CN115906310A - Drilling parameter optimization method based on variational self-encoder artificial intelligence model - Google Patents
Drilling parameter optimization method based on variational self-encoder artificial intelligence model Download PDFInfo
- Publication number
- CN115906310A CN115906310A CN202211400389.XA CN202211400389A CN115906310A CN 115906310 A CN115906310 A CN 115906310A CN 202211400389 A CN202211400389 A CN 202211400389A CN 115906310 A CN115906310 A CN 115906310A
- Authority
- CN
- China
- Prior art keywords
- data
- drilling
- encoder
- parameters
- parameter
- 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
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 116
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000005457 optimization Methods 0.000 title claims abstract description 26
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000012216 screening Methods 0.000 claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 239000012530 fluid Substances 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000010606 normalization Methods 0.000 claims description 14
- 230000004913 activation Effects 0.000 claims description 13
- 238000012217 deletion Methods 0.000 claims description 10
- 230000037430 deletion Effects 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 235000004257 Cordia myxa Nutrition 0.000 claims description 6
- 244000157795 Cordia myxa Species 0.000 claims description 6
- 238000009432 framing Methods 0.000 claims description 6
- 239000011435 rock Substances 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000012886 linear function Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000035515 penetration Effects 0.000 claims description 2
- 238000011161 development Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention relates to a drilling parameter optimization method based on a variational self-encoder artificial intelligence model, which comprises the following steps: firstly, acquiring drilling engineering parameter data, logging data and drilling fluid performance data; and carrying out data preprocessing on the acquired well history data, wherein the data preprocessing comprises data screening, outlier processing and data standardization processing, and finally obtaining the preprocessed drilling engineering parameter data, logging data and drilling fluid performance data. And taking the preprocessed data as input, constructing and training an artificial intelligence model of the unsupervised variational auto-encoder, and outputting a hidden variable spatial feature variable. And (4) correlating the implicit variable space characteristic variable with the drilling rate to realize the recommendation of the optimal drilling parameter combination from the target drilling rate to the implicit variable space characteristic variable. The method can provide the recommendation of the optimal drilling parameters under different well sections more accurately and more conveniently, thereby ensuring the smooth development of drilling speed and efficiency improvement.
Description
Technical Field
The invention relates to the field of drilling parameter optimization of drilling engineering, in particular to a drilling parameter optimization method based on a variational self-encoder artificial intelligence model.
Background
The development of drilling engineering technology to date, the speed and efficiency improvement is one of the goals pursued by drilling personnel. The key points of speed-up and efficiency-up are two points, the first equipment and the first tool are used as the basis of speed-up and play a vital role in the process of speed-up and efficiency-up; secondly, under the current conditions of tools and equipment, the matching of the equipment, the tools and drilling parameters is improved, the potential of the equipment and the tools is fully exerted, and the gripper is one of the key grippers for improving the speed and the efficiency. Currently, the problem of low matching of drilling parameters to drilling equipment exists. The problem of low matching of drilling parameters seriously restricts the drilling speed, and the serious problem can cause damage and failure of downhole tools and other downhole complex conditions, and the like, thereby causing great economic loss. For this reason, improving the matching between equipment, tools and drilling parameters has become one of the concerns of petroleum workers at home and abroad, and has also been one of the effective ways to improve the speed and efficiency.
As one of the key points of speed increase, the main methods for researching drilling parameter optimization currently include mechanical specific energy and drilling rate equation. The mechanical specific energy is obtained by calculating the rock breaking energy of the drill bit and judging the bottom hole working condition of the drill bit according to the rock breaking energy, so that drilling parameters such as the bit pressure, the rotating speed, the discharge capacity and the like are optimized; the drilling rate equation is to fit the regression relationship between drilling parameters such as the bit pressure, the drilling rate and the displacement, logging data and the like and the mechanical drilling rate for the roller bit or the PDC bit by a multiple regression means, and to determine the drilling parameter combination under different target drilling rates through the regression equation. However, the two methods still have the problems of overlarge human intervention and strong subjectivity in the field practical application process.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a drilling parameter optimization method based on a variational self-encoder artificial intelligence model, which can more accurately and conveniently recommend optimal drilling parameters under different well sections, thereby ensuring smooth development of drilling speed and efficiency improvement.
The technical scheme adopted by the invention for solving the technical problems is as follows: a drilling parameter optimization method based on a variational self-encoder artificial intelligence model is constructed, and the method comprises the following steps:
s1, obtaining 14 parameters required by a predicted leaking stoppage formula by arranging acquired well history data: bulk modulus, young's modulus, poisson's ratio, uniaxial compressive strength of rock, maximum horizontal principal stress, minimum horizontal principal stress, shear modulus, formation type, weight-on-bit, rotation speed, displacement, torque, drilling fluid density, and drilling fluid viscosity;
s2, carrying out data preprocessing on the acquired well history data, wherein the data preprocessing content comprises linear correlation data deletion, outlier processing and data normalization processing, and finally obtaining preprocessed engineering parameter data;
s3, taking the preprocessed data as input, and establishing and optimizing an unsupervised variational self-encoder artificial intelligence drilling parameter optimization model by taking the reconstruction error and the KL loss which are stable and minimum as standards;
s4, after the training of the variational self-encoder is finished, outputting an implicit variable parameter by the variational self-encoder, correspondingly framing the target mechanical drilling speed range and the implicit variable, and giving a corresponding drilling parameter combination according to the implicit variable parameter after framing by the variational self-encoder;
and S5, screening according to the geological parameter range of the specific interval and the preset engineering parameter range, and finally obtaining the recommended drilling parameter combination at the target drilling speed.
According to the scheme, in the step S2, the specific steps of deleting the data linearly related to the data are as follows:
calculating the correlation between the parameters by adopting a spearman correlation coefficient to obtain the correlation between the parameters;
evaluating the linear correlation among the parameters according to the correlation among the characteristic parameters, and rejecting the data with high linear correlation;
the specific steps of outlier processing in data preprocessing are as follows: performing outlier deletion processing on the features, and deleting standard deviation data deviating from the mean value by three times;
the normalization processing in the data processing comprises the following specific steps:
preprocessing data by adopting a maximum and minimum normalization method;
the maximum-minimum normalization method is to subtract the minimum value in the data from the data and then divide the data by the difference between the maximum value and the minimum value, and is as follows:
in the formula, X i As raw data, X norm For new data, X max For the maximum data value, X, in the feature column min Is the most significant in the feature columnA small data value.
According to the scheme, in the step S3, the method for establishing and optimizing the unsupervised variational self-encoder artificial intelligence drilling parameter optimization model comprises the following steps:
determining the number of hidden layer layers of a fully-connected neural network model in an encoder and a decoder, determining a fully-connected neural network activation function in the decoder and the encoder by using preprocessed data as encoder input data, realizing the nonlinear fitting of multidimensional data to low-dimensional data of a hidden variable space, and simultaneously comparing decoder output layer data with encoder input layer data to determine the fitting precision; and determining a gradient optimization algorithm, improving the training precision of the model, accelerating KL loss and minimizing reconstruction loss.
According to the scheme, the number of layers of the hidden layer of the fully-connected neural network model in the encoder and the decoder is determined by an empirical method.
According to the scheme, the activation function of the fully-connected neural network adopts a Selu activation function, and the activation function is as follows:
in the formula, λ and α are preset parameters of the model, x is the output value of the hidden layer node, selu (x) is the value processed by the hidden layer activation function, and the output layer adopts a linear function.
According to the scheme, the nonlinear fitting of the multidimensional data to the low-dimensional data of the hidden variable space is realized by performing normal distribution sampling on the output value of the encoder to obtain the hidden variable low-dimensional data.
According to the scheme, a gradient optimization algorithm is determined, the model training precision is improved, the KL loss is accelerated, and the reconstruction loss tends to be minimum, wherein the trained objective function is as follows:
wherein,it is used to characterize the closeness of the hidden variable distribution to the standard normal distribution, the reconstruction loss = - | q (z | x) log p (x | z) dz, which is used to characterize the closeness of the data set output by the hidden variable spatial upscaling to the original data set.
According to the scheme, in the step S4, the corresponding framing of the target mechanical drilling speed range and the hidden variable is carried out, and the corresponding drilling parameter combination is given through the variational self-encoder according to the hidden variable parameter after the framing range, and the specific steps are as follows:
determining a corresponding hidden variable parameter range in a hidden variable space by selecting a target mechanical drilling rate range; randomly selecting hidden variable points in a corresponding hidden variable space range, wherein the number of the needed hidden variable points can be set; through a decoder of the variational self-encoder, the hidden variable points can be subjected to one-to-one ascending dimension to the dimension of the actual drilling parameter combination, and the preliminary recommendation of the drilling parameter combination is realized.
According to the scheme, in the step S5, the specific steps of screening according to the geological parameter range of the specific interval and the preset engineering parameter range are as follows:
the geological screening is to delete the parameter combination of which the logging parameters exceed the original logging data range of the set stratum in the recommended drilling parameter combination;
the engineering screening is to carry out secondary screening in engineering on the drilling parameter combination after the geological screening; firstly, a recommended drilling combination is brought in through a theoretical model of traditional drilling speed calculation to calculate; deleting the recommended combinations with the target drilling speed difference larger than the user expectation; and (4) screening the reserved drilling parameter combinations again according to the application range of the engineering machinery.
The drilling parameter optimization method based on the variational self-encoder artificial intelligence model has the following beneficial effects:
according to the method, the drilling parameters are optimized by using an artificial intelligence method, the optimal drilling parameter recommendation under different well sections can be provided more accurately and more conveniently, the smooth development of drilling speed improvement and efficiency improvement is further ensured, and the problems of poor matching of the drilling parameters and strong optimization subjectivity of the drilling parameters in the drilling process are solved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a data processing block diagram of the drilling parameter optimization method of the present invention based on a variational self-encoder artificial intelligence model;
FIG. 2 is a schematic diagram of a variational self-coder model structure;
FIG. 3 is a flow chart of geological screening and engineering screening after a variational self-encoder recommends a combination of drilling parameters.
Detailed Description
For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
Fig. 1 is a block diagram of data preprocessing in the technical solution of the present invention, where the data preprocessing includes linear correlation data deletion, outlier processing, and normalization processing, where the linear correlation data deletion is to check the correlation between data, and since an encoder and a decoder in a model are all connected to a neural network, the highly co-linear characteristic cannot provide new information, and occupies a calculated amount, and the highly co-linear data deletion is performed by evaluating the size of linear correlation; the outlier processing means that under the premise of keeping the original appearance of the data as much as possible, the data is smoothened to the maximum extent, and outliers are deleted; the content of the data normalization process is to map all features of the data to the same scale through Min-Max standardization.
Fig. 2 is a schematic diagram of a neural network structure in the technical solution of the present invention, in this example, in order to fully consider the influence of the formation condition, the drilling parameters, and the drilling fluid performance on the rate of penetration of the machine, the preprocessed logging data, the drilling engineering parameters, and the drilling fluid performance parameter data are used as inputs, and a variational self-encoder model is trained to satisfy the nonlinear projection from a high-dimensional drilling parameter combination to a low-dimensional implicit variable space.
The drilling parameter optimization method based on the variational self-encoder artificial intelligence model comprises the following steps:
1) And acquiring historical data of the plugging formula well based on data mining.
The 14 parameters required by the optimization of the drilling parameters are obtained by arranging the collected drilling parameters, logging data and drilling fluid performance data: bulk modulus, young's modulus, poisson's ratio, uniaxial compressive strength of rock, maximum horizontal principal stress, minimum horizontal principal stress, shear modulus, formation, weight-on-bit, rotational speed, displacement, torque, drilling fluid density, drilling fluid viscosity; and using the collected data as a standard sample of the training model.
2) And carrying out data preprocessing on the acquired data, wherein the data preprocessing content comprises linear correlation data deletion, outlier processing and normalization processing, and finally obtaining the preprocessed data.
The above data preprocessing is further characterized by:
2-1) the data preprocessing is to complete data preprocessing by analyzing a raw data set, constructing a data cleaning model and an algorithm so as to ensure the training precision of the artificial intelligence model and the application stability.
Aiming at the data preprocessing, the data preprocessing mainly comprises linear correlation data deletion, outlier processing and normalization processing, and is an indispensable link before the artificial intelligence model is constructed. The method mainly comprises the following processing steps:
2-2-1) because the data characteristic column may not meet the normal distribution requirement, calculating the correlation between the characteristic columns by using a spearman correlation coefficient, and deleting the highly linear correlation characteristic column;
2-2-2) aiming at the data of the deleted linear correlation columns, in order to improve the signal to noise ratio, ensure the high robustness of the model and improve the generalization capability of the model, outlier deletion processing is carried out on the characteristics, and standard deviation data deviating from the mean value by three times is deleted from the data.
