[go: up one dir, main page]

CN112360341B - Machine learning-based tool face identification method for sliding directional drilling - Google Patents

Machine learning-based tool face identification method for sliding directional drilling Download PDF

Info

Publication number
CN112360341B
CN112360341B CN202011032164.4A CN202011032164A CN112360341B CN 112360341 B CN112360341 B CN 112360341B CN 202011032164 A CN202011032164 A CN 202011032164A CN 112360341 B CN112360341 B CN 112360341B
Authority
CN
China
Prior art keywords
tool face
face angle
tool
state
sequence
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.)
Active
Application number
CN202011032164.4A
Other languages
Chinese (zh)
Other versions
CN112360341A (en
Inventor
陶珉
郝家胜
张鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sichuan Wisdom Star Technology Co ltd
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202011032164.4A priority Critical patent/CN112360341B/en
Publication of CN112360341A publication Critical patent/CN112360341A/en
Application granted granted Critical
Publication of CN112360341B publication Critical patent/CN112360341B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B7/00—Special methods or apparatus for drilling
    • E21B7/04—Directional drilling
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B47/00—Survey of boreholes or wells
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B7/00—Special methods or apparatus for drilling
    • E21B7/04—Directional drilling
    • E21B7/10—Correction of deflected boreholes
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Geology (AREA)
  • Mining & Mineral Resources (AREA)
  • Physics & Mathematics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Fluid Mechanics (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geochemistry & Mineralogy (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Numerical Control (AREA)

Abstract

The invention discloses a machine learning-based sliding directional drilling tool face state identification method, which comprises the steps of collecting tool face angles and acquisition moments of sliding directional drilling operation, constructing a discrete state model of a tool face, normalizing the tool face angles to obtain a normalized tool face angle sequence, carrying out unique hot coding on marked states according to combinations of the tool face angles to obtain state codes corresponding to each tool face sequence, using the tool face angle sequence and the corresponding state codes as training data, and training a preset BP neural network to obtain a tool face state identification model; and in the sliding directional drilling operation process, obtaining a normalized tool face angle sequence at the current moment, inputting the normalized tool face angle sequence into a tool face state identification model to obtain a state code, and determining the state of the current equipment. The invention can accurately identify the states of tool face error data, stable state, change direction, change speed and the like, and realizes the real-time monitoring of the tool face state.

Description

Machine learning-based sliding directional drilling tool face state identification method
Technical Field
The invention belongs to the technical field of petroleum and gas drilling, and particularly relates to a machine learning-based sliding directional drilling tool face state identification method.
Background
Along with the exploitation of a large amount of oil and gas resources, the problems of low resource utilization rate, high exploitation difficulty and the like are more and more obvious. Particularly for the development of unconventional hydrocarbon reservoirs, it is often necessary to utilize sliding directional well techniques to drill to the desired horizon. FIG. 1 is a schematic diagram of a horizontal well drilling system. Currently, sliding directional drilling has been developed as the primary method of directional well trajectory control.
During sliding directional drilling, it is necessary to adjust the toolface (i.e., control the drilling toolface angle) at any time to maintain stable and dynamic control of the toolface to ensure that it can drill according to a predetermined trajectory. Fig. 2 is a schematic view of a toolface angle. As shown in fig. 2, the toolface angle is the angle of the deflecting toolface during sliding directional drilling and is divided into an inclinometer magnetometer face and an inclinometer face (unless otherwise specified, the toolface angle is usually referred to as the inclinometer face). The sliding directional drilling process is essentially an adjustment process for the toolface angle.
An important basis for adjusting the toolface control parameters is the current state of the toolface. The toolface conditions typically include erroneous data, steady state, rotational direction, rate of change, etc., and accurate identification of these conditions has a significant impact on the directional drilling process and benefits. However, due to the restriction of technologies such as sensors and data transmission, the real-time measurement and synchronous monitoring of the tool face angle cannot be achieved at present, and the state of the tool face is difficult to identify due to the delay or distortion in the information transmission process.
Currently, in an actual production process, in order to help an orientation engineer to make a tool face state judgment, a history tool face angle is recorded by a specific instrument panel (as shown in fig. 2 (c)). Engineers identify the current tool face state according to experience so as to guide the field to carry out directional operation, the individual factor difference is large, and the operation efficiency is different from person to person. Meanwhile, due to the influence of factors such as bit pressure, drilling tool combination, texture environment and the like, the change condition of the underground tool surface is complex, and data are discrete and nonlinear. The manual state identification is difficult and the accuracy is not high, so that the actual drilling track is easy to deviate from the preset drilling track.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a method for identifying the state of a sliding directional drilling tool face based on machine learning, and provides a discrete state model of the sliding directional drilling tool face.
In order to realize the purpose, the sliding directional drilling tool face state identification method based on the BP neural network comprises three stages: (1) establishing a tool surface discrete state model (S1-S3); (2) performing Onehot coding on various states to construct a state identification BP network (S4-S6); (3) performing online state identification (S7) by applying the BP network obtained by training, and specifically comprising the following steps:
s1: collecting tool face data of the sliding directional drilling operation, wherein the tool face data comprises a tool face angle and a collection time;
s2: extracting a plurality of tool face angle sequences from the tool face data, wherein each tool face angle sequence comprises tool face angles at continuous L acquisition moments;
s3: and carrying out state labeling on each tool face angle sequence, wherein the state labeling comprises error data, a stable state, a change direction and a change speed, and the specific method comprises the following steps:
judging error data of the tool face corner sequence, if no error data exists, marking the error data as 0, and if the error data exists, marking the error data as a serial number of the error data in the tool face corner sequence;
judging the stability of the tool face angle sequence, if the tool face angle sequence is unstable, marking the tool face angle sequence as 0, otherwise, marking the tool face angle sequence as 1;
judging the change direction of the tool face angle in the tool face angle sequence, if the change direction is anticlockwise change, marking the change direction as 0, and if the change direction is clockwise change, marking the change direction as 1;
dividing the change speed into M grades in advance, calculating the change speed of the tool face angle in the tool face angle sequence to obtain a corresponding grade, and marking the corresponding grade as the change speed grade of the tool face angle sequence;
s4: normalizing each tool face angle into data in a range of [0,1] to obtain a normalized tool face angle sequence;
s5: performing OneHot coding on the marked states according to the combination of the marked states to obtain state codes corresponding to each tool face sequence;
s6: constructing a BP neural network model, taking the normalized tool face angle sequence as input and the corresponding state code as expected output, and training the BP neural network model to obtain a tool face state identification network;
s7: in the process of the sliding directional drilling operation, tool face angles at the current time and the first L-1 times form a tool face angle sequence, normalization is carried out according to the method in the step S4, then the tool face angle sequence after normalization is input into the tool face state identification network trained in the step S6, and state codes are obtained, so that the error data, the stable state, the change direction and the change speed state of the tool face are determined.
Further, the judgment rule of the error data in step S3 is as follows:
sequentially judging each tool face angle in the tool face angle sequence, and acquiring the tool face angle with the acquisition time t
Figure BDA0002704078660000031
Calculating the face angle of the tool when determining the error data
Figure BDA0002704078660000032
The last correct tool face angle
Figure BDA0002704078660000033
Difference of (2)
Figure BDA0002704078660000034
If it is not
Figure BDA0002704078660000035
If the value is less than the preset threshold value, the face angle of the tool
Figure BDA0002704078660000036
If the data is correct, otherwise, respectively calculating the tool face angle at the next acquisition time t +1
Figure BDA0002704078660000037
And
Figure BDA0002704078660000038
if a difference is present between
Figure BDA0002704078660000039
Then judge the tool face angle with the collection time t
Figure BDA00027040786600000310
Recording the serial number of the error data, otherwise, the error data is correct data; and stopping when the tool face angle obtained by judgment is error data, otherwise, continuously judging the next tool face angle until the whole tool face angle sequence is judged.
Further, the stability determination rule in step S3 is as follows: and for the tool face angle sequence, respectively calculating the interval between adjacent tool faces and the interval between the tool face at the end of the sequence and the rest tool faces, if a certain interval is greater than a preset threshold value, judging that the tool face angle sequence is unstable, and if not, judging that the tool face angle sequence is stable.
Further, the normalization in step S5 is normalized using Min-Max normalization.
The invention relates to a method for identifying the state of a sliding directional drilling tool face based on machine learning, which comprises the steps of collecting tool face angles and acquisition moments of sliding directional drilling operation, constructing a discrete state model of the tool face, namely extracting a plurality of tool face angle sequences, carrying out state marking on each tool face angle sequence, wherein the state marking comprises error data, a stable state, a change direction and a change speed, then normalizing the tool face angles to obtain a normalized tool face angle sequence, carrying out unique hot coding on the marked states according to the combination of the marked states to obtain a state code corresponding to each tool face sequence, taking the normalized tool face angle sequence as input, taking the corresponding state code as expected output, and training a preset BP neural network to obtain a tool face state identification model; and in the sliding directional drilling operation process, obtaining a normalized tool face angle sequence at the current moment, inputting the normalized tool face angle sequence into a trained tool face state identification model to obtain a state code, and determining the state of the current equipment.
The invention adopts the BP neural network, can learn the state characteristics of the tool surface based on a large amount of historical data, avoids the limitation and difficulty of a tool surface physical model, is a convenient and effective tool surface state identification method, can identify the states of tool surface error data, stable state, change direction, change speed and the like, and can reduce the influence of human factors on the tool surface state judgment. The method is verified by the field operation process, has high accuracy and can meet the requirement of on-line automatic identification. The method has strong universality and popularization, and the established discrete state model of the sliding directional drilling tool face can be suitable for various mainstream MWD instruments at present.
Drawings
FIG. 1 is a schematic view of a horizontal well drilling system;
FIG. 2 is a schematic view of a toolface angle;
fig. 3 is a flowchart of an embodiment of a sliding directional drilling tool face state identification method based on a BP neural network according to the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the subject matter of the present invention.
Examples
FIG. 1 is a flow chart of an embodiment of a machine learning based method for identifying a tool face state of a sliding directional drilling tool according to the present invention. As shown in fig. 1, the method for identifying the tool face state of the sliding directional drilling based on machine learning according to the present invention can be divided into three stages, specifically as follows:
the first stage is as follows: establishing tool surface discrete state model
S301: tool face data collection:
tool face data for a sliding directional drilling operation is collected, including the tool face angle and the time of acquisition.
S302: tool face data serialization:
and extracting a plurality of tool face angle sequences from the tool face data, wherein each tool face angle sequence comprises tool face angles at continuous L acquisition moments. The specific value of L can be set according to actual needs, generally, the value range of L is 5-10, and L is 5 in this embodiment.
S303: tool surface data state labeling:
and carrying out state labeling on each tool face angle sequence, wherein the state labeling comprises error data, a stable state, a change direction and a change speed, and the specific method comprises the following steps:
and judging error data of the tool face corner sequence, marking the tool face corner sequence as 0 if no error data exists, and marking the tool face corner sequence as a serial number of the error data in the tool face corner sequence if the error data exists. Generally, only at most one error data needs to be considered, and the judgment rule of the error data in the embodiment is as follows:
sequentially judging each tool face angle in the tool face angle sequence, and acquiring the tool face angle with the acquisition time t
Figure BDA0002704078660000051
Calculating the face angle of the tool when determining the error data
Figure BDA0002704078660000052
The last correct tool face angle
Figure BDA00027040786600000510
Difference of (2)
Figure BDA0002704078660000053
If it is not
Figure BDA0002704078660000054
Less than a predetermined threshold, e.g. 50 deg., the toolface angle
Figure BDA0002704078660000055
If the data is correct, otherwise, respectively calculating the tool face angle at the next acquisition time t +1
Figure BDA0002704078660000056
And
Figure BDA0002704078660000057
if a difference is present between
Figure BDA0002704078660000058
Then judge the tool face angle with the collection time t
Figure BDA0002704078660000059
And recording the serial number of the error data, otherwise, recording the correct data. And stopping when the tool face angle obtained by judgment is error data, otherwise, continuously judging the next tool face angle until the whole tool face angle sequence is judged.
And (5) performing stability judgment on the tool face angle sequence, and marking the tool face angle sequence as 0 if the tool face angle sequence is unstable, otherwise, marking the tool face angle sequence as 1. The stability determination rule in this embodiment is as follows: and for the tool face angle sequence, respectively calculating the interval between adjacent tool faces and the interval between the tool face at the end of the sequence and the rest tool faces, if a certain interval is greater than a preset threshold value, judging that the tool face angle sequence is unstable, and if not, judging that the tool face angle sequence is stable.
And judging the change direction of the tool face angle in the tool face angle sequence, and marking the change direction as 0 if the change direction is anticlockwise change, and marking the change direction as 1 if the change direction is clockwise change.
Dividing the change speed into M grades in advance, calculating the change speed of the tool face angle in the tool face angle sequence to obtain a corresponding grade, and marking the corresponding grade as the change speed grade of the tool face angle sequence.
And a second stage: training a tool face state identification network:
s304: tool face angle normalization:
the data in the range of [0,1] is normalized for each toolface angle, resulting in a sequence of normalized toolface angles. In this embodiment, Min-Max normalization is used.
S305: and (3) state marking coding:
and performing OneHot (one-hot) coding on the marked states according to the combination of the marked states to obtain the state code corresponding to each tool face sequence. Since there are 4 kinds of marks, among which there are L +1 kinds of values for the error data mark, 2 kinds of values for the stable state mark, 2 kinds of values for the change direction mark, and M kinds of values for the change speed mark, there are (L +1) × 2 × 2 × M kinds in total for the mark combination, and the dimension of the state code is (L +1) × 2 × M.
S306: training a tool face state identification network:
and constructing a tool face state identification network, taking the normalized tool face angle sequence as input and the corresponding state code as expected output, and training the BP neural network to obtain the tool face state identification network. The BP neural network is a common deep learning model, and the specific principle thereof is not described herein. The BP neural network in the embodiment comprises an input layer, a hidden layer and an output layer, wherein the number of neurons in the input layer is L; the number of the neurons of the output layer is the dimension of one-hot coding, namely (L +1) multiplied by 2 multiplied by M, and the activation function is Softmax; the hidden layer is composed of a plurality of fully connected layers.
And a third stage: online identifying the tool face state:
s307: recognizing the state of the tool surface:
in the process of the sliding directional drilling operation, tool face angles at the current time and the first L-1 times form a tool face angle sequence, normalization is carried out according to the method in the step S304, then the tool face angle sequence after normalization is input into the tool face state identification network trained in the step S306, and state codes are obtained, so that the state of the tool face is determined.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined by the appended claims, and all matters of the invention which utilize the inventive concepts are protected.

Claims (2)

1.一种基于机器学习的滑动定向钻井工具面状态辨识方法，其特征在于，包括以下步骤：1. a sliding directional drilling tool face state identification method based on machine learning, is characterized in that, comprises the following steps: S1：收集滑动定向钻井作业的工具面数据，包括工具面角和采集时刻；S1: Collect tool face data of sliding directional drilling operations, including tool face angle and collection time; S2：从工具面数据中提取出若干工具面角序列，每个工具面角序列中包含连续L个采集时刻的工具面角；S2: Extract several tool-face angle sequences from the tool-face data, and each tool-face angle sequence includes tool-face angles at L consecutive acquisition moments; S3：对每个工具面角序列进行状态标注，包括错误数据、稳定状态、变化方向和变化速度，具体方法如下：S3: Label each tool face angle sequence, including error data, stable state, change direction and change speed, as follows: 对工具面角序列进行错误数据判断，如果不存在错误数据则标记为0，如果存在错误数据则标记为错误数据在工具面角序列中的序号；错误数据的判断规则如下：The tool face angle sequence is judged by the wrong data. If there is no wrong data, it will be marked as 0. If there is wrong data, it will be marked as the serial number of the wrong data in the tool face angle series. The judgment rules of wrong data are as follows: 依次对工具面角序列中各个工具面角进行判断，在对采集时刻为t的工具面角
Figure FDA0003272448470000011
进行错误数据判断时，计算工具面角
Figure FDA0003272448470000012
与之前最近一个正确工具面角
Figure FDA0003272448470000013
的差值
Figure FDA0003272448470000014
如果
Figure FDA0003272448470000015
小于预设阈值，则工具面角
Figure FDA0003272448470000016
为正确数据，否则分别计算后一采集时刻t+1工具面角
Figure FDA0003272448470000017
与
Figure FDA0003272448470000018
的差值，如果
Figure FDA0003272448470000019
则判断采集时刻为t的工具面角
Figure FDA00032724484700000110
为错误数据，记录其序号，否则为正确数据；当判断得到一个工具面角为错误数据时，则停止，否则继续判断下一个工具面角，直到整个工具面角序列判断完毕；The tool face angles in the tool face angle sequence are judged in turn, and the tool face angles at the collection time t
Figure FDA0003272448470000011
Calculate the tool face angle when judging wrong data
Figure FDA0003272448470000012
with the most recent correct tool face angle
Figure FDA0003272448470000013
difference
Figure FDA0003272448470000014
if
Figure FDA0003272448470000015
less than the preset threshold, the tool face angle
Figure FDA0003272448470000016
is the correct data, otherwise calculate the tool face angle of the next acquisition time t+1 respectively
Figure FDA0003272448470000017
and
Figure FDA0003272448470000018
difference, if
Figure FDA0003272448470000019
Then judge the tool face angle at the acquisition time t
Figure FDA00032724484700000110
If it is wrong data, record its serial number, otherwise it is correct data; when it is judged that a tool face angle is wrong data, stop, otherwise continue to judge the next tool face angle until the whole tool face angle sequence is judged; 对工具面角序列进行稳定性判断，如果不稳定则标记为0，否则标记为1；稳定性判断规则如下：对于工具面角序列，分别计算相邻工具面之间的间隔以及序列末尾工具面与其余工具面之间的间隔，如果存在某个间隔大于预设的阈值，则判断该工具面角序列为不稳定，否则为稳定；The stability judgment of the tool face angle sequence is carried out, if it is unstable, it is marked as 0, otherwise it is marked as 1; the stability judgment rule is as follows: For the tool face angle sequence, the interval between adjacent tool faces and the tool face at the end of the sequence are calculated respectively. The interval between the tool face and the rest of the tool face, if there is a certain interval greater than the preset threshold, the tool face angle sequence is judged to be unstable, otherwise it is stable; 对工具面角序列中工具面角的变化方向进行判断，如果是逆时针变化则标记为0，如果是顺时针变化则标记为1；Judge the change direction of the tool face angle in the tool face angle sequence, if it is a counterclockwise change, it is marked as 0, and if it is a clockwise change, it is marked as 1; 预先将变化速度划分为M个等级，计算工具面角序列中工具面角的变化速度得到对应等级，标记为该工具面角序列的变化速度等级；The change speed is divided into M grades in advance, and the change speed of the tool face angle in the tool face angle sequence is calculated to obtain the corresponding grade, which is marked as the change speed grade of the tool face angle sequence; S4：对每个工具面角归一化为[0,1]范围内的数据，得到归一化后的工具面角序列；S4: Normalize each tool face angle to data in the range of [0,1] to obtain the normalized tool face angle sequence; S5：对标记的状态按照其组合进行OneHot编码，得到每个工具面序列所对应的状态编码；S5: OneHot encoding is performed on the marked states according to their combinations, and the state encoding corresponding to each tool face sequence is obtained; S6：构建BP神经网络模型，将归一化后的工具面角序列作为输入，其对应的状态编码作为期望输出，对该BP神经网络模型进行训练，得到工具面状态辨识网络；S6: Build a BP neural network model, take the normalized tool face angle sequence as input, and the corresponding state code as the expected output, train the BP neural network model, and obtain a tool face state identification network; S7：在滑动定向钻井作业过程中，将当前时刻和前L-1个时刻的工具面角构成工具面角序列，按照步骤S4中的方法进行归一化，然后将归一化后的工具面角序列输入步骤S6训练好的工具面辨识网络中，得到状态编码，从而确定工具面的错误数据、稳定状态、变化方向和变化速度状态。S7: During the sliding directional drilling operation, the tool face angle sequence at the current moment and the previous L-1 moment is formed into a tool face angle sequence, normalized according to the method in step S4, and then the normalized tool face The angle sequence is input into the tool face identification network trained in step S6, and the state code is obtained, so as to determine the error data, stable state, change direction and change speed state of the tool face. 2.根据权利要求1所述的滑动定向钻井工具面状态辨识方法，其特征在于，所述步骤S5中归一化采用Min-Max标准化进行归一化。2 . The method for identifying the state of a sliding directional drilling tool face according to claim 1 , wherein in the step S5 , the normalization is performed by Min-Max normalization. 3 .
CN202011032164.4A 2020-09-27 2020-09-27 Machine learning-based tool face identification method for sliding directional drilling Active CN112360341B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011032164.4A CN112360341B (en) 2020-09-27 2020-09-27 Machine learning-based tool face identification method for sliding directional drilling

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011032164.4A CN112360341B (en) 2020-09-27 2020-09-27 Machine learning-based tool face identification method for sliding directional drilling

Publications (2)

Publication Number Publication Date
CN112360341A CN112360341A (en) 2021-02-12
CN112360341B true CN112360341B (en) 2021-12-24

Family

ID=74507025

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011032164.4A Active CN112360341B (en) 2020-09-27 2020-09-27 Machine learning-based tool face identification method for sliding directional drilling

Country Status (1)

Country Link
CN (1) CN112360341B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113464120B (en) * 2021-09-06 2021-12-03 中国石油集团川庆钻探工程有限公司 Tool face state prediction method and system, and sliding directional drilling method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109962909A (en) * 2019-01-30 2019-07-02 大连理工大学 Network intrusion anomaly detection method based on machine learning
CN110263972A (en) * 2019-05-15 2019-09-20 北京航空航天大学 Product quality prediction technique and device
CN110500034A (en) * 2019-08-30 2019-11-26 中国石油集团川庆钻探工程有限公司 Method for establishing neural network model, determining torsional pendulum drill string parameters and directionally drilling
CN111401399A (en) * 2019-12-24 2020-07-10 中国国家铁路集团有限公司 A method and device for early warning and classification of railway freight accidents
CN111462485A (en) * 2020-03-31 2020-07-28 电子科技大学 Traffic intersection congestion prediction method based on machine learning

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102010005264A1 (en) * 2010-01-20 2011-07-21 Smolka, Peter P., Dr., 48161 Chiselless drilling system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109962909A (en) * 2019-01-30 2019-07-02 大连理工大学 Network intrusion anomaly detection method based on machine learning
CN110263972A (en) * 2019-05-15 2019-09-20 北京航空航天大学 Product quality prediction technique and device
CN110500034A (en) * 2019-08-30 2019-11-26 中国石油集团川庆钻探工程有限公司 Method for establishing neural network model, determining torsional pendulum drill string parameters and directionally drilling
CN111401399A (en) * 2019-12-24 2020-07-10 中国国家铁路集团有限公司 A method and device for early warning and classification of railway freight accidents
CN111462485A (en) * 2020-03-31 2020-07-28 电子科技大学 Traffic intersection congestion prediction method based on machine learning

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
滑动定向钻探过程的智能决策算法研究与实现;张鑫;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20200715(第07期);B019-213 *

Also Published As

Publication number Publication date
CN112360341A (en) 2021-02-12

Similar Documents

Publication Publication Date Title
CN104239489B (en) A Method of Using Similarity Search and Improving BP Neural Network to Predict Water Level
CN112664174B (en) Tunnel surrounding rock grade determination method and system based on multiple drill holes
CN102944216B (en) Three-redundant ship dynamic positioning heading measurement method based on improved voting algorithm
CN104121014A (en) Method for diagnosing type of leakage of drilled well based on neural network fusion technique
CN108301823A (en) A method of identification reservoir hydrocarbons dessert
CN103114846B (en) A kind for the treatment of system afterwards of the deviational survey data based on optic fiber gyroscope inclinometer
CN114358434A (en) Prediction method of drilling machine ROP based on LSTM recurrent neural network model
CN112360341B (en) Machine learning-based tool face identification method for sliding directional drilling
CN104832299B (en) A kind of high oil consumption driving condition decision method, equipment and system
CN106382102A (en) Overflow early warning method based on clustering algorithm
CN119102578A (en) A downhole geosteering closed-loop control system and control method
CN106761685A (en) The method that well track pattern is recognized using tool face azimuth
CN111257588B (en) ORB and RANSAC-based oil phase flow velocity measurement method
CN111860863B (en) Active learning method based on covariance matrix
CN114086887A (en) Downhole planning method for track of borehole to be drilled based on artificial intelligence
CN118690271A (en) An intelligent identification method for early warning signals of pipe sticking accidents under long-tail distribution
CN117172113A (en) A rotary steerable drilling wellbore trajectory prediction method, system, equipment and medium
CN115434696B (en) Lithology recognition database construction method and lithology recognition method
US20210088396A1 (en) Method for assessing the quality of the coupling of two tubular components
CN114722717A (en) Method and device for obtaining drilling fluid flow rate
CN109653730A (en) One kind being used for drill-stem test operation downhole perforation well section depth calibration method
CN102953722B (en) Anti-collision obstacle bypassing method and system during drilling operation
CN114925247A (en) Lost circulation type identification method and device and related equipment
CN113464120B (en) Tool face state prediction method and system, and sliding directional drilling method and system
CN117077462B (en) Deep oil and gas precise navigation while drilling gamma logging curve tuning method and system

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
TR01 Transfer of patent right

Effective date of registration: 20220929

Address after: 646000 No. 19, 5 section of wine Valley Road, Luzhou high tech Zone, Sichuan

Patentee after: Sichuan Wisdom Star Technology Co.,Ltd.

Address before: 611731, No. 2006, West Avenue, Chengdu hi tech Zone (West District, Sichuan)

Patentee before: University of Electronic Science and Technology of China

TR01 Transfer of patent right