CN115329468A - Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm - Google Patents

Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm Download PDF

Info

Publication number
CN115329468A
CN115329468A CN202110507186.XA CN202110507186A CN115329468A CN 115329468 A CN115329468 A CN 115329468A CN 202110507186 A CN202110507186 A CN 202110507186A CN 115329468 A CN115329468 A CN 115329468A
Authority
CN
China
Prior art keywords
drilling
neural network
bas
model
parameters
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110507186.XA
Other languages
Chinese (zh)
Inventor
何璟彬
陈伟林
李小鹏
倪华峰
李录科
李德波
石崇东
赵莹
詹胜
何以晴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China National Petroleum Corp
CNPC Chuanqing Drilling Engineering Co Ltd
Original Assignee
China National Petroleum Corp
CNPC Chuanqing Drilling Engineering Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China National Petroleum Corp, CNPC Chuanqing Drilling Engineering Co Ltd filed Critical China National Petroleum Corp
Priority to CN202110507186.XA priority Critical patent/CN115329468A/en
Publication of CN115329468A publication Critical patent/CN115329468A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Hardware Design (AREA)
  • Pure & Applied Mathematics (AREA)
  • Biophysics (AREA)
  • Mathematical Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computational Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Optimization (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a drilling machine drilling rate prediction method based on a BAS algorithm improved BP neural network, which comprises the following steps: step one, drilling parameters are obtained; step two, noise reduction processing is carried out on the drilling parameters; step three, screening drilling parameters; step four, normalization processing is carried out on the screened parameters; step five, grouping the data; step six, determining a BP neural network architecture; step seven, optimizing the BP neural network by the BAS; step eight, training the model; step nine, testing the model; step ten, evaluating the model; and step eleven, storing the model. The drilling speed prediction method of the drilling machine takes the field measured data as the basis, takes a mixed algorithm as a means and takes a computer as a solving tool, and has important significance for improving the drilling efficiency and reducing the drilling cost.