2-2-3) in order to improve the model precision and accelerate the convergence rate of model training, a maximum and minimum normalization method is adopted to preprocess data;
the maximum-minimum normalization method is to subtract the minimum value in the data from the data and then divide the data by the difference between the maximum value and the minimum value, and is as follows:
in the above formula, wherein X i As raw data, X norm For new data, X max For the maximum data value in the feature column, X min Is the smallest data value in the feature column.
3) Taking the preprocessed characteristic data as input, and constructing an unsupervised artificial intelligence variational self-encoder model;
the technical scheme is further characterized in that the construction scheme of the variational self-coder model is as follows;
3-1) the variational self-encoder consists of an encoder and a decoder, which are all fully connected neural networks. Further, the encoder is a 4-layer fully-connected neural network, and the decoder is a 3-layer fully-connected neural network. And determining to use a gradient optimization algorithm, improving the model training precision, accelerating KL loss and minimizing reconstruction loss, wherein a trained objective function is as follows:
whereinIt is used to characterize the closeness of the hidden variable distribution to the standard normal distribution, the reconstruction loss = - | q (z | x) log p (x | z) dz, which is used to characterize the closeness of the data set output by the hidden variable spatial ascending dimension to the original data set. />
3-2) determining an activation function. The activation function of the hidden layer in the encoder and the decoder adopts a Selu activation function, and the output layer adopts a linear function.
Wherein, λ and α are preset parameters of the model, x is the output value of the hidden layer node, and selu (x) is the value processed by the hidden layer activation function. The output layer uses a linear function.
And 3-3) determining the statistical mean and variance characteristics of the input combined distribution through an encoder, and sampling the input combined distribution in a standard normal distribution to obtain hidden variable parameters under a low dimensionality.
And 3-4) obtaining an implicit variable space of the known mechanical drilling rate and the implicit variable parameters based on the obtained implicit variable parameters.
Fig. 3 is a drilling parameter combination screening implementation flow according to the technical scheme of the invention, which specifically comprises the following steps:
4) And (3) corresponding the hidden variable parameters given by the encoder with the known drilling rate, and further giving a corresponding hidden variable interval range in the target drilling rate range.
5) And in the range of the hidden variable interval, randomly selecting hidden variable points, and outputting the hidden variable points as the input of a decoder in the variational self-encoder and the output of the hidden variable points as the combination of the unscreened drilling parameters.
6) And performing geological screening in the drilling parameter combination. Drilling parameters that are outside of the known formation properties (bulk modulus, young's modulus, poisson's ratio, uniaxial rock compressive strength, maximum horizontal principal stress, minimum horizontal principal stress, shear modulus, formation type) are combined and removed. To ensure that the recommended parameter combinations are appropriate for the target formation.
7) And (4) combining the drilling parameters screened by the step 6) and inputting the drilling parameters into a traditional drilling speed calculation model so as to verify the drilling speed value recommended by the drilling parameter combination. And deleting the drilling parameter combinations exceeding the user allowable error.
8) And (4) screening the drilling parameter combinations screened in the step (7) according to objective applicable conditions of drilling mechanical equipment, and reserving the parameter combinations within the range of engineering construction capacity.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (9)
1. A drilling parameter optimization method based on a variational self-encoder artificial intelligence model is characterized by comprising the following steps:
s1, obtaining 14 parameters required by a prediction plugging formula by sorting collected well history data: bulk modulus, young's modulus, poisson's ratio, uniaxial compressive strength of rock, maximum horizontal principal stress, minimum horizontal principal stress, shear modulus, formation type, weight-on-bit, rotation speed, displacement, torque, drilling fluid density, and drilling fluid viscosity;
s2, carrying out data preprocessing on the acquired well history data, wherein the data preprocessing content comprises linear correlation data deletion, outlier processing and data normalization processing, and finally obtaining preprocessed engineering parameter data;
s3, taking the preprocessed data as input, and establishing and optimizing an unsupervised variational self-encoder artificial intelligence drilling parameter optimization model by taking the reconstruction error and the KL loss which are stable and minimum as standards;
s4, after the training of the variational self-encoder is finished, outputting an implicit variable parameter by the variational self-encoder, correspondingly framing the target mechanical drilling speed range and the implicit variable, and giving a corresponding drilling parameter combination according to the implicit variable parameter after framing by the variational self-encoder;
and S5, screening according to the geological parameter range of the specific interval and the preset range of the engineering parameters, and finally obtaining the recommended drilling parameter combination at the target drilling speed.
2. The method for optimizing drilling parameters based on the variational self-encoder artificial intelligence model according to claim 1, wherein in the step S2, the specific steps of deleting the data linearly related to the drilling parameters are as follows:
calculating the correlation between the parameters by adopting a spearman correlation coefficient to obtain the correlation between the parameters;
evaluating the linear correlation among the parameters according to the correlation among the characteristic parameters, and rejecting data with high linear correlation;
the specific steps of outlier processing in data preprocessing are as follows: performing outlier deletion processing on the features, and deleting standard deviation data deviating from the mean value by three times;
the normalization processing in the data processing comprises the following specific steps:
preprocessing data by adopting a maximum and minimum normalization method;
the maximum-minimum normalization method is to subtract the minimum value in the data from the data and then divide the data by the difference between the maximum value and the minimum value, and is as follows:
in the formula, X i As raw data, X norm For new data, X max For the maximum data value, X, in the feature column min Is the smallest data value in the feature column.
3. The method for optimizing drilling parameters based on variational self-encoder artificial intelligence model according to claim 1, wherein in step S3, the method for establishing and optimizing unsupervised variational self-encoder artificial intelligence drilling parameters optimization model is:
determining the number of hidden layer layers of a fully-connected neural network model in an encoder and a decoder, determining a fully-connected neural network activation function in the decoder and the encoder by using preprocessed data as encoder input data, realizing the nonlinear fitting of multidimensional data to low-dimensional data of a hidden variable space, and simultaneously comparing decoder output layer data with encoder input layer data to determine the fitting precision; and determining a gradient optimization algorithm, improving the training precision of the model, accelerating KL loss and minimizing reconstruction loss.
4. The method of claim 3, wherein the number of layers of the neural network model in the encoder and the decoder are determined empirically.
5. The method of claim 3, wherein the fully-connected neural network activation function is a Selu activation function, and is represented by the following formula:
in the formula, λ and α are preset parameters of the model, x is an output value of a hidden layer node, selu (x) is a value processed by an activation function of the hidden layer, and the output layer adopts a linear function.
6. The method for optimizing drilling parameters based on the variational self-encoder artificial intelligence model according to claim 3, wherein the nonlinear fitting of the multidimensional data to the low-dimensional data of the hidden variable space is realized by performing normal distribution sampling on the encoder output values to obtain the hidden variable low-dimensional data.
7. The drilling parameter optimization method based on the variational self-encoder artificial intelligence model according to claim 3, wherein in the method of determining the gradient optimization algorithm, improving the model training precision, accelerating the KL loss and minimizing the reconstruction loss, the trained objective function is as follows:
wherein,it is used for representing the approximation degree of hidden variable distribution and standard normal distribution and rebuilding lossLoss = - < q (z | x) logp (x | z) dz, which is used to characterize how close a dataset output by a hidden variable spatial dimensionality is to the original dataset.
8. The drilling parameter optimization method based on the variational auto-encoder artificial intelligence model according to claim 1, wherein in the step S4, the objective rate of penetration range and the hidden variable are correspondingly framed, and the variational auto-encoder is used again to give the corresponding drilling parameter combination according to the hidden variable parameter after the framed range, which comprises the specific steps of:
determining a corresponding hidden variable parameter range in a hidden variable space by selecting a target mechanical drilling rate range; randomly selecting hidden variable points in the corresponding hidden variable space range, wherein the number of the needed hidden variable points can be set; through a decoder of the variational self-encoder, the hidden variable points can be subjected to one-to-one ascending dimension to the dimension of the actual drilling parameter combination, and the preliminary recommendation of the drilling parameter combination is realized.
9. The drilling parameter optimization method based on the variational self-encoder artificial intelligence model according to claim 1, wherein in the step S5, the specific steps of screening according to the geological parameter range of the specific interval and the preset range of the engineering parameters are as follows:
the geological screening is to delete the parameter combination of which the logging parameters exceed the original logging data range of the set stratum in the recommended drilling parameter combination;
the engineering screening is to carry out secondary screening in engineering on the drilling parameter combination after the geological screening; firstly, a recommended drilling combination is brought in through a theoretical model of traditional drilling speed calculation to calculate; deleting the recommended combinations with the target drilling speed difference larger than the user expectation; and (4) screening the reserved drilling parameter combinations again according to the application range of the engineering machinery.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211400389.XA CN115906310A (en) | 2022-11-09 | 2022-11-09 | Drilling parameter optimization method based on variational self-encoder artificial intelligence model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211400389.XA CN115906310A (en) | 2022-11-09 | 2022-11-09 | Drilling parameter optimization method based on variational self-encoder artificial intelligence model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115906310A true CN115906310A (en) | 2023-04-04 |
Family
ID=86482783
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211400389.XA Pending CN115906310A (en) | 2022-11-09 | 2022-11-09 | Drilling parameter optimization method based on variational self-encoder artificial intelligence model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115906310A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116582134A (en) * | 2023-07-11 | 2023-08-11 | 江苏盖亚环境科技股份有限公司 | Drilling and testing integrated equipment data processing method |
CN118211492A (en) * | 2024-05-16 | 2024-06-18 | 青岛理工大学 | Well pattern well position optimization method based on knowledge migration |
-
2022
- 2022-11-09 CN CN202211400389.XA patent/CN115906310A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116582134A (en) * | 2023-07-11 | 2023-08-11 | 江苏盖亚环境科技股份有限公司 | Drilling and testing integrated equipment data processing method |
CN116582134B (en) * | 2023-07-11 | 2023-10-13 | 江苏盖亚环境科技股份有限公司 | Drilling and testing integrated equipment data processing method |
CN118211492A (en) * | 2024-05-16 | 2024-06-18 | 青岛理工大学 | Well pattern well position optimization method based on knowledge migration |
CN118211492B (en) * | 2024-05-16 | 2024-07-23 | 青岛理工大学 | Well pattern well position optimization method based on knowledge migration |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115906310A (en) | Drilling parameter optimization method based on variational self-encoder artificial intelligence model | |
CN112901137B (en) | Deep well drilling mechanical drilling speed prediction method based on deep neural network Sequential model | |
CN108573320B (en) | Method and system for calculating final recoverable reserves of shale gas reservoir | |
CN110952978B (en) | Drilling leakage fracture width prediction method based on neural network data mining | |
US20230169244A1 (en) | Method for evaluating fracture connectivity and optimizing fracture parameters based on complex network theory | |
US20220206174A1 (en) | Correction method for micro seismic interpretation fracturing fracture parameter result | |
CN115238861B (en) | Safe drilling fluid tightness determining method based on well wall collapse degree constraint | |
CN114358434A (en) | Drilling machine drilling speed prediction method based on LSTM recurrent neural network model | |
CN113073959B (en) | Drilling well loss mechanism prediction and auxiliary decision method based on fuzzy decision tree | |
CN110671095A (en) | Intelligent while-drilling soft measurement method for formation pressure | |
CN111738371B (en) | Stratum fracture number prediction method based on random forest data mining | |
CN116384554A (en) | Method and device for predicting mechanical drilling speed, electronic equipment and computer storage medium | |
CN109083641A (en) | Measure job execution effect evaluation method in a kind of underground workover treatment | |
CN114819677A (en) | Multi-factor fusion well control risk dynamic quantitative evaluation method and system | |
CN115438823A (en) | Borehole wall instability mechanism analysis and prediction method and system | |
Cao et al. | Feature investigation on the ROP machine learning model using realtime drilling data | |
CN114065909A (en) | Well logging curve completion method based on CNN _ AB _ Bi-LSTM | |
CN117035197A (en) | Intelligent lost circulation prediction method with minimized cost | |
CN116522583A (en) | Horizontal well stratum pressure prediction method based on linear regression method | |
CN111625916A (en) | Method and system for calculating stability value of well wall | |
US20240125959A1 (en) | Method for predicting the size range of lost circulation channel based on deep learning (dl) | |
CN114298369A (en) | Rock drillability grading comprehensive prediction model construction method | |
CN111749675A (en) | Stratum drillability prediction method and system based on cascade model algorithm | |
AL-Bahadly | Statistical regression model for estimating the rate of penetration of horizontal well in Al-Halfaya oil field | |
CN116070767B (en) | Drilling fluid leakage horizon prediction method based on capsule neural network |
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 |