of the completed well along the well trajectory after well-seismic matching calibration, and the corresponding three-dimensional spatial mechanical drilling rate label data. The support vector machine model is trained using the training sample set. The training ends when the loss function is stable or the number of iterations exceeds the limit, thus obtaining the mechanical drilling speed prediction model. The mechanical drilling rate prediction model is tested using a test training set, the model parameters are optimized, and a mechanical drilling rate prediction model that meets the test evaluation requirements is output.

6. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement the steps of the method as described in any one of claims 1 to 2.

7. A storage medium, characterized in that, The storage medium stores a computer program that can be read by a computer, the computer program being configured to execute the steps of the method as described in any one of claims 1 to 2 when it is run.

Images