Description

Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm
Technical Field
The invention belongs to the field of petroleum exploration and development and drilling engineering, and particularly relates to a drilling machine drilling speed prediction method for improving a BP neural network based on a BAS algorithm.
Background
In the development process of petroleum and natural gas, drilling engineering is a vital part, and the improvement of drilling efficiency and the reduction of drilling cost are one of the most important research contents in the field of drilling engineering. The mechanical drilling speed is an important index for evaluating the drilling efficiency, and the influence factors of the mechanical drilling speed are numerous and can be mainly divided into two main categories: one is an uncontrollable factor which mainly characterizes formation properties and rock physical properties, such as rock type, rock drillability, shale content, pore volume and the like; the other is controllable factors, mainly drilling fluid parameters and operation parameters of the drilling process, such as drilling fluid density, viscosity, displacement, weight on bit, rotating speed, torque and the like. The accurate establishment of the model between the drilling rate and the influencing factors is important for predicting the drilling rate and optimizing the engineering parameters. The existing drilling rate prediction method of the drilling machine lacks the application of field measured data, the existing drilling rate prediction model of the drilling machine based on an intelligent algorithm is easy to fall into local optimum, and the prediction result is stable and has a phase difference.
Disclosure of Invention
The invention provides a drilling machine drilling rate prediction method based on a BAS algorithm to improve a BP neural network, which is used for overcoming the problems or at least partially solving or relieving the problems.
Therefore, the invention provides a drilling machine drilling rate prediction method for improving a BP neural network based on a BAS algorithm, which comprises the following steps:
step one, drilling parameters are obtained; and collecting drilling parameters in the drilling construction process according to the sampling period.
Step two, noise reduction processing is carried out on the drilling parameters;
step three, screening drilling parameters;
step four, normalization processing is carried out on the screened parameters;
step five, grouping data;
step six, determining a BP neural network architecture;
step seven, optimizing the BP neural network by the BAS; optimizing the initial weight and the threshold in the BP neural network architecture established in the sixth step by using a BAS algorithm;
step eight, training the model;
step nine, testing the model; testing the test set data in the fifth step by applying the BAS improved BP neural network model trained in the eighth step;
step ten, evaluating the model;
step eleven, storing the model; the BAS improved BP neural network model that satisfies step ten is saved.
In the second step, the drilling parameters acquired in the first step are subjected to wavelet transformation as follows:
Figure BDA0003058896590000021
in the formula, a is more than 0 and is a scale factor, and the basic wavelet is realized
Figure BDA0003058896590000022
Performing telescopic transformation; tau is a translation factor, and translation transformation of the basic wavelet on a time axis is realized without dimension; t is an integral function argument; x (t) is a continuous function to be processed, namely each drilling parameter obtained in the step one,
Figure BDA0003058896590000023
is a basic wavelet function and is used for filtering X (t); w f And (tau, a) is a continuous signal after scale transformation and translation transformation, and here, each drilling parameter after processing.
In the third step, mutual information is used as a standard of correlation analysis, mutual information analysis is carried out on the data after noise reduction processing according to a mutual information definition formula, and the calculation formula is as follows:
Figure BDA0003058896590000024
in the formula, NMI (X; Y) is mutual information value after noise reduction; h (X, Y) is the joint entropy of the variables X, Y; h (X) and H (Y) are unconditional entropies of variables X and Y respectively; the higher the mutual information value after noise reduction between the two variables obtained in step three, the stronger the correlation between the two variables, and the weaker the correlation, otherwise. In the third step, the cross correlation is analyzed by the drilling parameters in the second step, and only one of the two variables with strong correlation is reserved according to the result.
In the fourth step, the drilling parameters screened in the third step are normalized, and numerical values of all the drilling parameters are mapped between-1 and 1, so that the excessive prediction error of the BAS improved BP neural network model caused by the excessive magnitude difference of all the drilling parameters is reduced; the normalization formula is as follows:
Figure BDA0003058896590000031
in the formula, x norm Normalized for dataValue, x is the value before the data is normalized, x max 、x min Maximum and minimum values before data normalization, y max 、y min Respectively, the maximum value and the minimum value after data normalization.
And in the fifth step, the drilling parameters screened in the third step are used as input variables, the mechanical drilling speed is used as an output variable, the data are grouped according to a certain proportion, and the data are divided into two groups, namely a training set and a test set.
In the sixth step, the screening result of the third step is used as an input variable of an input layer of the BP neural network, and the number n of the middle layer is determined; establishing an initial BP neural network architecture by taking the drilling speed of a drilling machine as an output layer; setting training times m, a training target epsilon and learning efficiency mu.
And in the step eight, training the training set data in the step five by taking the training times and the training target as model training termination conditions and applying the BP neural network improved by the step seven BAS.
In step ten, a decision coefficient R is selected 2 As an evaluation index, the calculation formula is as follows:
Figure BDA0003058896590000032
in the formula (I), the compound is shown in the specification,
Figure BDA0003058896590000033
is the predicted value of the ith sample; y is i The true value of the ith sample; n is the number of samples; wherein, the smaller the values of the root mean square and the average absolute error percentage are, the better the model is; the range of the determination coefficient is [0,1 ]]The closer to 1, the better the performance of the model and vice versa.
The drilling machine drilling rate prediction method based on the BAS algorithm improved BP neural network provided by the invention takes field measured data as a basis, takes a hybrid algorithm as a means, takes a computer as a solving tool to predict a drilling rate model, and has important significance for improving the drilling efficiency and reducing the drilling cost.
Drawings
FIG. 1 is a flow chart of a drilling rig penetration rate prediction method based on a BAS algorithm to improve a BP neural network in the invention;
FIG. 2 is a comparison graph before and after denoising of field measured data in the present invention;
FIG. 3 is a graph of the results of mutual information measurements of drilling parameters in accordance with the present invention;
FIG. 4 raw data without normalization in the present invention;
FIG. 5 illustrates the normalized data of the present invention;
FIG. 6 is a diagram of neural network model training using MATLAB in the present invention;
FIG. 7 is a line graph of heating section pressure drop heat exchange calculation results in the present invention;
FIG. 8 is a line diagram of a pressure drop heat exchange resolving result of an outflow section in the invention;
FIG. 9 BAS-improved BP neural network model decision coefficient values in the present invention.
Detailed Description
The invention is further illustrated by the following specific examples.
Referring to fig. 1, the invention provides a drilling machinery drilling rate prediction method for improving a BP neural network based on a BAS algorithm, comprising the following steps:
step one, drilling parameters are obtained; and collecting drilling parameters in the drilling construction process according to the sampling period. The data collected by the method comprises the following steps: the drilling speed of a machine, the rotating speed, the drilling pressure, the flow, the pressure of a vertical pipe, the well depth, the well diameter, the consistency coefficient and the like.
Step two, noise reduction processing is carried out on the drilling parameters; performing wavelet transformation on the drilling parameters acquired in the step one as follows:
Figure BDA0003058896590000041
in the formula, a is more than 0 and is a scale factor, and the basic wavelet is realized
Figure BDA0003058896590000042
Performing telescopic transformation; tau is a translation factor, and translation transformation of the basic wavelet on a time axis is realized without dimension; t is an independent variable of an integral function; x (t) is a continuous function to be processed, namely each drilling parameter obtained in the step one,
Figure BDA0003058896590000043
is a basic wavelet function and is used for filtering X (t); w f And (tau, a) is a continuous signal after scale transformation and translation transformation, and here, each drilling parameter after processing.
As shown in fig. 2, soft threshold filtering in the wavelet filtering method is applied to perform denoising processing on the field data obtained in step one, and a decomposition result meeting the requirement is selected as an output result. The method selects the m-th layer decomposition result as the data noise reduction result, and m is selected according to the requirement, generally 3 or 4.
Step three, screening drilling parameters; referring to fig. 3, the cross-correlation analysis results, the cross information is applied as a criterion for the correlation analysis. And C, performing mutual information analysis on the data subjected to the normalization processing in the step two according to a mutual information definition formula, wherein the calculation formula is as follows:
Figure BDA0003058896590000051
where NMI (X; Y) is a normalized mutual information value (i.e., the value is normalized to between 0 and 1); h (X, Y) is the joint entropy of the variables X, Y; h (X) and H (Y) are unconditional entropies of variables X and Y respectively. The higher the mutual information value after noise reduction between the two variables obtained in step three, the stronger the correlation between the two variables, and the weaker the correlation, otherwise. In the method, X and Y are two drilling parameters for analyzing the correlation, the drilling parameters with strong linear correlation except the mechanical drilling rate are removed, and one drilling parameter is reserved.
As shown in fig. 3, the linear correlation between the flow rate and the consistency coefficient and between the riser pressure and the depth are strong, and in this case, the flow rate and the riser pressure are eliminated, so that the drilling rate, the rotation speed, the drilling pressure, the depth, the hole diameter and the consistency coefficient are drilling parameters for final screening.
Step four, performing normalization processing on the screened parameters, and mapping numerical values of various drilling parameters to a range from-1 to 1, so that the excessive prediction error of the BAS-improved BP neural network model caused by the excessive magnitude difference of various drilling parameters is reduced; referring to fig. 4 and 5, the normalization formula is as follows:
Figure BDA0003058896590000052
in the formula, x norm Is the value after data normalization, x is the value before data normalization, x max 、x min Maximum and minimum values before data normalization, y max 、y min The maximum value and the minimum value after data normalization are respectively.
Step five, grouping the data; and step three, the screened drilling parameters are input variables, the mechanical drilling speed is output variables, the data are grouped according to a certain proportion, the data are divided into two groups, namely a training set and a test set, in the case, 80% of all the data are used as a model training set, and the rest 20% of all the data are used as a model test set.
Step six, determining a BP neural network architecture; referring to fig. 6, the screening result of step three is the input variables of the input layer of the BP neural network, i.e., the rotation speed, the weight on bit, the well depth, the well diameter and the consistency coefficient; determining the number n of the middle layers; establishing an initial BP neural network architecture by taking the drilling speed of a drilling machine as an output layer; setting training times m, a training target epsilon and learning efficiency mu. In this case n =14,m =10000, epsilon =10 6 ，μ＝0.01。
Seventhly, optimizing the BP neural network by the BAS; and (5) optimizing the initial weight and the threshold in the BP neural network architecture established in the step five by using a BAS algorithm.
Step eight, training the model; referring to fig. 6, the training set data in step five is trained by applying the BP neural network modified by step seven BAS with the training times and the training target as the model training termination conditions.
Step nine, testing the model; referring to fig. 7, the test set data in step five is tested by applying the BAS improved BP neural network model trained in step eight.
Step ten, evaluating the model; referring to FIG. 8, a decision coefficient (R) is selected 2 ) As an evaluation index, the calculation formula is as follows:
Figure BDA0003058896590000061
in the formula (I), the compound is shown in the specification,
Figure BDA0003058896590000062
is the predicted value of the ith sample; y is i The real value of the ith sample; n is the number of samples. Wherein, the smaller the values of the root mean square and the average absolute error percentage are, the better the model is; the range of the determination coefficient is [0,1 ]]The closer to 1, the better the performance of the model and vice versa.
σ is the minimum value of the coefficient of determination set to meet the demand, when R 2 When the model is larger than sigma, the established BAS improved BP neural network model is the model established by the method. If R is 2 If the sum of the parameters is less than sigma, modifying the number of middle layers of the model in the step six, and repeating the steps six to nine until R 2 Meets the requirements.
In the case, σ =0.9 is selected, and referring to fig. 9,0.948 > 0.9, it is obvious that the BAS improved BP neural network established by the scheme meets the requirement.
Step eleven, storing the model; the BAS improved BP neural network model that satisfies step ten is saved.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims (8)

