CN116663203A - Drilling parameter optimization method and device - Google Patents
Drilling parameter optimization method and device Download PDFInfo
- Publication number
- CN116663203A CN116663203A CN202310942767.5A CN202310942767A CN116663203A CN 116663203 A CN116663203 A CN 116663203A CN 202310942767 A CN202310942767 A CN 202310942767A CN 116663203 A CN116663203 A CN 116663203A
- Authority
- CN
- China
- Prior art keywords
- drilling
- predicted
- friction
- speed
- 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.)
- Granted
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 355
- 238000005457 optimization Methods 0.000 title claims abstract description 108
- 238000000034 method Methods 0.000 title claims abstract description 107
- 238000004364 calculation method Methods 0.000 claims abstract description 56
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 52
- 230000008569 process Effects 0.000 claims abstract description 47
- 230000001965 increasing effect Effects 0.000 claims abstract description 15
- 238000007637 random forest analysis Methods 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 28
- 238000004590 computer program Methods 0.000 claims description 27
- 230000011218 segmentation Effects 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 16
- 238000003066 decision tree Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 230000010354 integration Effects 0.000 claims description 8
- 238000013450 outlier detection Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000035515 penetration Effects 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 abstract description 15
- 230000000694 effects Effects 0.000 abstract description 14
- 238000010586 diagram Methods 0.000 description 14
- 238000012360 testing method Methods 0.000 description 9
- 230000002159 abnormal effect Effects 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000009467 reduction Effects 0.000 description 6
- 101100001678 Emericella variicolor andM gene Proteins 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 230000002068 genetic effect Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 241000219000 Populus Species 0.000 description 3
- 230000033228 biological regulation Effects 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 101100001671 Emericella variicolor andF gene Proteins 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000005452 bending Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011010 flushing procedure Methods 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000002002 slurry Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Earth Drilling (AREA)
Abstract
The invention discloses a drilling parameter optimization method and a drilling parameter optimization device, wherein the method comprises the following steps: according to drilling speed prediction parameters of the depth to be drilled, a plurality of first predicted drilling speeds are obtained through a drilling speed prediction model; acquiring a first predicted friction corresponding to each first predicted drilling rate through a friction calculation model according to the friction calculation parameters; and adopting an NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the predicted friction resistances as optimal drilling weight and rotating speed parameter combinations, and rotating speeds being the rotating speeds of the ground turntables. According to the invention, the optimal weight on bit and rotation speed parameter combination is obtained through the drilling speed prediction model and the friction calculation model, so that the optimization effect of a multi-objective optimization algorithm in the drilling process, especially on complex well type such as a large-displacement horizontal well, can be improved.
Description
Technical Field
The invention relates to the technical field of oil and gas drilling, in particular to a drilling parameter optimization method and device.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
With the development of oil and gas resources to deep and complex stratum, the number of ultra-deep wells and large-displacement horizontal wells is gradually increased, and higher requirements are put forward on drilling engineering technology. The drilling parameters are important controllable factors in the drilling process, and the scientific selection of the drilling parameters directly affects the drilling speed, the cost of petroleum drilling and the quality of the drilled well. In the current drilling process of 'one-time drilling', the setting of some controllable parameters of the actual ground is mostly set by ground personnel according to experience, and is not changed for a long time. But different stratum conditions should be matched with different drilling parameter combinations so as to achieve the purposes of improving drilling efficiency, reducing drilling risks and reducing drilling cost. When a large-displacement horizontal well is developed for one drilling, the drilling process is more complex, and drilling risks caused by pipe column factors such as drill sticking are particularly important. The traditional well drilling optimization method mostly takes drilling speed and the like as single targets, and can not meet the site requirements of safe well drilling. In addition, the traditional well drilling process optimization method is mostly based on mechanism models such as a poplar drilling rate equation, and the accuracy and efficiency of the well drilling process cannot meet the requirements after the well drilling process optimization method is applied to the field due to the complexity of the well drilling process.
Disclosure of Invention
The embodiment of the invention provides a drilling parameter optimization method, which is used for optimizing drilling parameters through a pre-constructed drilling rate prediction model and a friction calculation model to obtain an optimal drilling pressure and rotating speed parameter combination, and improving the optimization effect of a multi-objective optimization algorithm in the drilling process, especially on complex well types such as a large-displacement horizontal well, and the method comprises the following steps:
acquiring a drilling speed prediction parameter and a friction calculation parameter of the depth to be drilled in the drilling process;
according to the drilling speed prediction parameters, a plurality of first predicted drilling speeds are obtained through a pre-constructed drilling speed prediction model;
acquiring a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters;
adopting NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and first predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
The embodiment of the invention also provides a drilling parameter optimization device, which is used for optimizing drilling parameters through a pre-constructed drilling rate prediction model and a friction calculation model to obtain an optimal drilling pressure and rotating speed parameter combination, and improving the optimization effect of a multi-objective optimization algorithm in the drilling process, especially on complex wells such as a large-displacement horizontal well, and the device comprises:
the acquisition unit is used for acquiring drilling speed prediction parameters and friction calculation parameters of the depth to be drilled in the drilling process;
the drilling speed prediction unit is used for obtaining a plurality of first predicted drilling speeds through a pre-constructed drilling speed prediction model according to drilling speed prediction parameters;
a friction prediction unit for obtaining a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters;
the optimizing unit is used for carrying out multi-objective collaborative optimization on a plurality of groups of first predicted drilling rates and first predicted friction resistances which are in one-to-one correspondence to speed increasing and drag reducing to obtain an optimal group of second predicted drilling rates and second predicted friction resistances corresponding to the second predicted drilling rates, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling rates and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the drilling parameter optimization method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the drilling parameter optimization method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the drilling parameter optimization method when being executed by a processor.
In the embodiment of the invention, the drilling parameter optimization scheme works as follows: acquiring a drilling speed prediction parameter and a friction calculation parameter of the depth to be drilled in the drilling process; according to the drilling speed prediction parameters, a plurality of first predicted drilling speeds are obtained through a pre-constructed drilling speed prediction model; acquiring a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters; adopting NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and first predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotational speed parameter combination is used to guide the drilling operation of the depth to be drilled.
Compared with the technical scheme that the traditional drilling optimization method is mainly used for optimizing the drilling process by taking drilling speed and the like as single targets and based on the mechanism models such as the poplar drilling speed equation and the like, the drilling parameter optimization scheme provided by the embodiment of the invention is used for optimizing drilling parameters through the pre-constructed drilling speed prediction model and the friction calculation model to obtain the optimal drilling pressure and rotating speed parameter combination, and the optimization effect of the multi-target optimization algorithm in the drilling process, especially on complex wells such as large-displacement horizontal wells and the like can be improved, so that the drilling period is shortened, the drilling cost is reduced, the drilling quality and efficiency are improved, the drilling safety and controllability are enhanced, and the one-trip drilling technology is more efficient and safe.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a schematic diagram of training principle of random forest in the embodiment of the invention;
FIG. 2 is a schematic diagram of an NSGA-II algorithm in accordance with an embodiment of the present invention;
FIG. 3 is a schematic diagram of the effect of linear interpolation of well deviation in an embodiment of the present invention;
FIG. 4 is a schematic diagram of detecting density scan anomalies in an embodiment of the present invention;
FIG. 5 is a schematic diagram of detection of anomalies in a well depth-while-drilling density scan in accordance with an embodiment of the present invention;
FIG. 6 is a diagram of NSGA-II algorithm minimum drilling-minimum friction search results in an embodiment of the present invention;
FIG. 7 is a flow chart of a multi-objective multi-parameter collaborative optimization method for speed-up drag reduction in a drilling process in accordance with an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a multi-objective multi-parameter collaborative optimization device for speed-up and drag-reduction in a drilling process in an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
The inventors found that the technical problem is: the friction resistance represents the friction force between the drill string and the well wall and the additional resistance generated by various factors such as key grooves, rock debris beds, slurry viscosity and the like, can represent the underground state of the pipe column, is an important basis for judging the drilling sticking condition, and therefore, needs to be paid attention to in the drilling process of a large-displacement well and a horizontal well. The traditional well drilling optimization method mostly takes drilling speed and the like as single targets, friction is not considered, the site requirement of safe well drilling cannot be met, and cooperative multi-target optimization of drilling speed and friction is required.
In view of the above technical problems, the present inventors propose a multi-objective multi-parameter collaborative optimization scheme for speed-up and drag-reduction in a drilling process, which includes: integrating the data, detecting abnormal values, interpolating and the like, and establishing a complete well history-logging database; constructing an intelligent mechanical drilling speed prediction model based on a random forest algorithm; based on the hard rod model, establishing a real-time calculation model of the friction coefficient of the drill string; the method comprises the steps of taking the maximum drilling speed (drilling speed of an underground drill bit during drilling) and the minimum friction as optimization targets, taking two engineering parameters of drilling pressure (pressure applied to the drill bit) and rotating speed (rotating speed of a ground turntable during drilling) as optimization variables, and after boundary conditions are established, adopting an NSGA-II (non-dominant order genetic algorithm) multi-target optimization algorithm to obtain the optimal drilling pressure and rotating speed combination. The proper optimal combination is selected for regulation and control according to the actual regulatable and controllable conditions, so that the speed-increasing and drag-reducing multi-target multi-parameter intelligent cooperative optimization in the drilling process can be realized, the drill sticking risk of one-trip drilling operation is reduced, and the drilling efficiency is improved.
According to the scheme, the artificial intelligence technology is considered to have a better performance effect on modeling problems of high-dimensional and nonlinear complex problems than a traditional mechanism model, the artificial intelligence technology is introduced into multi-objective optimization research of drilling parameters, and the optimization of the drilling parameters is carried out through the intelligent drilling rate prediction model and the friction calculation model, so that the optimization effect of a multi-objective optimization algorithm on a large-displacement horizontal well can be improved, the drilling period is shortened, the drilling cost is reduced, the drilling quality and the drilling efficiency are improved, the drilling safety and controllability are enhanced, and the one-pass drilling technology is more efficient and safer. The following describes the multi-objective multi-parameter collaborative optimization scheme for accelerating and reducing drag in the drilling process in detail.
FIG. 7 is a schematic flow chart of a multi-objective multi-parameter collaborative optimization method for speed-up and drag reduction in a drilling process according to an embodiment of the present invention, as shown in FIG. 7, the method includes the following steps:
step 101: acquiring a drilling speed prediction parameter and a friction calculation parameter of the depth to be drilled in the drilling process;
step 102: according to the drilling speed prediction parameters, a plurality of first predicted drilling speeds are obtained through a pre-constructed drilling speed prediction model;
step 103: acquiring a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters;
step 104: adopting NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and first predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
In the embodiment of the invention, the drilling parameter optimization scheme works as follows: acquiring a drilling speed prediction parameter and a friction calculation parameter of the depth to be drilled in the drilling process; according to the drilling speed prediction parameters, a plurality of first predicted drilling speeds are obtained through a pre-constructed drilling speed prediction model; acquiring a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters; adopting NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and first predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotational speed parameter combination is used to guide the drilling operation of the depth to be drilled.
The drilling parameter optimization method provided by the embodiment of the invention can be applied to complex wells such as a large-displacement horizontal well, the well depth structure of the large-displacement well is more complex than that of a straight well, and the drilling parameter optimization method is more prone to the risk of jamming caused by overlarge friction.
Compared with the technical scheme that the traditional drilling optimization method in the prior art is mainly used for optimizing the drilling process by taking drilling speed and the like as single targets and based on the mechanism models such as the poplar drilling speed equation and the like, the drilling parameter optimization method provided by the embodiment of the invention is used for optimizing drilling parameters through the pre-constructed drilling speed prediction model and the friction calculation model to obtain the optimal drilling pressure and rotating speed parameter combination, and the optimization effect of the multi-target optimization algorithm in the drilling process, especially on complex wells such as large-displacement horizontal wells and the like can be improved, so that the drilling period is shortened, the drilling cost is reduced, the drilling quality and efficiency are improved, the drilling safety and controllability are enhanced, and the one-trip drilling technology is more efficient and safer. The multi-objective multi-parameter collaborative optimization method for accelerating and reducing drag in the drilling process is described in detail below.
The embodiment of the invention aims to provide a speed-increasing and drag-reducing multi-objective multi-parameter intelligent collaborative optimization method in a drilling process, which can be used for carrying out parameter optimization on drilling pressure (pressure applied to a drill bit) and rotating speed (rotating speed of a ground turntable in drilling) in the drilling process, so as to take the speed-increasing and drag-reducing as optimization targets and realize safe and efficient drilling, and particularly, the method has the aim of reducing cost and enhancing efficiency for one-trip drilling operation. Specifically, the method comprises the following technical scheme: firstly, optimizing a data preprocessing method, forming a perfect well history-logging data processing flow, establishing a complete database, and providing a basis for model training and prediction; then, a machine learning method is optimized, and an intelligent mechanical drilling speed prediction model is constructed; preferably, a drill string friction torque calculation model is adopted, and a real-time calculation model (friction calculation model) of the drill string friction coefficient is established; and finally, on the basis of a mechanical drilling speed model and a friction torque model, optimizing a multi-objective optimization algorithm, taking the maximum drilling speed and the minimum friction as optimization targets, setting up boundary conditions, optimizing algorithm super-parameters, outputting a plurality of optimal weight on bit and rotating speed combinations, and finally realizing the intelligent collaborative optimization of speed-increasing and drag-reducing multi-objective multi-parameters in the drilling process. The following is a detailed description.
The embodiment of the invention provides a speed-increasing and drag-reducing multi-objective multi-parameter intelligent collaborative optimization method in a drilling process.
Step 1: the well history-logging data is subjected to data preprocessing to form data processing methods including integration, outlier detection, interpolation and the like, and related data are shown in the following table 1.
TABLE 1 well history-logging integration data
The integrated data may comprise 15 data in the above tableThe characteristics basically meet the data requirements of building the intelligent model. In the data integration process, the logging data and the gamma data are data with a meter interval, and the interval of the well track data in the well history is larger, which is several meters or tens of meters. Therefore, in order to facilitate integration and ensure the quantity of data, the well deviation data is filled by a linear interpolation method, and the linear interpolation principle is as follows: suppose that the well depth and well deviation data of two points are (h) 0 ,l 0 )，(h 1 ,l 1 ) According to the linear interpolation principle, the well depth (h, l) of any point between the two points meets the following formula:
；
the interpolation effect is shown in fig. 3.
Then, abnormal point detection is carried out, and a density scanning abnormal detection method is adopted, and the principle is shown in figure 4.
(1) Setting a set X, and storing n drilling data which are continuous according to depth, namely: x= { X h1 ,x h2 ,x h3 ,...,x hn And x is max = max{x hi }，x min = min{x hi }。
(2) When data is acquired on a new well depth and drilling, i.e. x h(n+1) Taking the radius r=0.2 (x max -x min ) The density ρ=0, the data in set X is traversed if X h(n+1) -r<x hi <x h(n+1) +r, ρ=ρ+1. If ρ after traversing>0.2n, then x hn+1 Normal values, otherwise abnormal values.
(3) If x h(n+1) If the value is normal, updating the set X, removing the first data, and adding X hn+1 To the collection as the last data.
(4) Repeating (2) (3) until all data are detected.
The application effect is shown in fig. 5.
A dataset is ultimately established. After the data processing operation, a well history-logging data set is established and is used for intelligent model training and testing.
It can be seen from the foregoing that in one embodiment, the drilling parameter optimization method may further include performing data preprocessing on logging engineering parameters, gamma logging parameters, and well history data parameters, where the data preprocessing includes integration, outlier detection, and interpolation.
In the specific implementation, in order to fully consider engineering and geological factors of the intelligent model of the mechanical drilling speed, the embodiment of the invention utilizes well history data, gamma rock debris logging data and logging engineering data to combine the data existing in different documents into a data table according to depth. Including well depth, drilling time, weight on bit, sling weight, rotational speed, torque, displacement, drilling fluid equivalent density, gamma, cutting tooth diameter, blade number, tooth distribution density, etc. Meanwhile, because the logging data and the gamma data are data with an interval of one meter, the interval of the well track data in the well history is larger, and most of the logging data and the gamma data are data with an interval of several meters or tens of meters. Therefore, in order to facilitate integration and ensure the quantity of data, the well deviation data is filled by a linear interpolation method. Then, the abnormal value is removed, and data cleaning is performed.
Step 2: and constructing a mechanical drilling speed prediction model.
The drilling speed data has the characteristics of high data dimension and unbalanced distribution of the labels during drilling. The random forest algorithm randomly selects the features of the high-dimensional data during training, does not need to consider the importance degree of the features, and has strong stability. And each decision tree is constructed by only needing random partial data, so that the method is suitable for training unbalanced samples.
Random forest algorithm principle. The decision tree is a tree-shaped classifier, and can be divided into a classification tree and a regression tree according to different output variables, the drilling rate prediction problem is a regression problem, and the embodiment of the invention constructs a regression binary decision tree. When training the binary decision tree model, how to select the segmentation variable (feature) and the segmentation point and how to measure the quality of one segmentation variable and the segmentation point need to be considered. For the quality of the segmentation variable and the segmentation point, the degree of the segmentation variable and the segmentation point are generally measured by the degree of the node's non-purity after the segmentation, namely the weighted sum of the non-purity of each child node, the calculation formula (the optimization function of the random forest drilling rate prediction model) is as follows:
； （1）
wherein ,x i in order to cut the variable(s),ν ij for the segmentation value of the segmentation variable,n l the number of training samples is predicted for the drilling rate of the left child node after the segmentation,n r the number of training samples is predicted for the drilling rate of the right child node after the segmentation, N s Predicting the number of training samples for all drilling rates of the current node,X l predicting a training sample set for the drilling speed of the left child node,X r A set of training samples is predicted for the bit rate of the right child node,H(X)as a function of the measured node non-purity.
The process of training the mechanical drilling speed prediction model is as follows:
(1) Dividing the data set: dividing normal data after abnormal value processing into a training set and a testing set, wherein the proportion is 0.8:0.2.
(2) Model training: the data set is used for selecting well depth, hanging weight, drilling pressure, rotating speed of a rotary table, pressure of a vertical pipe, total pump flushing, gamma, well deviation, cutting tooth diameter, blade number and tooth distribution density as input characteristics and drilling as output, so that an initial trained model is obtained.
(3) Model prediction: and (3) introducing the test set into the model which is initially trained in the step (2), optimizing the super parameters of the model, and verifying the model to obtain the final mechanical drilling speed prediction model.
From the foregoing, in one embodiment, the drilling parameter optimization method may further include pre-constructing a drilling rate prediction model according to the following method:
acquiring logging engineering parameters, gamma logging parameters and well history data parameters, and constructing a drilling rate prediction sample set;
and based on a random forest algorithm, establishing a random forest drilling rate prediction model by utilizing the drilling rate prediction sample set.
Step 3: and constructing a friction calculation model.
Model principle: the classical drill string friction torque model comprises a soft rod model and a hard rod model, the soft rod model is early in application, simple and has certain precision, but the rigidity of the drill string is not considered, and the friction torque model is suitable for well types with smaller vertical wells and well inclination angles. The hard rod model is more complex, but considers the rigidity of the drilling column, and is more suitable for non-vertical well type wells such as large-displacement wells, horizontal wells and the like. On the basis of a classical hard rod model, a plurality of students further complement and perfect the model on the basis of the classical hard rod model, and the drill string friction torque model is developed to be more mature, so that the real-time calculation of the drill string friction coefficient is performed by adopting the model. The two parameters of "engineering parameters" and "wellbore trajectory" in table 1 above are input to the friction torque model to build the model or calculate the friction.
Differential equation of overall stress of drill string:
；
；
wherein ,“In the description, a negative sign represents lifting and a positive sign represents dropping;Fn, an axial pressure on the drill string;Sis the well depth, m;qgravity of the drill string in unit length, N/m;αis the well bevel, rad;EIn.m for flexural rigidity of drill string 2 ；n t N/m is the contact force between the drill string and the well wall;μ 1 the friction coefficient is the axial friction coefficient of the drill string; k b For borehole axis curvature, m -1 。
Curvature of wellborek b The calculation formula of (2) is as follows:
；
contact force on drill stringn t The calculation formula of (2) is as follows:
；
；
；
；
the method is characterized in that a finite difference numerical solution method is utilized to write the whole stress model of the drill string into the following form (friction calculation model):
the pressure is positive:
；（2）
；（3）
wherein ,F i andF i+1 the axial force of the ith section of drill column near the ground and near one end of the drill bit is respectively N;M Ti andM Ti+1 torque at two ends of the ith section of drill string is N.m;M bi andM bi+1 wellbore curvatures at both ends of the ith section, m -1 ；q i 、EI i 、n ti 、Δs i AndD bi the line weights (N/m) and the bending rigidities (N.m) of the i-th section of the drill string 2 ) The contact force (N/m) with the well wall, the length (m) of the unit (i.e. the length of the unit of the i-th section divided into a plurality of units) and the outer diameter (m).
From the above, in one embodiment, the friction calculation model is a friction calculation model established based on a hard bar model.
Step 4: acceleration-drag reduction collaborative optimization based on NSGA-II.
(1) Selecting an optimization function, optimizing a target, setting boundary conditions and selecting an optimization algorithm.
Optimization function: a random forest drilling speed prediction model and a tubular column friction resistance calculation model.
Optimization target: the drilling speed is the largest, namely the drilling time is the smallest, and the friction resistance is the smallest.
In the oil gas drilling process, in order to increase the drilling speed, the drilling pressure is often increased, namely the force applied to the drill bit is increased, the drill bit is allowed to eat more stratum, the drilling speed is increased to a certain extent, but the too high drilling pressure can cause the pipe column to generate serious buckling, the friction resistance of the pipe column is increased, the friction of the drill rod is caused to rub against the well wall, the friction force of the drill rod is increased, the transmission efficiency of the drilling pressure from the ground to the underground is reduced, the drilling speed is reduced instead, the drilling speed and the friction resistance are inversely related, the construction efficiency is influenced, meanwhile, the risk of jamming and the like is easy to occur due to the too high friction resistance, and accidents are caused. Therefore, the drilling speed and the friction are mutually influenced, and the friction is small while the drilling speed is kept large in the actual operation process so as to ensure safe and efficient drilling. The embodiments of the present invention set the optimization objectives to maximum rate of penetration, minimum friction, and seek a combination of parameters that satisfy a compromise of both requirements.
The speed-up-drag reduction collaborative optimization model based on NSGA-II can be:
；
wherein ,t dt min/m for the drill time predicted by the intelligent model (drilling rate intelligent prediction model);Fand N is the column friction resistance obtained by column stress calculation.
Boundary conditions: when the drilling pressure and the rotating speed in a certain well depth are respectively w and r, and the two parameters are optimized, the optimization range is w-20< w+20, and r-5< r+5.
Optimization algorithm: the NSGA-II multi-objective optimization algorithm is described above.
(2) And (3) starting optimization: setting super parameters of NSGA-II multi-objective optimization algorithm, wherein the super parameters comprise the number of individuals of each population, the iteration times of the population, the individual cross probability and the individual variation probability. In the embodiment of the invention, the population individuals refer to a plurality of initial weight-on-bit and rotating speed parameter combinations, the genetic algorithm is continuously varied to perform optimization selection on the basis of the initial population, and finally an optimal group of weight-on-bit and rotating speed parameter combinations is obtained to provide drilling operation references.
(3) Optimizing the result: n Pareto front optimal combinations can be selectively output, n is less than the population individual number, the weight on bit and rotation speed optimal combinations are obtained, and proper optimal combinations are selected for regulation and control according to the actual site regulatable conditions. The search results are shown in fig. 6.
In specific implementation, the multi-objective multi-parameter collaborative optimization method for accelerating and reducing drag in the drilling process has the beneficial technical effects that:
firstly, the intelligent drilling rate prediction model in the embodiment of the invention can be a random forest drilling rate prediction model established based on a random forest algorithm, and the specific reasons are as follows: aiming at the drilling speed prediction problem, the prediction method can use a random forest model, an XGBoost model, a support vector machine model, a neural network model and the like. However, the random forest model performs optimally on horizontal well rate of penetration prediction problems. The random forest algorithm randomly selects the characteristics of the high-dimensional data during training, does not need to consider the importance degree of the characteristics, has good stability, and is suitable for being applied to the drilling process with severe data fluctuation. The random forest is a classical integrated learning algorithm, a binary decision tree is selected as a weak learner, and the average of the results of a plurality of binary decision trees in a regression task is the result of a random forest model. The training principle of random forests is shown in fig. 1, wherein: x (N) is the whole data set, X (N) 1 ) Is the nth 1 Subsampling a constructed dataset, X (n 2 ) Is the nth 2 Subsampled constructed dataset … …, X (n n ) Is the nth n Subsampling the constructed data set.
Secondly, the friction calculation model in the embodiment of the invention is a model (drill string friction torque physical model) built based on a hard rod model, and the specific reasons are that: the classical drill string friction torque model comprises a soft rod model and a hard rod model, the soft rod model is early in application, simple and has certain precision, but the rigidity of the drill string is not considered, and the friction torque model is suitable for well types with smaller vertical wells and well inclination angles. The hard rod model is complex, but considers the rigidity of the drilling column, and is more suitable for non-vertical well type wells such as large-displacement wells, horizontal wells and the like, so the embodiment of the invention adopts the model to calculate the friction coefficient of the drilling column in real time.
Again, in the embodiment of the present invention, the multi-objective optimization algorithm is NSGA-II algorithm, which is an improved version of non-dominant ranking genetic algorithm NSGA (non-dominant ranking genetic algorithm), and this algorithm mainly solves three pain points of NSGA:
(1) high computational complexity of non-dominant ordering;
(2) the sharing parameters are difficult to determine;
(3) there is a lack of preservation elite strategies.
The NSGA-II algorithm is improved against the deficiencies of NSGA by the following three aspects:
(1) The rapid non-dominant sorting algorithm is provided, so that the complexity of calculating the non-dominant order is reduced;
(2) an elite strategy is introduced, so that the sampling space is enlarged, and the accuracy of an optimization result is improved;
(3) a congestion level and congestion level comparison operator is introduced.
Fast non-dominant ordering: fast non-dominant ordering is the process of decomposing a solution set into Pareto fronts in a different order, which can be described as:
(1) two key quantities are assigned to each solution p: the number of solutions that govern p and the set of solutions Sp that are governed by p;
(2) setting i=1, and classifying individuals with np=0 into Fi;
(3) traversing Sp for each solution p for individuals in Fi, subtracting np for each solution by one;
(4) i+=1, assigning the solution in np to Fi;
(5) repeating (3) (4) until more individuals in the solution set are classified as Fi.
And (3) calculating the crowding degree: in NSGA-II, to measure the quality of each solution in the same front, a congestion distance is assigned to each solution. The idea behind this is to have the best solution of Pareto found as dispersed as possible in the target space. There is also a greater likelihood that the solutions will be evenly distributed over the Pareto optimal front.
Elite retention strategy: ranking the individuals in the population according to the results of the fast non-dominant ranking and the crowding distance calculation: for two solutions i e Ft, j e Fs, if t < s, i is better than j; if t=s and dist [ i ] > dist [ j ], then i is better than j.
The NSGA-II not only overcomes the defect that the shared parameters are required to be manually specified in the NSGA algorithm, but also takes the crowding degree as a comparison criterion among individuals in the population, so that the population individuals in the quasi-Pareto domain can be uniformly expanded to the whole Pareto domain, the diversity of the population is ensured, as shown in fig. 2, and the effect of multi-objective multi-parameter collaborative optimization of speed-up drag reduction in the drilling process can be improved.
After the model is constructed in advance, parameters such as well depth, well deviation, weight on bit, rotating speed and the like at a certain depth (depth to be drilled) in the drilling process are acquired, an adjustable weight on bit and rotating speed range of the drilling machine is set, boundary conditions are established, and a NSGA-II algorithm is used for searching a weight on bit-rotating speed combination which enables the drilling speed (the drilling speed of an underground drill bit in drilling) to be maximum (namely minimum in drilling) and friction to be minimum in the adjustable range.
In order to facilitate an understanding of how the invention may be practiced, examples are described in their entirety.
After the data processing operation, 99720 pieces of data are collected from the 30 well logging data, logging data and well history data of the H platform, wherein each piece of data comprises 11 feature columns; of these, 1-25 wells construct 82786 training sets, 26-30 wells construct 16934 test sets, and the constructed data sets will be used for intelligent model training and testing of this platform.
The data set selects well depth, hanging weight, drilling pressure, rotating disk rotating speed, vertical pipe pressure, total pump flushing, gamma, well deviation, cutting tooth diameter, blade number and tooth distribution density as input characteristics, and the drilling time is used as output. After tuning, the random forest model parameters were selected as shown in table 2 below.
TABLE 2 random forest drilling rate prediction model superparameter selection
From the foregoing, in one embodiment, the hyper-parameters of the random forest drilling rate prediction model may be: the value range of the number of decision trees is 180-190, the value range of the maximum depth of the decision trees is 125-135, the value range of the minimum sample number required for splitting the decision tree nodes is 1-4, the value range of the number of characteristic variables considered by the best nodes is found to be 7-11, the non-purity evaluation function is a square average error MSE function, and the accuracy of rotation speed prediction can be further improved by the aid of the super-parameter random forest drilling rate prediction model.
A certain well data is selected as a test, and friction torque at the positions of 1000, 1500, 2000, 2500, 3000 and 3500 m of well depth are calculated respectively as shown in the following table 3:
TABLE 3 partial depth friction torque calculation
Selecting an optimization function, optimizing a target, setting boundary conditions, constructing an NSGA-II optimization algorithm, setting super parameters of the NSGA-II multi-target optimization algorithm, wherein the super parameters comprise the individual number of each population, the iteration number of the population, the individual cross probability and the individual variation probability, and after grid search optimization, the parameters are selected as shown in the following table 4:
Table 4 NSGA-II Multi-objective optimization Algorithm Supermameters selection
From the foregoing, in one embodiment, the super parameters of the NSGA-II multi-objective optimization algorithm may be: the value range of the population individual number is 18-22, the value range of the population iteration times is 23-27, the value range of the individual cross probability is 0.5-0.9, the value range of the individual variation probability is 0.1-0.3, and the random forest drilling rate prediction model comprising the super parameters can further improve the optimization effect.
Five depth data were selected from a well as test subjects, the test results are shown in table 5 below:
TABLE 5 acceleration-drag reduction results for certain well test points
As can be seen from table 5, the technical solution provided by the embodiment of the present invention at least includes: the drilling rate prediction model is established through the intelligent algorithm, the friction calculation model is combined for drilling parameter optimization, and the optimization effect of the multi-objective optimization algorithm on the large-displacement horizontal well can be improved, so that the drilling period is shortened, the drilling cost is reduced, the drilling quality and the drilling efficiency are improved, the drilling safety and controllability are enhanced, and the one-pass drilling technology is more efficient and safer.
The embodiment of the invention also provides a multi-objective multi-parameter cooperative optimization device for accelerating and reducing drag in the drilling process, as described in the following embodiment. Because the principle of the device for solving the problems is similar to that of a multi-target multi-parameter collaborative optimization method for accelerating and reducing drag in the drilling process, the implementation of the device can be referred to the implementation of the multi-target multi-parameter collaborative optimization method for accelerating and reducing drag in the drilling process, and the repetition is omitted.
FIG. 8 is a schematic structural diagram of a drilling parameter optimizing apparatus according to an embodiment of the present invention, as shown in FIG. 8, the apparatus includes:
an obtaining unit 01, configured to obtain a drilling rate prediction parameter and a friction calculation parameter of a depth to be drilled in a drilling process;
a drilling rate prediction unit 02, configured to obtain a plurality of first predicted drilling rates according to drilling rate prediction parameters through a pre-constructed drilling rate prediction model;
a friction prediction unit 03 for obtaining a first predicted friction corresponding to each first predicted drilling rate through a friction calculation model constructed in advance according to friction calculation parameters;
the optimizing unit 04 is used for carrying out multi-objective collaborative optimization on a plurality of groups of first predicted drilling rates and first predicted friction resistances which are in one-to-one correspondence to speed increasing and drag reducing to obtain an optimal group of second predicted drilling rates and second predicted friction resistances corresponding to the second predicted drilling rates, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling rates and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
In one embodiment, the drilling parameter optimizing apparatus may further include a first construction unit for constructing a drilling rate prediction model in advance according to the following method:
Acquiring logging engineering parameters, gamma logging parameters and well history data parameters, and constructing a drilling rate prediction sample set;
and based on a random forest algorithm, establishing a random forest drilling rate prediction model by utilizing the drilling rate prediction sample set.
In one embodiment, the optimization function of the random forest drilling rate prediction model is:
；
wherein ,x i in order to cut the variable(s),ν ij for the segmentation value of the segmentation variable,n l the number of training samples is predicted for the drilling rate of the left child node after the segmentation,n r the number of training samples is predicted for the drilling rate of the right child node after the segmentation,N s predicting the number of training samples for all drilling rates of the current node,X l predicting a training sample set for the drilling speed of the left child node,X r A set of training samples is predicted for the bit rate of the right child node,H(X)as a function of the measured node non-purity.
In one embodiment, the hyper-parameters of the random forest drilling rate prediction model are: the value range of the number of decision trees is 180-190, the value range of the maximum depth of the decision trees is 125-135, the value range of the minimum sample number required for splitting the decision tree nodes is 1-4, the value range of the number of characteristic variables considered by the best node is found to be 7-11, and the non-purity evaluation function is a square average error MSE function.
In one embodiment, the friction calculation model is a friction calculation model established based on a hard rod model.
In one embodiment, the friction calculation model is:
；
；
wherein ,F i andF i+1 the axial force of the ith section of drill column near the ground and near one end of the drill bit is respectively N;M Ti andM Ti+1 torque at two ends of the ith section of drill string is N.m;M bi andM bi+1 wellbore curvatures at both ends of the ith section, m -1 ；q i 、EI i 、n ti 、Δs i AndD bi the line weights (N/m) and the bending rigidities (N.m) of the i-th section of the drill string 2 ) The contact force (N/m) with the well wall, the length (m) of the unit (i.e. the length of the unit of the i-th section divided into a plurality of units) and the outer diameter (m).
In one embodiment, the drilling parameter optimizing apparatus may further include a preprocessing unit, configured to perform data preprocessing on logging engineering parameters, gamma logging parameters, and well history data parameters, where the data preprocessing mode includes integration, outlier detection, and interpolation processing.
In one embodiment, the boundary conditions of weight on bit and rotational speed are: when the drilling pressure and the rotating speed in the depth of the preset well are respectively w and r, and the two parameters of the drilling pressure and the rotating speed are optimized, the optimization range is w-20< w+20, and r-5< r+5.
In one embodiment, the super parameters of the NSGA-II algorithm are: the number of individuals in the population is 18-22, the number of iterations of the population is 23-27, the cross probability of the individuals is 0.5-0.9, and the variation probability of the individuals is 0.1-0.3.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the drilling parameter optimization method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the drilling parameter optimization method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program realizes the drilling parameter optimization method when being executed by a processor.
The speed-increasing and drag-reducing multi-objective multi-parameter intelligent collaborative optimization method of the drilling process is implemented by firstly integrating data, detecting abnormal values, interpolating and the like, and establishing a complete well history-logging database; constructing a mechanical drilling speed prediction model based on a random forest algorithm; based on the hard rod model, establishing a real-time calculation model of the friction coefficient of the drill string; and setting boundary conditions by taking the maximum drilling speed and the minimum friction as optimization targets, and adopting an NSGA-II multi-target optimization algorithm to perform algorithm super-parameter optimization to obtain the optimal drilling weight and rotating speed parameter combination. The proper optimal combination is selected for regulation and control according to the actual regulatable and controllable conditions, so that the speed-increasing and drag-reducing multi-target multi-parameter intelligent cooperative optimization in the drilling process can be realized, the drill sticking risk of one-trip drilling operation is reduced, and the drilling efficiency is improved.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (12)
1. A method of optimizing drilling parameters, comprising:
acquiring a drilling speed prediction parameter and a friction calculation parameter of the depth to be drilled in the drilling process;
according to the drilling speed prediction parameters, a plurality of first predicted drilling speeds are obtained through a pre-constructed drilling speed prediction model;
acquiring a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters;
adopting NSGA-II algorithm to carry out multi-objective collaborative optimization of speed increasing and drag reducing on a plurality of groups of first predicted drilling speeds and first predicted friction resistances which are in one-to-one correspondence to obtain an optimal group of second predicted drilling speeds and second predicted friction resistances corresponding to the second predicted drilling speeds, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling speeds and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
2. The method of claim 1, further comprising pre-constructing a rate of penetration prediction model according to the following method:
acquiring logging engineering parameters, gamma logging parameters and well history data parameters, and constructing a drilling rate prediction sample set;
And based on a random forest algorithm, establishing a random forest drilling rate prediction model by utilizing the drilling rate prediction sample set.
3. The method of claim 2, wherein the optimization function of the random forest bit rate prediction model is:
；
wherein ,x i in order to cut the variable(s),ν ij for the segmentation value of the segmentation variable,n l the number of training samples is predicted for the drilling rate of the left child node after the segmentation,n r the number of training samples is predicted for the drilling rate of the right child node after the segmentation,N s predicting the number of training samples for all drilling rates of the current node,X l predicting a training sample set for the drilling speed of the left child node,X r A set of training samples is predicted for the bit rate of the right child node,H (X)as a function of the measured node non-purity.
4. The method of claim 2, wherein the hyper-parameters of the random forest bit rate prediction model are: the value range of the number of decision trees is 180-190, the value range of the maximum depth of the decision trees is 125-135, the value range of the minimum sample number required for splitting the decision tree nodes is 1-4, the value range of the number of characteristic variables considered by the best node is found to be 7-11, and the non-purity evaluation function is a square average error MSE function.
5. The method of claim 1, wherein the friction calculation model is a friction calculation model established based on a hard bar model.
6. The method of any one of claims 2 to 5, further comprising: and carrying out data preprocessing on logging engineering parameters, gamma logging parameters and well history data parameters, wherein the data preprocessing mode comprises integration, outlier detection and interpolation processing.
7. The method of claim 1, wherein the boundary conditions of weight on bit and rotational speed are: and if the weight on bit and the rotating speed in the depth of the preset well are w and r respectively, when the two parameters of the weight on bit and the rotating speed are optimized, the optimization range is w-20 < w+20, and r-5 < r+5.
8. The method of claim 1, wherein the super parameters of the NSGA-II algorithm are: the number of individuals in the population is 18-22, the number of iterations of the population is 23-27, the cross probability of the individuals is 0.5-0.9, and the variation probability of the individuals is 0.1-0.3.
9. A drilling parameter optimizing apparatus, comprising:
the acquisition unit is used for acquiring drilling speed prediction parameters and friction calculation parameters of the depth to be drilled in the drilling process;
the drilling speed prediction unit is used for obtaining a plurality of first predicted drilling speeds through a pre-constructed drilling speed prediction model according to drilling speed prediction parameters;
A friction prediction unit for obtaining a first predicted friction corresponding to each first predicted drilling rate through a pre-constructed friction calculation model according to friction calculation parameters;
the optimizing unit is used for carrying out multi-objective collaborative optimization on a plurality of groups of first predicted drilling rates and first predicted friction resistances which are in one-to-one correspondence to speed increasing and drag reducing to obtain an optimal group of second predicted drilling rates and second predicted friction resistances corresponding to the second predicted drilling rates, and taking the drilling weights and the rotating speeds corresponding to the second predicted drilling rates and the second predicted friction resistances as optimal drilling weight and rotating speed parameter combinations; the optimal weight on bit and rotation speed parameter combination is used for guiding the drilling work of the depth to be drilled; the drilling speed is the drilling speed of the drill bit, and the rotating speed is the rotating speed of the ground turntable.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 8 when executing the computer program.
11. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 8.
12. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310942767.5A CN116663203B (en) | 2023-07-28 | 2023-07-28 | Drilling parameter optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310942767.5A CN116663203B (en) | 2023-07-28 | 2023-07-28 | Drilling parameter optimization method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116663203A true CN116663203A (en) | 2023-08-29 |
| CN116663203B CN116663203B (en) | 2023-10-27 |
Family
ID=87715717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310942767.5A Active CN116663203B (en) | 2023-07-28 | 2023-07-28 | Drilling parameter optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116663203B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117150828A (en) * | 2023-10-31 | 2023-12-01 | 中石化西南石油工程有限公司 | Ultra-deep well drilling column stuck-unlocking safety assessment method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110174541A1 (en) * | 2008-10-03 | 2011-07-21 | Halliburton Energy Services, Inc. | Method and System for Predicting Performance of a Drilling System |
| CN111520122A (en) * | 2020-03-27 | 2020-08-11 | 中国石油化工股份有限公司石油工程技术研究院 | Mechanical drilling speed prediction method, device and equipment |
| CN111738371A (en) * | 2020-08-26 | 2020-10-02 | 西南石油大学 | Stratum fracture number prediction method based on random forest data mining |
| CN112052875A (en) * | 2020-07-30 | 2020-12-08 | 华控清交信息科技(北京)有限公司 | Method and device for training tree model |
| CN112796747A (en) * | 2021-03-03 | 2021-05-14 | 中国石油大学(北京) | Oil and gas drilling strategy prediction method and device based on multiple targets |
| CN116151101A (en) * | 2022-12-27 | 2023-05-23 | 西南石油大学 | Horizontal well drilling parameter optimization chart establishing method |
-
2023
- 2023-07-28 CN CN202310942767.5A patent/CN116663203B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110174541A1 (en) * | 2008-10-03 | 2011-07-21 | Halliburton Energy Services, Inc. | Method and System for Predicting Performance of a Drilling System |
| CN111520122A (en) * | 2020-03-27 | 2020-08-11 | 中国石油化工股份有限公司石油工程技术研究院 | Mechanical drilling speed prediction method, device and equipment |
| CN112052875A (en) * | 2020-07-30 | 2020-12-08 | 华控清交信息科技(北京)有限公司 | Method and device for training tree model |
| CN111738371A (en) * | 2020-08-26 | 2020-10-02 | 西南石油大学 | Stratum fracture number prediction method based on random forest data mining |
| CN112796747A (en) * | 2021-03-03 | 2021-05-14 | 中国石油大学(北京) | Oil and gas drilling strategy prediction method and device based on multiple targets |
| CN116151101A (en) * | 2022-12-27 | 2023-05-23 | 西南石油大学 | Horizontal well drilling parameter optimization chart establishing method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117150828A (en) * | 2023-10-31 | 2023-12-01 | 中石化西南石油工程有限公司 | Ultra-deep well drilling column stuck-unlocking safety assessment method |
| CN117150828B (en) * | 2023-10-31 | 2024-02-27 | 中石化西南石油工程有限公司 | Ultra-deep well drilling column stuck-unlocking safety assessment method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116663203B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116663203B (en) | Drilling parameter optimization method and device | |
| CN113689055B (en) | Oil-gas drilling machinery drilling speed prediction and optimization method based on Bayesian optimization | |
| CA3093668C (en) | Learning based bayesian optimization for optimizing controllable drilling parameters | |
| CN110619353B (en) | Multi-scale logging curve automatic identification method based on deep learning | |
| CN110807557A (en) | Drilling rate prediction method based on BP neural network and drilling rate optimization method based on BP neural network and particle swarm optimization | |
| CN113326660B (en) | Tunnel surrounding rock extrusion deformation prediction method based on GA-XGboost model | |
| CN112100927A (en) | Prediction method for slope deformation and soft soil foundation settlement based on GA-BP neural network | |
| Anemangely et al. | Determination of constant coefficients of Bourgoyne and Young drilling rate model using a novel evolutionary algorithm | |
| WO2017011469A1 (en) | Ensemble based decision making | |
| CN117035197B (en) | Intelligent lost circulation prediction method with minimized cost | |
| CN114723095A (en) | Missing well logging curve prediction method and device | |
| CN114548482A (en) | Creep type landslide kinetic energy change rate face-slip early warning method | |
| CN106326624A (en) | Method for predicating stratum fracture pressure | |
| CN117786794A (en) | Shield tunneling existing tunnel deformation optimization control method and system | |
| CN116776717A (en) | Drilling parameter multi-objective dynamic optimization method based on improved NSGA-III algorithm | |
| CN116822971B (en) | Well wall risk level prediction method | |
| CN110633868A (en) | Method for predicting properties of oil and gas in exploration well oil testing layer by optimizing neural network through genetic algorithm | |
| Moradi et al. | Drilling rate prediction using an innovative soft computing approach | |
| CN118013344A (en) | Tunneling parameter classification optimization method and system based on tunneling performance surrounding rock classification | |
| CN114862071B (en) | Method, device and equipment for predicting reaming torque of horizontal directional drilling and storage medium | |
| CN116644844A (en) | Stratum pressure prediction method based on neural network time sequence | |
| Mnati et al. | Prediction of penetration rate and cost with artificial neural network for alhafaya oil field | |
| Duan et al. | A ROP optimization approach based on improved BP neural network PSO | |
| CN114065603B (en) | Mechanical drilling speed prediction method and device | |
| Duan et al. | A ROP prediction approach based on improved BP 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |