Method for predicting drillability of well to be drilled

By calculating the time domain-related seismic properties of wells to be drilled and wells that have been drilled, a drillability prediction model is established using feedforward backpropagation BP neural network, which solves the problem of large prediction error of rock drillability in the existing technology and improves drilling Efficiency and economic benefits.

CN118940598APending Publication Date: 2024-11-12PETROCHINA CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310525343.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-10
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the prediction of drillability of rocks to be drilled, the calculation method of single seismic attribute data or good single well correlation has problems such as large error and low accuracy, resulting in increased risks and increased costs during drilling.

Method used

The time-domain related seismic properties of the well to be drilled and drilled were calculated, and the drillability prediction model was established through feedforward backpropagation BP neural network training, and the normalized seismic attribute data were used to predict it, combining with the drilled rocks. The relationship between drilling level value and time-depth, establish a pre-drill prediction model for rock drillability.

Benefits of technology

It improves the accuracy of rock drillability prediction, reduces drilling risks, provides reasonable drilling methods and parameter selection, and reduces drilling costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940598A_ABST
    Figure CN118940598A_ABST
Patent Text Reader

Abstract

The invention provides a to-be-drilled well drillability prediction method, which comprises the following steps of: firstly, calculating a related seismic attribute of a time domain of a to-be-drilled well and a related seismic attribute of a drilled well, and performing correlation analysis on a related target interval of the to-be-drilled well and the drilled well; then, training of the feedforward back propagation BP neural network of each layer section in a to-be-drilled well is completed, a normalization calculation formula of related seismic attributes is obtained, a drillability pre-drilling prediction model of rock to be drilled is established, and finally, the drillability pre-drilling prediction model of the rock to be drilled is obtained through the trained feedforward back propagation BP neural network. And performing normalization processing on the seismic attributes of the different stratums of the to-be-drilled well, taking the seismic attributes as input for calculation, and predicting the drillability of the to-be-drilled well. According to the method, the stratum to be drilled can be evaluated, an important basis is provided for reasonably selecting a drilling mode and a drill bit type, designing drilling parameters and the like, and the method has important significance in improving the drilling speed and reducing the drilling cost.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of oil and gas exploration, and in particular to a method for predicting the drillability of a well to be drilled. Background Art

[0002] In recent years, the exploration and development of marine shale gas and conventional natural gas in the Sichuan Basin has made great progress. For example, in the Changning-Weiyuan block, Weirong shale gas exploration area, Jiaoshiba and Pingqiao shale gas exploration blocks in the Sichuan Basin, a large number of shale gas horizontal wells have obtained industrial gas flow after fracturing, achieving considerable economic benefits. In the process of drilling in continental and marine formations, the relevant rock drillability evaluation is very important. This is because the rock physical properties of the continental and marine formations in the Sichuan Basin are quite different, and the fracture system is relatively developed, the formation pressure coefficient is high and the well temperature is high. In addition, the treatment of the drilled gypsum salt layer is also a relatively important link.

[0003] Rock drillability is an important formation drilling characteristic parameter in oil drilling. It is a comprehensive indicator for evaluating the difficulty of rock crushing and a basic factor in determining drilling efficiency. Rock drillability parameters are an important basis for rationally selecting drilling methods, drill bit types and designing drilling parameters. Correctly evaluating formation drillability is one of the important foundations for achieving high-quality and efficient drilling, and is of great significance for increasing drilling speed and reducing drilling costs.