1. A drilling machine drilling rate prediction method for improving a BP neural network based on a BAS algorithm is characterized by comprising the following steps:
step one, drilling parameters are obtained; and collecting drilling parameters in the drilling construction process according to the sampling period.
Step two, noise reduction processing is carried out on the drilling parameters;
step three, screening drilling parameters;
step four, carrying out normalization processing on the screened parameters;
step five, grouping the data;
step six, determining a BP neural network architecture;
step seven, optimizing the BP neural network by the BAS; optimizing the initial weight and the threshold in the BP neural network architecture established in the sixth step by using a BAS algorithm;
step eight, training the model;
step nine, testing the model; testing the test set data in the fifth step by applying the BAS improved BP neural network model trained in the eighth step;
step ten, evaluating the model;
step eleven, storing the model; the BAS improved BP neural network model that satisfies step ten is saved.
2. The method of claim 1, wherein in step two, the drilling parameters obtained in step one are wavelet transformed as follows:
Figure FDA0003058896580000011
in the formula, a is more than 0 and is a scale factor, and the basic wavelet is realized
Figure FDA0003058896580000012
Performing telescopic transformation; tau is a translation factor, and realizes translation transformation of basic wavelet on a time axis without any translation transformationDimension; t is an integral function argument; x (t) is a continuous function to be processed, namely each drilling parameter obtained in the step one,
Figure FDA0003058896580000013
is a basic wavelet function and is used for filtering X (t); w f And (tau, a) is a continuous signal after scale transformation and translation transformation, and here, each drilling parameter after processing.
3. The BAS algorithm-based drilling rig penetration rate prediction method for improving the BP neural network according to claim 1, wherein in step three, mutual information is applied as a standard for correlation analysis, and mutual information analysis is performed on the noise-reduced data according to a mutual information definition formula, wherein the calculation formula is as follows:
Figure FDA0003058896580000021
in the formula, NMI (X; Y) is mutual information value after noise reduction; h (X, Y) is the joint entropy of the variables X, Y; h (X) and H (Y) are unconditional entropies of variables X and Y respectively; the higher the mutual information value obtained in the third step after the noise reduction between the two variables is, the stronger the correlation between the two variables is, otherwise, the weaker the correlation is; in the third step, the cross correlation is analyzed by the drilling parameters in the second step, and only one of the two variables with strong correlation is reserved according to the result.
4. The drilling machinery drilling rate prediction method for improving the BP neural network based on the BAS algorithm as claimed in claim 1, wherein the drilling parameters screened in the third step are normalized in the fourth step, and numerical values of all drilling parameters are mapped to a range from-1 to 1, so as to reduce the situation that prediction errors of the BAS improved BP neural network model are too large due to too large magnitude differences of all drilling parameters, and the normalization formula is as follows:
Figure FDA0003058896580000022
in the formula, x norm Is the value after data normalization, x is the value before data normalization, x max 、x min Maximum and minimum values before data normalization, y max 、y min Respectively, the maximum value and the minimum value after data normalization.
5. The BAS algorithm-based drilling rate prediction method for improving the BP neural network according to claim 1, wherein in the fifth step, the drilling parameters screened in the third step are input variables, the drilling rate is output variables, and the data are grouped according to a certain proportion and are divided into two groups, namely a training set and a test set.
6. The BAS algorithm-based drilling rig penetration rate prediction method for improving the BP neural network according to claim 1, wherein in step six, the screening result of step three is used as an input variable of an input layer of the BP neural network to determine the number n of intermediate layer layers; establishing an initial BP neural network architecture by taking the drilling speed of a drilling machine as an output layer; setting training times m, a training target epsilon and learning efficiency mu.
7. The drilling machine drilling rate prediction method based on the BAS algorithm-improved BP neural network as claimed in claim 1, wherein in step eight, training set data in step five is trained by applying a BP neural network improved by step seven BAS with training times and training targets as model training termination conditions.
8. The method of claim 1, wherein in step ten, a decision factor R is selected 2 As an evaluation index, the calculation formula is as follows:
Figure FDA0003058896580000031
in the formula (I), the compound is shown in the specification,
Figure FDA0003058896580000032
is the predicted value of the ith sample; y is i The true value of the ith sample; n is the number of samples; wherein, the smaller the values of the root mean square and the average absolute error percentage are, the better the model is; the range of the determination coefficient is [0,1 ]]The closer to 1, the better the performance of the model, and vice versa.
CN202110507186.XA 2021-05-10 2021-05-10 Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm Pending CN115329468A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110507186.XA CN115329468A (en) 2021-05-10 2021-05-10 Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110507186.XA CN115329468A (en) 2021-05-10 2021-05-10 Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm

Publications (1)

Publication Number Publication Date
CN115329468A true CN115329468A (en) 2022-11-11

Family

ID=83912774

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110507186.XA Pending CN115329468A (en) 2021-05-10 2021-05-10 Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm

Country Status (1)

Country Link
CN (1) CN115329468A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117349798A (en) * 2023-12-05 2024-01-05 西南石油大学 Unbalanced regression-based mechanical drilling rate prediction method and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117349798A (en) * 2023-12-05 2024-01-05 西南石油大学 Unbalanced regression-based mechanical drilling rate prediction method and system
CN117349798B (en) * 2023-12-05 2024-02-23 西南石油大学 Unbalanced regression-based mechanical drilling rate prediction method and system

Similar Documents

Publication Publication Date Title
CN112901137B (en) Deep well drilling mechanical drilling speed prediction method based on deep neural network Sequential model
CN111046341B (en) Unconventional natural gas fracturing effect evaluation and productivity prediction method based on principal component analysis
CN112487582B (en) Oil-gas drilling machinery drilling speed prediction and optimization method based on CART algorithm
CN111425193B (en) Reservoir compressibility evaluation method based on clustering analysis logging rock physical facies division
CN109214026A (en) A kind of shale gas horizontal well initial productivity prediction technique
CN113553780B (en) Stratum pore pressure prediction method based on machine learning
CN114861563B (en) Method, device, medium and equipment for predicting formation pressure in physical embedding deep learning
CN112257254B (en) Stratum drillability evaluation method based on grey prediction
CN108595803A (en) Shale gas well liquid loading pressure prediction method based on recurrent neural network
CN114358434A (en) Drilling machine drilling speed prediction method based on LSTM recurrent neural network model
CN111767674A (en) Well logging lithology identification method based on active domain adaptation
CN110288233A (en) A kind of deep layer shale gas compressibility evaluation method based on fuzzy gray correlation theroy method
CN116644284A (en) Stratum classification characteristic factor determining method, system, electronic equipment and medium
CN115329468A (en) Drilling machine drilling speed prediction method for improving BP neural network based on BAS algorithm
CN116384554A (en) Method and device for predicting mechanical drilling speed, electronic equipment and computer storage medium
CN113627639A (en) Well testing productivity prediction method and system for carbonate fracture-cave reservoir
CN113762614A (en) Shale gas well productivity prediction method based on static oil and gas field data
CN114991757A (en) Formation pressure measurement and recording integrated monitoring method and system while drilling
Anifowose et al. A functional networks-type-2 fuzzy logic hybrid model for the prediction of porosity and permeability of oil and gas reservoirs
CN115049173B (en) Deep learning and Eaton method coupling driving stratum pore pressure prediction method
CN111625916A (en) Method and system for calculating stability value of well wall
CN116432891A (en) Comprehensive evaluation method and system for application efficiency of drill bit
CN112465347A (en) Method for cooperatively predicting roof stability based on cluster analysis and improved neural network
Feng et al. Interpretable Lost Circulation Analysis: Labeled, Identified, and Analyzed Lost Circulation in Drilling Operations
CN113627640A (en) Productivity well testing prediction method and system for fracture-cavity type oil reservoir oil and gas well

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination