CN116975728A - Safety management method and system for coal bed methane drilling engineering - Google Patents
Safety management method and system for coal bed methane drilling engineering Download PDFInfo
- Publication number
- CN116975728A CN116975728A CN202310981644.2A CN202310981644A CN116975728A CN 116975728 A CN116975728 A CN 116975728A CN 202310981644 A CN202310981644 A CN 202310981644A CN 116975728 A CN116975728 A CN 116975728A
- Authority
- CN
- China
- Prior art keywords
- vector
- pressure
- feature
- coalbed methane
- methane concentration
- 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
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 454
- 238000005553 drilling Methods 0.000 title claims abstract description 102
- 238000007726 management method Methods 0.000 title claims abstract description 53
- 239000003245 coal Substances 0.000 title claims abstract description 51
- 239000013598 vector Substances 0.000 claims description 335
- 239000011159 matrix material Substances 0.000 claims description 92
- 238000000605 extraction Methods 0.000 claims description 43
- 238000000034 method Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 14
- 238000000354 decomposition reaction Methods 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 7
- 230000008859 change Effects 0.000 description 14
- 238000004590 computer program Methods 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 235000019580 granularity Nutrition 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
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
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
- E21B49/08—Obtaining fluid samples or testing fluids, in 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
- E21B47/00—Survey of boreholes or wells
- E21B47/06—Measuring temperature or pressure
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Geology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mining & Mineral Resources (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Geochemistry & Mineralogy (AREA)
- Fluid Mechanics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Geophysics (AREA)
- Drilling And Exploitation, And Mining Machines And Methods (AREA)
- Earth Drilling (AREA)
Abstract
The application relates to the field of intelligent detection, and particularly discloses a safety management method and a safety management system for coal bed methane drilling engineering. Therefore, the safety condition in the well drilling can be intuitively known, the safety problem can be found in time, measures can be taken, and the safety accident can be prevented.
Description
Technical Field
The application relates to the field of intelligent detection, in particular to a safety management method and system for coal bed methane drilling engineering.
Background
Safety management of coalbed methane drilling engineering is a key step in ensuring personnel and equipment safety during drilling operations. The safety condition in the drilling operation process is required to be monitored and recorded, the abnormal condition can be found timely through the monitoring and the recording, and corresponding measures are adopted to adjust and process, so that the safety in the drilling operation process is ensured.
The drilling site can be regularly subjected to safety inspection and checking, so that potential safety hazards can be found in time, but the safety of the drilling site is completely insufficient only by means of the evaluation.
Thus, there is a need for an optimized safety management scheme for coal bed methane drilling projects.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a safety management method and a system thereof for coal bed methane drilling engineering, which adopt an artificial intelligent detection technology based on deep learning, and judge whether the inside of a well is in a safety condition or not by extracting and fusing characteristics of a coal bed methane concentration value and a well drilling internal pressure value. Therefore, the safety condition in the well drilling can be intuitively known, the safety problem can be found in time, measures can be taken, and the safety accident can be prevented.
According to one aspect of the present application, there is provided a safety management method for a coalbed methane drilling project, comprising:
acquiring coalbed methane concentration values at a plurality of preset time points in a preset time period and pressure values in a well;
arranging the coalbed methane concentration values and the drilling pressure values at a plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension respectively;
The coalbed methane concentration input vector is passed through a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector;
inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector;
fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector; and
the classification feature vector is passed through a classifier to obtain a classification result that indicates whether a safe condition is present within the borehole.
In the safety management method for the coal bed methane drilling engineering, the coal bed methane concentration input vector is transmitted through a system comprising a full-connection layer and a one-dimensional convolution layerA time series encoder to obtain a coalbed methane concentration profile, comprising: performing full-connection coding on the coalbed methane concentration input vector by using a full-connection layer of the time sequence coder according to the following full-connection coding formula so as to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector; wherein, the full-connection coding formula is: wherein />Is the coalbed methane concentration input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++ >Representing a matrix multiplication; and performing one-dimensional convolution coding on the coalbed methane concentration input vector by using a one-dimensional convolution layer of the time sequence coder according to the following one-dimensional convolution coding formula so as to extract high-dimensional implicit correlation features among feature values of all positions in the input vector; wherein, the one-dimensional convolution coding formula is: /> wherein ,ais convolution kernel inxWidth in direction, ++>For the convolution kernel parameter vector, +.>For a local vector matrix that operates with a convolution kernel,wfor the size of the convolution kernel +.>Representing the coalbed methane concentration input vector, +.>Representing one-dimensional convolutional encoding of the coalbed methane concentration input vector.
In the above method for safety management of coal bed methane drilling engineering, the inputting the pressure input vector into the multi-scale domain feature extraction module to obtain a pressure feature vector includes: inputting the pressure input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first-scale pressure feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel of a first length; and concatenating the first scale pressure feature vector and the second scale pressure feature vector to obtain the pressure feature vector.
In the above safety management method for coal bed methane drilling engineering, the fusing the coal bed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector includes: constructing a global similarity matrix between the coalbed methane concentration feature vector and the pressure feature vector, wherein the feature value of each position on the non-diagonal position in the global similarity matrix is the variance between the feature values of the coalbed methane concentration feature vector and the corresponding two positions in the pressure feature vector; constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector; the global similarity matrix and the covariance matrix are aggregated along the channel dimension and then pass through an inter-manifold multi-granularity association feature extractor based on a convolution layer to obtain an inter-manifold global association multi-granularity feature matrix; performing matrix decomposition based on characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold related characteristic values; and extracting a predetermined number of manifold-related feature values from the plurality of manifold-related feature values, and arranging the predetermined number of manifold-related feature values to obtain the classification feature vector.
In the above safety management method for coal bed methane drilling engineering, the step of passing the classification feature vector through a classifier to obtain a classification result, where the classification result indicates whether the drilling is in a safe condition, includes: processing the classification feature vector using the classifier in a classification formula to generate the classification result; wherein, the classification formula is: wherein />For the classification result, < >>The classification feature vector is represented as such,weight matrix for full connection layer, +.>Representing the deflection vector of the fully connected layer, +.>Is a normalized exponential function.
According to another aspect of the present application, there is provided a safety management system for a coalbed methane drilling project, comprising:
the data acquisition module is used for acquiring coalbed methane concentration values and drilling pressure values at a plurality of preset time points in a preset time period;
the data structuring module is used for respectively arranging the coalbed methane concentration values and the drilling pressure values at a plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension;
the coalbed methane concentration characteristic extraction module is used for enabling the coalbed methane concentration input vector to pass through a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector;
The drilling pressure characteristic extraction module is used for inputting the pressure input vector into the multi-scale domain characteristic extraction module to obtain a pressure characteristic vector;
the fusion module is used for fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector; and
and the safety condition judging module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result represents whether the drilling well is in a safety condition or not.
Compared with the prior art, the safety management method and the system for the coal bed methane drilling engineering provided by the application adopt an artificial intelligent detection technology based on deep learning, and judge whether the drilling is in a safety condition or not by extracting and fusing the characteristics of the coal bed methane concentration value and the drilling pressure value. Therefore, the safety condition in the well drilling can be intuitively known, the safety problem can be found in time, measures can be taken, and the safety accident can be prevented.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a flow chart of a method for safety management for coal bed methane drilling projects according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a security management method for coal bed methane drilling engineering according to an embodiment of the present application.
Fig. 3 is a flowchart of inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector in a safety management method for coal bed methane drilling engineering according to an embodiment of the present application.
Fig. 4 is a flowchart of a method for safety management of a coalbed methane drilling project according to an embodiment of the present application, in which the coalbed methane concentration feature vector and the pressure feature vector are fused to obtain a classification feature vector.
FIG. 5 is a system block diagram of a security management system for coal bed methane drilling projects in accordance with an embodiment of the present application.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the application.
Description of the embodiments
Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Summary of the application
As noted above in the background section, ensuring personnel and equipment safety during drilling operations is an important task in coalbed methane drilling projects. The effective safety management can be implemented to monitor and record the safety condition in the drilling operation, discover abnormality in time and take corresponding measures to adjust and manage so as to ensure the safety of the drilling operation. At the drilling site, regular safety inspection and examination is necessary, and potential safety hazards can be found in time. However, merely relying on this approach to evaluate the safety of the drilling site is inadequate, and thus, an optimized safety management scheme for coal bed methane drilling projects is desired.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
Deep learning and safety management for coalbed methane drilling engineering provide new solutions and solutions.
Specifically, in the technical scheme of the application, firstly, coal bed gas concentration values and drilling pressure values at a plurality of preset time points in a preset time period are obtained. It should be appreciated that the concentration of coalbed methane and the pressure in the well bore are important indicators of whether the well bore operation is safe or not, and that changes can reflect potentially dangerous and dangerous situations that may exist. By collecting the coalbed methane concentration data at a plurality of preset time points, the time sequence change and fluctuation condition of the coalbed methane concentration data can be observed, and whether the coalbed methane concentration exceeds an abnormal or safe range or not can be judged. High concentration coalbed methane may cause dangerous events such as explosion, fire and the like, so that it is necessary to timely acquire the change of the coalbed methane concentration and take corresponding measures. In addition, the pressure in the well is one of key parameters for measuring the safety of the well drilling operation, and too high or too low pressure in the well drilling can cause serious accidents such as well wall collapse, blowout and the like. Thus, the value of the pressure in the well is obtained and monitored in time, and the stability and safety of the well operation can be evaluated so as to take necessary safety measures.
And then, arranging the coalbed methane concentration values and the pressure values in the well at the plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension respectively. By arranging the data according to the time dimension, the time sequence relation of the data can be reserved, and the change condition of the coal bed gas concentration and the drilling pressure at different time points can be reflected. This facilitates capturing trends, periodic or sudden events over a period of time, helping and predicting potential safety hazards and anomalies. Meanwhile, the concentration value of the coal bed gas and the pressure value in the well are combined into a vector form, so that data processing and analysis can be better carried out.
And then, the coalbed methane concentration input vector is passed through a time sequence encoder comprising a fully-connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector. It should be appreciated that the one-dimensional convolution layer is capable of performing sliding window convolution operations on the time dimension of the coalbed methane concentration data, so as to effectively capture local time sequence patterns and features of the coalbed methane concentration; the full-connection layer is used for further processing and mapping the coalbed methane concentration feature vector, and can convert the multidimensional feature vector output by the one-dimensional convolution layer into a higher-level abstract representation, so that the coalbed methane concentration feature vector with more characterization capability is extracted by considering coalbed methane concentration features at different time points. The input vector of the concentration of the coal bed gas can be processed layer by layer and extracted by a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer, and finally a compact characteristic vector of the concentration of the coal bed gas is obtained.
And simultaneously, inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector. In particular, the pressure input vector contains pressure values at a plurality of predetermined points in time, which may have different laws of variation and periodicity. By applying the multi-scale domain feature extraction module, pressure change features on different time scales, such as rapid fluctuation on a short time scale and trend change on a long time scale, can be captured. By extracting the multi-scale features, dynamic change conditions of pressure can be more fully described, and potential safety risks and abnormal conditions can be revealed.
And further, fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector. It should be understood that different information sources can be combined by fusing the characteristic vector of the concentration of the coalbed methane and the characteristic vector of the pressure, so that characteristic information with more representation and comprehensiveness is obtained, and the relevance and the change trend between the concentration of the coalbed methane and the pressure can be better reflected by fusing the characteristic vectors. Finally, the classification feature vector is passed through a classifier to obtain a representation of whether a safe condition is in the well. Therefore, engineers can intuitively know the safety condition of drilling, and can timely take necessary safety measures to prevent accidents and ensure the safe operation of drilling engineering.
In particular, in the technical scheme of the application, if cascade connection or weighting is directly adopted to fuse the coalbed methane concentration characteristic vector and the pressure characteristic vector, the correlation between the coalbed methane concentration characteristic vector and the pressure characteristic vector is ignored, and the fused characteristic vector cannot fully represent the safety condition in the well drilling. At the same time, because coalbed methane concentration characteristics and pressure characteristics tend to have different manifold structures, i.e., their distribution patterns in the feature space may be different. If cascading and weighted fusion are directly used, the situation that main information is lost and a large amount of redundancy exists can occur, which is unfavorable for final classification.
Based on this, further, fusing the coalbed methane concentration feature vector and the pressure feature vector based on a dimensional perspective correlation of feature manifolds between the coalbed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector, specifically, comprising: constructing a global similarity matrix between the coalbed methane concentration feature vector and the pressure feature vector, wherein the feature value of each position on the non-diagonal position in the global similarity matrix is the variance between the feature values of the coalbed methane concentration feature vector and the corresponding two positions in the pressure feature vector; constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector; the global similarity matrix and the covariance matrix are aggregated along the channel dimension and then pass through an inter-manifold multi-granularity association feature extractor based on a convolution layer to obtain an inter-manifold global association multi-granularity feature matrix; performing matrix decomposition based on characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold related characteristic values; and extracting a predetermined number of manifold-related feature values from the plurality of manifold-related feature values, and arranging the predetermined number of manifold-related feature values to obtain the classification feature vector.
In the technical scheme of the application, the global similarity matrix between the coalbed methane concentration characteristic vector and the pressure characteristic vector is constructed and used for representing position manifold similarity correlation of the characteristic manifold of the coalbed methane concentration characteristic vector and the characteristic manifold of the pressure characteristic vector, namely manifold correlation information of element granularity between the coalbed methane concentration characteristic vector and the characteristic manifold of the pressure characteristic vector. In order to represent the characteristic manifold correlation of the vector granularity between the coalbed methane concentration characteristic vector and the pressure characteristic vector, in the technical scheme of the application, a covariance matrix between the coalbed methane concentration characteristic vector and the pressure characteristic vector is further calculated. Further, the local correlation pattern features implicit in the global similarity matrix and the covariance matrix are captured through an inter-manifold correlation feature extractor based on a convolution layer, wherein the local correlation pattern features are used for representing implicit feature representations of feature manifold correlations with different granularities of coal bed gas concentration feature vectors and pressure feature vectors. After the manifold global-association feature matrix is obtained, performing matrix decomposition on the manifold global-association feature matrix based on feature values to capture feature main dimensions from the manifold global-association feature matrix and reject unnecessary feature dimensions to obtain the classification feature vector.
In this way, the coalbed methane concentration feature vector and the pressure feature vector are fused based on the dimensional view angle correlation of the feature manifold between the coalbed methane concentration feature vector and the pressure feature vector, and the intrinsic structure and distribution of the data of the coalbed methane concentration feature vector and the pressure feature vector in the high-dimensional feature space can be utilized to capture the characteristics and rules of the data distribution between the coalbed methane concentration feature vector and the pressure feature vector from the main direction of the feature manifold of the coalbed methane concentration feature vector and the pressure feature vector. Meanwhile, the data in the coalbed methane concentration characteristic vector and the pressure characteristic vector can be projected on the most important main dimension based on the characteristic decomposition of the characteristic value of the matrix, so that redundant information and noise information are removed, the most essential information is reserved, the quality and the efficiency of the data can be improved, and the separability and the distinguishability of the data are enhanced.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary method
Fig. 1 is a flow chart of a method for safety management for coal bed methane drilling projects according to an embodiment of the present application. As shown in fig. 1, a safety management method for coal bed methane drilling engineering according to an embodiment of the present application includes: s110, acquiring coalbed methane concentration values at a plurality of preset time points in a preset time period and pressure values in a well; s120, arranging the coalbed methane concentration values and the pressure values in the well at a plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension respectively; s130, passing the coalbed methane concentration input vector through a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector; s140, inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector; s150, fusing the coalbed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector; and S160, passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result indicates whether the drilling is in a safe condition.
Fig. 2 is a schematic diagram of a security management method for coal bed methane drilling engineering according to an embodiment of the present application. In this architecture, as shown in fig. 2, first, coalbed methane concentration values and intra-well pressure values are acquired at a plurality of predetermined time points over a predetermined period of time. And then, arranging the coalbed methane concentration values and the pressure values in the well at the plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension respectively. And then, the coalbed methane concentration input vector is passed through a time sequence encoder comprising a fully-connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector. And simultaneously, inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector. And further, fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector. Finally, the classification feature vector is passed through a classifier to obtain a classification result, the classification result being indicative of whether a safe condition is present within the borehole.
In step S110, coalbed methane concentration values and intra-well pressure values are obtained at a plurality of predetermined time points over a predetermined period of time. It should be appreciated that the concentration of coalbed methane and the pressure in the well bore are important indicators of whether the well bore operation is safe or not, and that changes can reflect potentially dangerous and dangerous situations that may exist. By collecting the coalbed methane concentration data at a plurality of preset time points, the time sequence change and fluctuation condition of the coalbed methane concentration data can be observed, and whether the coalbed methane concentration exceeds an abnormal or safe range or not can be judged. High concentration coalbed methane may cause dangerous events such as explosion, fire and the like, so that it is necessary to timely acquire the change of the coalbed methane concentration and take corresponding measures. In addition, the pressure in the well is one of key parameters for measuring the safety of the well drilling operation, and too high or too low pressure in the well drilling can cause serious accidents such as well wall collapse, blowout and the like. Thus, the value of the pressure in the well is obtained and monitored in time, and the stability and safety of the well operation can be evaluated so as to take necessary safety measures. The values of coalbed methane concentration at a plurality of predetermined points in time over the predetermined time period described herein may be obtained by collecting data from the gas sensor and the values of intra-well pressure at the plurality of predetermined points in time may be obtained by collecting data from the pressure sensor.
In step S120, the coalbed methane concentration values and the pressure values in the well at the plurality of predetermined time points are respectively arranged into a coalbed methane concentration input vector and a pressure input vector according to a time dimension. It should be appreciated that by arranging the data in a time dimension, the sequential relationship of the data in time can be preserved, and the change condition of the concentration of the coalbed methane and the drilling pressure at different time points can be reflected, which is beneficial to capturing trends, periodicity or emergency events in a time period, and helping and predicting potential safety hazards and abnormal conditions possibly existing. Meanwhile, the concentration value of the coal bed gas and the pressure value in the well are combined into a vector form, so that data processing and analysis can be better carried out.
In step S130, the coalbed methane concentration input vector is passed through a time series encoder comprising a fully connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration feature vector. It should be appreciated that the one-dimensional convolution layer is capable of performing sliding window convolution operations on the time dimension of the coalbed methane concentration data, so as to effectively capture local time sequence patterns and features of the coalbed methane concentration; the full-connection layer is used for further processing and mapping the coalbed methane concentration feature vector, and can convert the multidimensional feature vector output by the one-dimensional convolution layer into a higher-level abstract representation, so that the coalbed methane concentration feature vector with more characterization capability is extracted by considering coalbed methane concentration features at different time points. The input vector of the concentration of the coal bed gas can be processed layer by layer and extracted by a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer, and finally a compact characteristic vector of the concentration of the coal bed gas is obtained.
Specifically, the step of passing the coalbed methane concentration input vector through a time sequence encoder comprising a fully-connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector comprises the following steps: performing full-connection coding on the coalbed methane concentration input vector by using a full-connection layer of the time sequence coder according to the following full-connection coding formula so as to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector; wherein, the full-connection coding formula is: wherein />Is the coalbed methane concentration input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing a matrix multiplication; and performing one-dimensional convolution coding on the coalbed methane concentration input vector by using a one-dimensional convolution layer of the time sequence coder according to the following one-dimensional convolution coding formula so as to extract high-dimensional implicit correlation features among feature values of all positions in the input vector; wherein, the one-dimensional convolution coding formula is: /> wherein ,ais convolution kernel inxWidth in direction, ++>For the convolution kernel parameter vector, +.>To run with convolution kernelThe matrix of the calculated local vectors,wfor the size of the convolution kernel +.>Representing the coalbed methane concentration input vector, +.>Representing one-dimensional convolutional encoding of the coalbed methane concentration input vector.
In step S140, the pressure input vector is input to a multi-scale domain feature extraction module to obtain a pressure feature vector. The multi-scale domain feature extraction module carries out multi-scale convolution operation on input data by setting one-dimensional convolution kernels with different sizes, so that features under different scales are captured. In particular, the pressure input vector contains pressure values at a plurality of predetermined points in time, which may have different laws of variation and periodicity. By applying the multi-scale domain feature extraction module, pressure change features on different time scales, such as rapid fluctuation on a short time scale and trend change on a long time scale, can be captured. By extracting the multi-scale features, dynamic change conditions of pressure can be more fully described, and potential safety risks and abnormal conditions can be revealed.
Fig. 3 is a flowchart of inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector in a safety management method for coal bed methane drilling engineering according to an embodiment of the present application. As shown in fig. 3, the inputting the pressure input vector into the multi-scale domain feature extraction module to obtain a pressure feature vector includes: s141, inputting the pressure input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first-scale pressure feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length; and S143, cascading the first scale pressure characteristic vector and the second scale pressure characteristic vector to obtain the pressure characteristic vector.
In step S150, the coalbed methane concentration feature vector and the pressure feature vector are fused to obtain a classification feature vector. It should be understood that different information sources can be combined by fusing the characteristic vector of the concentration of the coalbed methane and the characteristic vector of the pressure, so that characteristic information with more representation and comprehensiveness is obtained, and the relevance and the change trend between the concentration of the coalbed methane and the pressure can be better reflected by fusing the characteristic vectors.
In particular, in the technical scheme of the application, if cascade connection or weighting is directly adopted to fuse the coalbed methane concentration characteristic vector and the pressure characteristic vector, the correlation between the coalbed methane concentration characteristic vector and the pressure characteristic vector is ignored, and the fused characteristic vector cannot fully represent the safety condition in the well drilling. At the same time, because coalbed methane concentration characteristics and pressure characteristics tend to have different manifold structures, i.e., their distribution patterns in the feature space may be different. If cascading and weighted fusion are directly used, the situation that main information is lost and a large amount of redundancy exists can occur, which is unfavorable for final classification. Thus, there is a need to fuse the coalbed methane concentration feature vector and the pressure feature vector based on the dimensional perspective correlation of the feature manifold between the coalbed methane concentration feature vector and the pressure feature vector.
Fig. 4 is a flowchart of a method for safety management of a coalbed methane drilling project according to an embodiment of the present application, in which the coalbed methane concentration feature vector and the pressure feature vector are fused to obtain a classification feature vector. As shown in fig. 4, the fusing the coalbed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector includes: s151, constructing a global similarity matrix between the coalbed methane concentration feature vector and the pressure feature vector, wherein the feature value of each position on the non-diagonal position in the global similarity matrix is the variance between the feature values of the coalbed methane concentration feature vector and the corresponding two positions in the pressure feature vector; s152, constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector; s153, aggregating the global similarity matrix and the covariance matrix along a channel dimension, and then obtaining a manifold-to-manifold global association multi-granularity feature matrix through a manifold multi-granularity association feature extractor based on a convolution layer; s154, performing matrix decomposition based on characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold association characteristic values; and S155, extracting a predetermined number of manifold related feature values from the manifold related feature values, and arranging the manifold related feature values to obtain the classification feature vector.
In the technical scheme of the application, the global similarity matrix between the coalbed methane concentration characteristic vector and the pressure characteristic vector is constructed and used for representing position manifold similarity correlation of the characteristic manifold of the coalbed methane concentration characteristic vector and the characteristic manifold of the pressure characteristic vector, namely manifold correlation information of element granularity between the coalbed methane concentration characteristic vector and the characteristic manifold of the pressure characteristic vector. In order to represent the characteristic manifold correlation of the vector granularity between the coalbed methane concentration characteristic vector and the pressure characteristic vector, in the technical scheme of the application, a covariance matrix between the coalbed methane concentration characteristic vector and the pressure characteristic vector is further calculated. Further, the local correlation pattern features implicit in the global similarity matrix and the covariance matrix are captured through an inter-manifold correlation feature extractor based on a convolution layer, wherein the local correlation pattern features are used for representing implicit feature representations of feature manifold correlations with different granularities of coal bed gas concentration feature vectors and pressure feature vectors. After the manifold global-association feature matrix is obtained, performing matrix decomposition on the manifold global-association feature matrix based on feature values to capture feature main dimensions from the manifold global-association feature matrix and reject unnecessary feature dimensions to obtain the classification feature vector.
In this way, the coalbed methane concentration feature vector and the pressure feature vector are fused based on the dimensional view angle correlation of the feature manifold between the coalbed methane concentration feature vector and the pressure feature vector, and the intrinsic structure and distribution of the data of the coalbed methane concentration feature vector and the pressure feature vector in the high-dimensional feature space can be utilized to capture the characteristics and rules of the data distribution between the coalbed methane concentration feature vector and the pressure feature vector from the main direction of the feature manifold of the coalbed methane concentration feature vector and the pressure feature vector. Meanwhile, the data in the coalbed methane concentration characteristic vector and the pressure characteristic vector can be projected on the most important main dimension based on the characteristic decomposition of the characteristic value of the matrix, so that redundant information and noise information are removed, the most essential information is reserved, the quality and the efficiency of the data can be improved, and the separability and the distinguishability of the data are enhanced.
In step S160, the classification feature vector is passed through a classifier to obtain a classification result that indicates whether a safe condition is present in the borehole. A classifier is a machine learning model that can learn the mapping from input features to output classes. The classification result has two categories, namely safe and unsafe, so that the safety condition in the drilling well can be more intuitively known, and corresponding measures are taken for processing, thereby ensuring the safety of drilling operation.
Specifically, the step of passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result indicates whether the drilling is in a safe condition or not, and the step of including: processing the classification feature vector using the classifier in a classification formula to generate the classification result; wherein, the classification formula is: wherein />For the classification result, < >>Representing the classification feature vector,/->Weight matrix for full connection layer, +.>Representing the deflection vector of the fully connected layer, +.>Is a normalized exponential function.
In summary, the safety management method for coal bed methane drilling engineering according to the embodiment of the application is explained, which adopts an artificial intelligence detection technology based on deep learning, and judges whether the drilling is in a safety condition or not by extracting and fusing characteristics of the coal bed methane concentration value and the drilling pressure value. Therefore, the safety condition in the well drilling can be intuitively known, the safety problem can be found in time, measures can be taken, and the safety accident can be prevented.
Exemplary System
FIG. 5 is a system block diagram of a security management system for coal bed methane drilling projects in accordance with an embodiment of the present application. As shown in fig. 5, a safety management system 100 for coal bed methane drilling engineering according to an embodiment of the present application includes: a data acquisition module 110 for acquiring coalbed methane concentration values and intra-well pressure values at a plurality of predetermined time points within a predetermined time period; the data structuring module 120 is configured to arrange the coalbed methane concentration values and the pressure values in the well at the plurality of predetermined time points into a coalbed methane concentration input vector and a pressure input vector according to a time dimension, respectively; the coalbed methane concentration feature extraction module 130 is configured to pass the coalbed methane concentration input vector through a time sequence encoder including a fully-connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration feature vector; a drilling pressure feature extraction module 140 for inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector; a fusion module 150, configured to fuse the coalbed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector; and a safety condition judgment module 160 for passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result represents whether the drilling is in a safe condition.
In one example, in the safety management system 100 described above for coal bed methane drilling engineering, the coal isThe layer gas concentration feature extraction module 130 is configured to: performing full-connection coding on the coalbed methane concentration input vector by using a full-connection layer of the time sequence coder according to the following full-connection coding formula so as to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector; wherein, the full-connection coding formula is: wherein />Is the coalbed methane concentration input vector, +.>Is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing a matrix multiplication; and performing one-dimensional convolution coding on the coalbed methane concentration input vector by using a one-dimensional convolution layer of the time sequence coder according to the following one-dimensional convolution coding formula so as to extract high-dimensional implicit correlation features among feature values of all positions in the input vector; wherein, the one-dimensional convolution coding formula is: /> wherein ,ais convolution kernel inxWidth in direction, ++>For the convolution kernel parameter vector, +.>For a local vector matrix that operates with a convolution kernel,wfor the size of the convolution kernel +.>Representing the coalbed methane concentration input vector, +.>Representing one-dimensional convolutional encoding of the coalbed methane concentration input vector.
In one example, in the safety management system 100 for coal bed methane drilling engineering described above, the drilling pressure feature extraction module 140 includes: the device comprises a multi-scale neighborhood feature extraction module, a first scale pressure feature extraction unit, a second scale pressure feature extraction unit and a first scale pressure feature extraction unit, wherein the pressure input vector is input into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first scale pressure feature vector, the first convolution layer is provided with a first one-dimensional convolution kernel with a first length; and a multi-scale cascading unit, configured to cascade the first-scale pressure feature vector and the second-scale pressure feature vector to obtain the pressure feature vector.
In one example, in the safety management system 100 for coal bed methane drilling engineering described above, the fusion module 150 includes: the global similarity matrix construction unit is used for constructing a global similarity matrix between the coalbed methane concentration eigenvector and the pressure eigenvector, wherein the eigenvalues of each position on the non-diagonal positions in the global similarity matrix are variances between the eigenvalues of the coalbed methane concentration eigenvector and the eigenvalues of the corresponding two positions in the pressure eigenvector; the covariance matrix construction unit is used for constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector; the inter-manifold multi-granularity association feature extraction unit is used for aggregating the global similarity matrix and the covariance matrix along the channel dimension and then obtaining an inter-manifold global association multi-granularity feature matrix through a convolution layer-based inter-manifold multi-granularity association feature extractor; the matrix decomposition unit is used for performing matrix decomposition based on the characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold association characteristic values; and a manifold related feature value arrangement unit configured to extract a predetermined number of manifold related feature values from the plurality of manifold related feature values, and arrange the predetermined number of manifold related feature values to obtain the classification feature vector.
In one example, in the safety management system 100 for coal bed methane drilling engineering, the safety condition determining module 160 is configured to: processing the classification feature vector using the classifier in a classification formula to generate the classification result; wherein, the classification formula is: wherein />For the classification result, < >>Representing the classification feature vector,/->Weight matrix for full connection layer, +.>Representing the deflection vector of the fully connected layer,is a normalized exponential function.
In summary, the safety management system 100 for coal bed methane drilling engineering according to the embodiment of the application is illustrated, which adopts an artificial intelligence detection technology based on deep learning, and performs feature extraction and fusion on the coal bed methane concentration value and the pressure value in the well to determine whether the well is in a safe condition. Therefore, the safety condition in the well drilling can be intuitively known, the safety problem can be found in time, measures can be taken, and the safety accident can be prevented.
As described above, the safety management system 100 for coal-bed methane drilling engineering according to an embodiment of the present application may be implemented in various wireless terminals, such as a server or the like for safety management for coal-bed methane drilling engineering. In one example, the security management system 100 for coal bed methane drilling projects according to embodiments of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the security management system 100 for coal-bed methane drilling engineering may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the safety management system 100 for coal bed methane drilling engineering may also be one of the many hardware modules of the wireless terminal.
Alternatively, in another example, the security management system 100 for a coalbed methane drilling project may be a separate device from the wireless terminal, and the security management system 100 for a coalbed methane drilling project may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a agreed data format.
Exemplary electronic device
Next, an electronic device according to an embodiment of the present application is described with reference to fig. 6.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the application.
As shown in fig. 6, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that may be executed by the processor 11 to implement the safety management methods for coal bed methane drilling engineering and/or other desired functions of the various embodiments of the present application described above. Various content such as values of coalbed methane concentration and values of pressure in the wellbore at a plurality of predetermined points in time over a predetermined period of time may also be stored in the computer readable storage medium.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
The input means 13 may comprise, for example, a keyboard, a mouse, etc.
The output device 14 can output various information to the outside, including the result of whether the well is in a safe condition or not, etc. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 6 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium
In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in a safety management method for coal-bed methane drilling engineering according to the various embodiments of the application described in the "exemplary methods" section of this specification.
The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the application may also be a computer-readable storage medium, having stored thereon computer program instructions, which when executed by a processor, cause the processor to perform the steps in a safety management method for coal-bed gas drilling engineering according to various embodiments of the application described in the "exemplary methods" section of the specification above.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Claims (10)
1. A method for safety management of a coalbed methane drilling project, comprising:
acquiring coalbed methane concentration values at a plurality of preset time points in a preset time period and pressure values in a well;
arranging the coalbed methane concentration values and the drilling pressure values at a plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension respectively;
the coalbed methane concentration input vector is passed through a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector;
inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector;
fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector; and
the classification feature vector is passed through a classifier to obtain a classification result that indicates whether a safe condition is present within the borehole.
2. The method of claim 1, wherein passing the coalbed methane concentration input vector through a time series encoder comprising a fully connected layer and a one-dimensional convolution layer to obtain a coalbed methane concentration feature vector comprises:
Performing full-connection coding on the coalbed methane concentration input vector by using a full-connection layer of the time sequence coder according to the following full-connection coding formula so as to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector;
wherein, the full-connection coding formula is: wherein />Is the input vector of the concentration of the coalbed methane,is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing a matrix multiplication; and
using a one-dimensional convolution layer of the time sequence encoder to perform one-dimensional convolution encoding on the coalbed methane concentration input vector according to the following one-dimensional convolution encoding formula so as to extract high-dimensional implicit correlation features among feature values of each position in the input vector;
wherein, the one-dimensional convolution coding formula is: wherein ,ais convolution kernel inxWidth in direction, ++>For the convolution kernel parameter vector, +.>For a local vector matrix that operates with a convolution kernel,wfor the size of the convolution kernel +.>Representing the coalbed methane concentration input vector, +.>Representing one-dimensional convolutional encoding of the coalbed methane concentration input vector.
3. The method of claim 2, wherein inputting the pressure input vector into a multi-scale domain feature extraction module to obtain a pressure feature vector comprises:
Inputting the pressure input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first-scale pressure feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length;
inputting the pressure input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second scale pressure feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and concatenating the first scale pressure feature vector and the second scale pressure feature vector to obtain the pressure feature vector.
4. A safety management method for a coalbed methane drilling project according to claim 3, wherein fusing the coalbed methane concentration feature vector and the pressure feature vector to obtain a classification feature vector comprises:
constructing a global similarity matrix between the coalbed methane concentration feature vector and the pressure feature vector, wherein the feature value of each position on the non-diagonal position in the global similarity matrix is the variance between the feature values of the coalbed methane concentration feature vector and the corresponding two positions in the pressure feature vector;
Constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector;
the global similarity matrix and the covariance matrix are aggregated along the channel dimension and then pass through an inter-manifold multi-granularity association feature extractor based on a convolution layer to obtain an inter-manifold global association multi-granularity feature matrix;
performing matrix decomposition based on characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold related characteristic values; and
extracting a predetermined number of manifold-related feature values from the plurality of manifold-related feature values, and arranging the predetermined number of manifold-related feature values to obtain the classification feature vector.
5. The method of claim 4, wherein passing the classification feature vector through a classifier to obtain a classification result, the classification result being indicative of whether a condition within the borehole is safe, comprises: processing the classification feature vector using the classifier in a classification formula to generate the classification result;
wherein, the classification formula is: wherein />For the classification result, < > >Representing the classification feature vector,/->Weight matrix for full connection layer, +.>Representing the deflection vector of the fully connected layer, +.>Is a normalized exponential function.
6. A security management system for coal bed methane drilling engineering, comprising:
the data acquisition module is used for acquiring coalbed methane concentration values and drilling pressure values at a plurality of preset time points in a preset time period;
the data structuring module is used for respectively arranging the coalbed methane concentration values and the drilling pressure values at a plurality of preset time points into a coalbed methane concentration input vector and a pressure input vector according to the time dimension;
the coalbed methane concentration characteristic extraction module is used for enabling the coalbed methane concentration input vector to pass through a time sequence encoder comprising a full-connection layer and a one-dimensional convolution layer to obtain a coalbed methane concentration characteristic vector;
the drilling pressure characteristic extraction module is used for inputting the pressure input vector into the multi-scale domain characteristic extraction module to obtain a pressure characteristic vector;
the fusion module is used for fusing the coalbed methane concentration characteristic vector and the pressure characteristic vector to obtain a classification characteristic vector; and
and the safety condition judging module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result represents whether the drilling well is in a safety condition or not.
7. The safety management system for a coalbed methane drilling project of claim 6, wherein the coalbed methane concentration profile extraction module comprises:
performing full-connection coding on the coalbed methane concentration input vector by using a full-connection layer of the time sequence coder according to the following full-connection coding formula so as to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector;
wherein, the full-connection coding formula is: wherein />Is the input vector of the concentration of the coalbed methane,is the output vector, +.>Is a weight matrix, < >>Is a bias vector, ++>Representing a matrix multiplication; and
using a one-dimensional convolution layer of the time sequence encoder to perform one-dimensional convolution encoding on the coalbed methane concentration input vector according to the following one-dimensional convolution encoding formula so as to extract high-dimensional implicit correlation features among feature values of each position in the input vector;
wherein, the one-dimensional convolution coding formula is: wherein ,ais convolution kernel inxWidth in direction, ++>For the convolution kernel parameter vector, +.>For a local vector matrix that operates with a convolution kernel,wfor the size of the convolution kernel +.>Representing the coalbed methane concentration input vector, +.>Representing one-dimensional convolutional encoding of the coalbed methane concentration input vector.
8. The safety management system for coal bed methane drilling project of claim 7, wherein the drilling pressure feature extraction module comprises:
a first scale pressure feature extraction unit, configured to input the pressure input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first scale pressure feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length;
a second scale pressure feature extraction unit configured to input the pressure input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second scale pressure feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and
and the multi-scale cascading unit is used for cascading the first scale pressure characteristic vector and the second scale pressure characteristic vector to obtain the pressure characteristic vector.
9. The safety management system for coal bed methane drilling project of claim 8, wherein the fusion module comprises:
the global similarity matrix construction unit is used for constructing a global similarity matrix between the coalbed methane concentration eigenvector and the pressure eigenvector, wherein the eigenvalues of each position on the non-diagonal positions in the global similarity matrix are variances between the eigenvalues of the coalbed methane concentration eigenvector and the eigenvalues of the corresponding two positions in the pressure eigenvector;
The covariance matrix construction unit is used for constructing a covariance matrix between the coalbed methane concentration eigenvector and the pressure eigenvector;
the inter-manifold multi-granularity association feature extraction unit is used for aggregating the global similarity matrix and the covariance matrix along the channel dimension and then obtaining an inter-manifold global association multi-granularity feature matrix through a convolution layer-based inter-manifold multi-granularity association feature extractor;
the matrix decomposition unit is used for performing matrix decomposition based on the characteristic values on the manifold inter-domain association multi-granularity characteristic matrix to obtain a plurality of manifold association characteristic values; and
and a manifold correlation feature value arrangement unit configured to extract a predetermined number of manifold correlation feature values from the plurality of manifold correlation feature values, and arrange the predetermined number of manifold correlation feature values to obtain the classification feature vector.
10. The system for safety management of a coalbed methane drilling project according to claim 9, wherein the safety condition determination module is configured to: processing the classification feature vector using the classifier in a classification formula to generate the classification result; wherein, the classification formula is: wherein />For the classification result, < >>Representing the classification feature vector,/->Weight matrix for full connection layer, +.>Representing the deflection vector of the fully connected layer, +.>Is a normalized exponential function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310981644.2A CN116975728B (en) | 2023-08-07 | 2023-08-07 | Safety management method and system for coal bed methane drilling engineering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310981644.2A CN116975728B (en) | 2023-08-07 | 2023-08-07 | Safety management method and system for coal bed methane drilling engineering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116975728A true CN116975728A (en) | 2023-10-31 |
| CN116975728B CN116975728B (en) | 2024-01-26 |
Family
ID=88481135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310981644.2A Active CN116975728B (en) | 2023-08-07 | 2023-08-07 | Safety management method and system for coal bed methane drilling engineering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116975728B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117189071A (en) * | 2023-11-07 | 2023-12-08 | 克拉玛依市远山石油科技有限公司 | Automatic control method for core drilling rig operation |
| CN118309395A (en) * | 2024-06-07 | 2024-07-09 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Intelligent monitoring system and method for coal bed gas drainage and production |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2020104006A4 (en) * | 2020-12-10 | 2021-02-18 | Naval Aviation University | Radar target recognition method based on feature pyramid lightweight convolutional neural network |
| CN115099684A (en) * | 2022-07-18 | 2022-09-23 | 江西中科冠物联网科技有限公司 | Enterprise safety production management system and management method thereof |
| CN115128663A (en) * | 2022-07-11 | 2022-09-30 | 江西中科冠物联网科技有限公司 | Mine safety production detection system based on Internet of things and detection method thereof |
| CN115759658A (en) * | 2022-11-24 | 2023-03-07 | 浙江智慧信息产业有限公司 | Enterprise energy consumption data management system suitable for smart city |
| CN116184972A (en) * | 2023-04-28 | 2023-05-30 | 福建德尔科技股份有限公司 | Control system and method for air separation device |
| CN116526670A (en) * | 2023-04-27 | 2023-08-01 | 杭州保信智能设备有限公司 | Information fusion method for power big data visualization |
| CN116520182A (en) * | 2023-05-11 | 2023-08-01 | 珠海中瑞电力科技有限公司 | DCS device power supply module abnormality early diagnosis method |
-
2023
- 2023-08-07 CN CN202310981644.2A patent/CN116975728B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2020104006A4 (en) * | 2020-12-10 | 2021-02-18 | Naval Aviation University | Radar target recognition method based on feature pyramid lightweight convolutional neural network |
| CN115128663A (en) * | 2022-07-11 | 2022-09-30 | 江西中科冠物联网科技有限公司 | Mine safety production detection system based on Internet of things and detection method thereof |
| CN115099684A (en) * | 2022-07-18 | 2022-09-23 | 江西中科冠物联网科技有限公司 | Enterprise safety production management system and management method thereof |
| CN115759658A (en) * | 2022-11-24 | 2023-03-07 | 浙江智慧信息产业有限公司 | Enterprise energy consumption data management system suitable for smart city |
| CN116526670A (en) * | 2023-04-27 | 2023-08-01 | 杭州保信智能设备有限公司 | Information fusion method for power big data visualization |
| CN116184972A (en) * | 2023-04-28 | 2023-05-30 | 福建德尔科技股份有限公司 | Control system and method for air separation device |
| CN116520182A (en) * | 2023-05-11 | 2023-08-01 | 珠海中瑞电力科技有限公司 | DCS device power supply module abnormality early diagnosis method |
Non-Patent Citations (5)
| Title |
|---|
| HONGTAI CHEN ETC.: "Automatic Modulation Classification Using Multi-Scale Convolutional Neural Network", 2020 IEEE 31ST ANNUAL INTERNATIONAL SYMPOSIUM ON PERSONAL, INDOOR AND MOBILE RADIO COMMUNICATIONS * |
| 乐守群;梁海波;赵庆;李建民;靳文博;: "钻井风险实时诊断技术概念设计", 钻采工艺, no. 01 * |
| 吴俊;管鲁阳;鲍明;许耀华;叶炜;: "基于多尺度一维卷积神经网络的光纤振动事件识别", 光电工程, no. 05 * |
| 常亮 等: "基于相似规律和神经网络的多级多相混输泵气液增压性能预测", 应用数学和力学, vol. 44, no. 6 * |
| 马永杰;程时升;马芸婷;陈敏;: "多尺度特征融合与极限学习机结合的交通标志识别", 液晶与显示, no. 06 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117189071A (en) * | 2023-11-07 | 2023-12-08 | 克拉玛依市远山石油科技有限公司 | Automatic control method for core drilling rig operation |
| CN118309395A (en) * | 2024-06-07 | 2024-07-09 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Intelligent monitoring system and method for coal bed gas drainage and production |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116975728B (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116975728B (en) | Safety management method and system for coal bed methane drilling engineering | |
| US10706229B2 (en) | Content aware heterogeneous log pattern comparative analysis engine | |
| CN115783923A (en) | Elevator fault mode identification system based on big data | |
| CN115171317A (en) | Internet of things smart home method and system and electronic equipment | |
| CN116309580B (en) | Oil and gas pipeline corrosion detection method based on magnetic stress | |
| CN116055293B (en) | Remote fault monitoring method of router and router | |
| CN116866054A (en) | Public information safety monitoring system and method thereof | |
| CN117874633B (en) | Network data asset portrayal generation method and device based on deep learning algorithm | |
| CN116759053A (en) | Medical system prevention and control method and system based on Internet of things system | |
| CN117596057A (en) | Network information security management system and method | |
| CN117115743A (en) | Mining safety production monitoring system and method thereof | |
| CN117124138A (en) | Automatic processing system and method for mold parts | |
| CN117176433A (en) | Abnormal behavior detection system and method for network data | |
| CN113469247B (en) | Network asset abnormity detection method | |
| CN116821831A (en) | Intelligent electric power inspection system and method thereof | |
| CN117316462A (en) | Medical data management method | |
| CN117527524A (en) | Communication network fault detection system and method based on big data technology | |
| CN117231590A (en) | Fault prediction system and method for hydraulic system | |
| CN116958697A (en) | Image-based power safety inspection system and method thereof | |
| Gámiz et al. | Dynamic reliability and sensitivity analysis based on HMM models with Markovian signal process | |
| CN117421655A (en) | Industrial Internet data stream anomaly detection method and system | |
| CN117268760A (en) | Bearing state monitoring system and method | |
| CN114580472B (en) | Large-scale equipment fault prediction method with repeated cause and effect and attention in industrial internet | |
| CN116311739A (en) | Multi-sensor fire detection method based on long-short-term memory network and environment information fusion | |
| CN118054111B (en) | Lithium battery pack safety management method and device, storage medium and electronic equipment |
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 |