[0004] Many scholars and experts have made great efforts in the evaluation and calculation of rock drillability. For example, in the article "Using Multiple Logging Parameters to Calculate Rock Drillability" published by Lu Baoping et al. in 1998, multiple logging parameters were used to calculate rock drillability, and a mathematical model for predicting rock drillability was given; in the article "New Method for Drill Bit Optimization and Its Application" published by Li Zhonghui et al. in 2008, single-factor calculation models of rock drillability and acoustic time difference, rock drillability and formation density, and calculation models of rock drillability and multivariate logging parameters were established using field logging data and core experiment results. The relationship model between multivariate logging parameters and rock drillability was used to correct the error of the rock drillability grade calculated by a single logging parameter, and the horizontal formation rock drillability profile of the block and the vertical rock drillability profile of the entire block were established. Based on their research results, the drill bit type suitable for the block was selected to solve the problems of slow drilling speed and long drilling cycle. In the actual drilling process, special attention should be paid to the evaluation of reservoir location, fault or fracture system, lithology and its brittleness, formation fracturing coefficient, etc. In the oil and gas exploration in the Sichuan Basin, these evaluations are crucial to the drillability and safety of the wells. In addition, there are a large number of wells that have been drilled in the basin and have accumulated a lot of experience in the drilling process. Therefore, the parameter data of these wells drilled in the relevant formations can be used for reference to predict and evaluate the drillability of the wells to be drilled. According to relevant research results, it is found that the conventional rock drillability has the following problems:

[0005] (1) For the drillability of the rock to be drilled, using a single seismic attribute data, such as seismic attribute data related to a certain well logging attribute, to calculate the rock drillability level value will result in large errors and low accuracy.

[0006] (2) Regarding the drillability of the rock to be drilled, some conventional methods for calculating the drillability grade of the rock only use a single well with a good correlation to calculate the drillability grade of the rock to be drilled. This also has the problem of insufficient calculation accuracy, resulting in large errors. Summary of the invention

[0007] The purpose of the present invention is to provide a method for predicting the drillability of rocks to be drilled before drilling, to evaluate the relevant strata to be drilled, thereby reducing drilling risks and improving the economic benefits of related oil and gas exploration.

[0008] The technical solution adopted by the present invention to achieve its invention object is to provide a method for predicting the drillability of a well to be drilled, the steps of which are:

[0009] S1. Calculate the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled, and perform correlation analysis on the relevant target layers of the well to be drilled and the wells that have been drilled;

[0010] S2. Complete the feedforward back propagation BP neural network training of each layer in the well to be drilled, obtain the normalized calculation formula of the relevant seismic attributes, and establish a pre-drilling prediction model for the drillability of the rock to be drilled;

[0011] S3. Using the trained feedforward back propagation BP neural network, the seismic attributes of different strata to be drilled are normalized according to the normalization calculation formula of the relevant seismic attributes in step S2 and then used as input for calculation to predict the drillability of the well to be drilled.

[0012] Further, the step S1 includes the following specific steps:

[0013] S11, calculating relevant seismic attributes in the time domain of the well to be drilled and relevant seismic attributes of the wells already drilled;

[0014] S12, interpreting relevant stratigraphic data of the relevant drilled area and the area to be drilled, and obtaining stratigraphic data of the target layer;

[0015] S13, selecting a seismic attribute curve according to a target layer section to be drilled, performing correlation analysis with the selected seismic attribute curve and the sliding interval parameter and the same seismic attribute curve of the corresponding target layer section that has been drilled, and calculating a correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled;

[0016] S14. Determine the drillability grade of the rock in the well that has been drilled, and convert the rock drillability grade curve in the depth domain into the time domain by using the time-depth relationship to obtain the drillability grade curve of the rock in the time domain.

[0017] During operation, it is necessary to obtain the time-depth relationship of the relevant wells based on the calibration results of the synthetic records of the relevant drilling wells, convert the rock drillability grade value curve in the relevant wells from the depth domain to the time domain, and make the rock drillability grade value curve consistent with the sampling rate of the seismic attribute data to form a correspondence between the rock drillability grade value and the seismic attribute data.

[0018] Furthermore, in step S11, calculating the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the well that has been drilled includes calculating the seismic attributes related to lithology, fractures, rock brittleness and gas content.

[0019] These seismic attributes can be calculated from relevant pre-stack or post-stack seismic data, logging data and geological data, or they can be obtained by fusion processing of multiple seismic attribute data bodies. The same seismic attribute calculation method and parameters are used to calculate the relevant seismic data body in the study area to be drilled or other study areas where wells have been drilled, so as to obtain the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled. If the wells to be drilled and the wells that have been drilled belong to the same seismic work area, it is only necessary to calculate the relevant seismic attribute data for the seismic work area; otherwise, the relevant attribute data calculation method and parameters are used to calculate the seismic data body where each well is located, so as to obtain the seismic attribute data of the relevant wells.

[0020] Further, the specific process of step S12 is: according to the post-stack seismic data, logging data and geological data, well seismic calibration is performed, the target layer position is determined by using the calibration result, the interpretation grid is designed, the layer position is tracked and interpreted, and the interpretation result is interpolated and smoothed to obtain the layer data of the target layer section. The target layer section refers to the two-way reflection time period between two layer data.

[0021] Further, the specific calculation process of step S13 includes: starting from the top of the layer section, selecting a seismic attribute curve to be drilled as the standard curve, starting from the top of the relevant target layer section that has been drilled, and calculating the correlation coefficient value within the standard curve time length of the layer section, that is, within the standard section, with a fixed sliding interval. When the bottom of the standard curve touches the bottom of the target layer that has been drilled, the calculation is terminated, and the correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled is obtained, and a correlation coefficient database of the standard layers of related different layers is established. The size of the fixed sliding interval can be determined according to the standard curve, expert experience, the characteristics of the time length of the target layer to be drilled and the drilled target layer, and the actual operation situation.

[0022] Furthermore, when the standard curve time length in the target layer to be drilled is greater than the standard curve time length of the drilled target layer, the seismic attribute curve of the target layer to be drilled is divided into two or more sections, and the correlation coefficient values ​​are calculated using fixed sliding intervals.

[0023] Furthermore, the specific process of determining the drillability grade of the rock in the drilled well in step S14 is: according to the well logging data including the acoustic wave and the shear wave of the drilled well, the well logging data is calculated and converted into the rock drillability grade through the functional relationship between the well logging data and the rock drillability grade. The well logging data can be obtained by measuring the relevant well logging parameters in the well using a well logging instrument, or by using an indoor measuring instrument such as the Huashi-III drillability measuring instrument.

[0024] Further, the step S2 includes the following specific steps:

[0025] S21, using the calculation result of the correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled obtained in step S13, preferably taking the seismic attribute data of the standard section when the correlation coefficient value between the target layer section that has been drilled and the target layer section to be drilled is the largest and performing normalization processing as input, training a feedforward back propagation BP neural network, and after normalization processing, obtaining the normalized calculation formula of the relevant seismic attributes at the same time;

[0026] S22. According to the method in step S21, the training of the feedforward back propagation BP neural network of each layer section in the well to be drilled is completed, and a pre-drilling prediction model of the drillability of the rock to be drilled is established.

[0027] The beneficial effects of the present invention are:

[0028] The present invention provides a method for pre-drilling prediction of the drillability of a rock to be drilled, calculates relevant seismic attribute data of the well to be drilled and the wells that have been drilled, and applies an artificial neural network algorithm to perform pre-drilling prediction of the rock drillability, thereby finding the relationship between the relevant seismic attributes and the rock drillability, which can provide an important basis for reasonably selecting the drilling method, drill bit type, designing drilling parameters, etc., and is of great significance for improving the drilling speed and reducing the drilling cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flow chart of the method for predicting the drillability of a well to be drilled of the present invention. DETAILED DESCRIPTION

[0030] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] Example 1

[0032] A specific implementation of the present invention is to use the method for predicting the drillability of a well to be drilled of the present invention, comprising the following steps:

[0033] Step S1, calculating the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled, and performing correlation analysis on the relevant target layers of the well to be drilled and the wells that have been drilled. Step S1 includes the following specific steps:

[0034] S11, calculating relevant seismic attributes in the time domain of the well to be drilled and relevant seismic attributes of the wells already drilled;

[0035] S12, interpreting relevant stratigraphic data of the relevant drilled area and the area to be drilled, and obtaining stratigraphic data of the target layer;

[0036] S13, selecting a seismic attribute curve according to a target layer section to be drilled, performing correlation analysis with the selected seismic attribute curve and the sliding interval parameter and the same seismic attribute curve of the corresponding target layer section that has been drilled, and calculating a correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled;

[0037] S14. Determine the drillability grade of the rock in the well that has been drilled, and convert the rock drillability grade curve in the depth domain into the time domain by using the time-depth relationship to obtain the drillability grade curve of the rock in the time domain.

[0038] In this embodiment, according to relevant exploration experience and expert experience, a fixed sliding interval is set to 6ms, and the standard curve of the relevant target layer segment is divided into three segments in an unequal manner. Starting from the layer top, the relevant drilled layer segment is implemented according to the sliding interval to calculate the correlation coefficient value of the same seismic attribute data of the relevant sampling points of each well segment, and a series of correlation value data about the standard layer in the target layer is obtained.

[0039] Step S2, complete the feedforward back propagation BP neural network training of each layer section in the well to be drilled, obtain the normalized calculation formula of the relevant seismic attributes, and establish a pre-drilling prediction model for the drillability of the rock to be drilled. Step S2 includes the following specific steps:

[0040] S21, using the calculation result of the correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled obtained in step S13, preferably taking the seismic attribute data of the standard section when the correlation coefficient value between the target layer section that has been drilled and the target layer section to be drilled is the largest and performing normalization processing as input, training a feedforward back propagation BP neural network, and after normalization processing, obtaining the normalized calculation formula of the relevant seismic attributes at the same time;

[0041] S22. According to the method in step S21, the training of the feedforward back propagation BP neural network of each layer section in the well to be drilled is completed, and a pre-drilling prediction model of the drillability of the rock to be drilled is established.

[0042] Step S3, using the trained feedforward back propagation BP neural network, normalize the seismic attributes of different strata to be drilled according to the normalization calculation formula of the relevant seismic attributes in step S2, and then use them as input for calculation to predict the drillability of the well to be drilled.

[0043] Example 2

[0044] For a special exploration well for marine shale exploration in a certain three-dimensional work area, a drill bit scheme design and drilling measures are formulated, and a pre-drilling prediction study of rock drillability is conducted on the special exploration well. There are three special exploration wells for marine shale gas in the work area, and pre-drilling prediction of drillability of the wells to be drilled can be implemented. A specific implementation of the present invention is to use the method for predicting the drillability of the wells to be drilled of the present invention, including the following steps:

[0045] Step S1, calculating the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled, and performing correlation analysis on the relevant target layers of the well to be drilled and the wells that have been drilled. Step S1 includes the following specific steps:

[0046] S11, calculating relevant seismic attributes in the time domain of the well to be drilled and relevant seismic attributes of the wells already drilled;

[0047] S12, interpreting relevant stratigraphic data of the relevant drilled area and the area to be drilled, and obtaining stratigraphic data of the target layer;

[0048] S13, selecting a seismic attribute curve according to a target layer section to be drilled, performing correlation analysis with the selected seismic attribute curve and the sliding interval parameter and the same seismic attribute curve of the corresponding target layer section that has been drilled, and calculating a correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled;

[0049] S14. Determine the drillability grade of the rock in the well that has been drilled, and convert the rock drillability grade curve in the depth domain into the time domain by using the time-depth relationship to obtain the drillability grade curve of the rock in the time domain.

[0050] In this embodiment, the calculation of the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled includes calculating the seismic attributes related to lithology, fractures, rock brittleness and gas content. After calculation, three seismic attribute data bodies are obtained, namely wave impedance, curvature and Poisson's ratio, and the post-stack seismic data body is calibrated by well seismic, the target layer is determined, the interpretation grid is designed, the layer is tracked and interpreted, and the interpretation results are interpolated and smoothed to obtain relevant layer interpretation results. In actual operation, 12 layers are determined, thereby obtaining 11 target layer sections. The three seismic attribute data of the 11 target layers of the well to be drilled and the three wells that have been drilled are extracted, and the seismic attribute data of the 11 target layer sections of the relevant four wells are determined.

[0051] According to relevant exploration experience and expert experience, the fixed sliding interval is set to 4ms, and the standard curve of the relevant target layer is equally divided into two sections. Starting from the top of the layer, the relevant drilling layer section is implemented according to the sliding interval. The correlation coefficient value calculation of the same seismic attribute data of the relevant sampling points of each well section is implemented to obtain a series of correlation value data about the standard layer in the target layer, thereby establishing a correlation coefficient database for the standard layer. According to this method, the correlation coefficient value calculation of the two sample curves of 11 layers of three seismic attribute data is implemented, and 22 correlation coefficient databases for different layers are obtained, that is, one layer section has two correlation coefficient databases.

[0052] In addition, the rock drillability grade value of the drilled wells is calculated using logging data or indoor tests to obtain the rock drillability grade value curve in the drilled depth domain, which is then converted into the time domain after resampling.

[0053] Step S2, complete the feedforward back propagation BP neural network training of each layer section in the well to be drilled, obtain the normalized calculation formula of the relevant seismic attributes, and establish a pre-drilling prediction model for the drillability of the rock to be drilled. Step S2 includes the following specific steps:

[0054] S21, using the calculation result of the correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled obtained in step S13, preferably taking the seismic attribute data of the standard section when the correlation coefficient value between the target layer section that has been drilled and the target layer section to be drilled is the largest and performing normalization processing as input, training a feedforward back propagation BP neural network, and after normalization processing, obtaining the normalized calculation formula of the relevant seismic attributes at the same time;

[0055] S22. According to the method in step S21, the training of the feedforward back propagation BP neural network of each layer section in the well to be drilled is completed, and a pre-drilling prediction model of the drillability of the rock to be drilled is established.

[0056] In this embodiment, according to the two sample curves of the relevant layer section to be drilled, the standard layer position corresponding to the maximum correlation coefficient value is determined from the correlation coefficient database of each relevant layer section, and the normalized value range is determined according to the rock drillability grade value of the three wells. After extracting and normalizing the three seismic attribute data, the drillability grade value corresponding to the standard layer of the relevant well is used as the training data of the neural network, and the feedforward back propagation BP neural network training is implemented for the relevant target layer sections respectively, so as to establish the pre-drilling prediction model of the drillability of each layer section of the relevant target layer section to be drilled. By analogy, the training of the feedforward back propagation BP neural network of each layer section to be drilled is completed. In actual operation, in the evaluation of the target layer section to be drilled, the drillability grade value of the relevant extracted standard layer of the well that has been drilled is used as the training data of the neural network, and the three seismic attribute curve data corresponding to it are used as input, and the network model is 3-12-1. After 16,800 iterations, the sum of squares of the system error is 0.002, which is less than the expected error of 0.01, and the learning is completed. The seismic attribute data of the drilled wells are normalized so that their attribute data values ​​are all located in [2.1, 6.8], and the relevant normalized calculation formula is obtained.

[0057] Step S3, using the trained feedforward back propagation BP neural network, normalize the seismic attributes of different strata to be drilled according to the normalization calculation formula of the relevant seismic attributes in step S2, and then use them as input for calculation to predict the drillability of the well to be drilled.

[0058] In this embodiment, the three seismic attribute data of the well to be drilled are normalized according to the above normalization calculation formula and used as the input of the feedforward back propagation BP neural network for each layer. The feedforward back propagation BP neural network of each layer calculates the three relevant seismic attribute data through the corresponding calculation model to obtain a time domain rock drillability grade value curve data. Then, according to the rock drillability grade value curve data in the time domain of the well to be drilled and the determined time-depth relationship curve of the well to be drilled, after conversion, a depth domain rock drillability grade value curve of the well to be drilled is obtained.

[0059] The prediction model established according to the present embodiment makes a pre-drilling prediction of the drillability of the well to be drilled. The prediction results are compared with the logging data and laboratory test results. The maximum relative error between the prediction results and the measured results is 10.3%, and the average relative error is 4.7%, which meets the relevant geological and drilling requirements.

[0060] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although the present invention has been described in detail in this specification with reference to the above embodiments, the present invention is not limited to the above specific implementation modes. Therefore, any modification or equivalent replacement of the present invention and all technical solutions and improvements that do not depart from the spirit and scope of the invention are all covered by the scope of the claims of the present invention.

Claims

1. A method for predicting the drillability of a well to be drilled, characterized in that The method comprises the following steps: S1. Calculate the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the wells that have been drilled, and perform correlation analysis on the relevant target layers of the well to be drilled and the wells that have been drilled; S2. Complete the feedforward back propagation BP neural network training of each layer in the well to be drilled, obtain the normalized calculation formula of the relevant seismic attributes, and establish a pre-drilling prediction model for the drillability of the rock to be drilled; S3. Using the trained feedforward back propagation BP neural network, the seismic attributes of different strata to be drilled are normalized according to the normalization calculation formula of the relevant seismic attributes in step S2 and then used as input for calculation to predict the drillability of the well to be drilled.

2. A method for predicting the drillability of a well to be drilled according to claim 1, characterized in that: The step S1 comprises the following specific steps: S11, calculating relevant seismic attributes in the time domain of the well to be drilled and relevant seismic attributes of the wells already drilled; S12, interpreting relevant stratigraphic data of the relevant drilled area and the area to be drilled, and obtaining stratigraphic data of the target layer; S13, selecting a seismic attribute curve according to a target layer section to be drilled, performing correlation analysis with the selected seismic attribute curve and the sliding interval parameter and the same seismic attribute curve of the corresponding target layer section that has been drilled, and calculating a correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled; S14. Determine the drillability grade of the rock in the well that has been drilled, and convert the rock drillability grade curve in the depth domain into the time domain by using the time-depth relationship to obtain the drillability grade curve of the rock in the time domain.

3. A method for predicting the drillability of a well to be drilled according to claim 2, characterized in that: In the step S11, calculating the relevant seismic attributes in the time domain of the well to be drilled and the relevant seismic attributes of the well that has been drilled includes calculating the seismic attributes related to lithology, fractures, rock brittleness and gas content.

4. A method for predicting the drillability of a well to be drilled according to claim 2, characterized in that: The specific process of step S12 is: perform well-seismic calibration based on post-stack seismic data, logging data and geological data, use the calibration results to determine the target layer position, design an interpretation grid, track and interpret the layer, and perform interpolation and smoothing on the interpretation results to obtain the layer data of the target layer segment.

5. The method for predicting the drillability of a well to be drilled according to claim 2, characterized in that: The specific calculation process of step S13 includes: starting from the top of the layer segment, selecting a seismic attribute curve to be drilled as the standard curve, starting from the top of the relevant target layer segment that has been drilled, and calculating the correlation coefficient value within the standard curve time length of the layer segment, that is, within the standard segment, at a fixed sliding interval; when the bottom of the standard curve touches the bottom of the target layer that has been drilled, the calculation is terminated, and the correlation coefficient value of a seismic attribute of the target layer segment to be drilled relative to the drilled target layer segment is obtained, and a correlation coefficient database of standard layers of related different layer segments is established.

6. A method for predicting the drillability of a well to be drilled according to claim 5, characterized in that: When the standard curve time length in the target layer to be drilled is longer than that of the drilled target layer, the seismic attribute curve of the target layer to be drilled is divided into two or more sections, and the correlation coefficient values ​​are calculated using fixed sliding intervals.

7. A method for predicting the drillability of a well to be drilled according to claim 2, characterized in that: The specific process of determining the drillability grade of the rock in the drilled well in step S14 is: according to the logging data of the drilled well including sonic waves and shear waves, the logging data is calculated and converted into the rock drillability grade through the functional relationship between the logging data and the rock drillability grade.

8. The method for predicting the drillability of a well to be drilled according to claim 2, characterized in that: The step S2 comprises the following specific steps: S21, using the calculation result of the correlation coefficient value of a seismic attribute of the target layer section to be drilled relative to the target layer section that has been drilled obtained in step S13, preferably taking the seismic attribute data of the standard section when the correlation coefficient value between the target layer section that has been drilled and the target layer section to be drilled is the largest and performing normalization processing as input, training a feedforward back propagation BP neural network, and after normalization processing, obtaining the normalized calculation formula of the relevant seismic attributes at the same time; S22. According to the method in step S21, the training of the feedforward back propagation BP neural network of each layer section in the well to be drilled is completed, and a pre-drilling prediction model of the drillability of the rock to be drilled is established.