CN116680557A - Real-time monitoring system and method for coal bed gas drilling engineering - Google Patents
Real-time monitoring system and method for coal bed gas drilling engineering Download PDFInfo
- Publication number
- CN116680557A CN116680557A CN202310971828.0A CN202310971828A CN116680557A CN 116680557 A CN116680557 A CN 116680557A CN 202310971828 A CN202310971828 A CN 202310971828A CN 116680557 A CN116680557 A CN 116680557A
- Authority
- CN
- China
- Prior art keywords
- matrix
- feature
- parameter
- vector
- full
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005553 drilling Methods 0.000 title claims abstract description 136
- 238000012544 monitoring process Methods 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 79
- 239000003245 coal Substances 0.000 title claims abstract description 42
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims abstract description 142
- 230000008569 process Effects 0.000 claims abstract description 40
- 239000013598 vector Substances 0.000 claims description 254
- 239000011159 matrix material Substances 0.000 claims description 201
- 238000000605 extraction Methods 0.000 claims description 42
- 238000009826 distribution Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 21
- 238000010586 diagram Methods 0.000 claims description 18
- 230000004927 fusion Effects 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 17
- 230000004913 activation Effects 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000007480 spreading Effects 0.000 claims description 4
- 238000003892 spreading Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 12
- 239000007789 gas Substances 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000007405 data analysis Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 230000018109 developmental process Effects 0.000 description 5
- 239000012530 fluid Substances 0.000 description 5
- 238000003062 neural network model Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 238000013433 optimization analysis Methods 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000013480 data collection Methods 0.000 description 3
- 230000001934 delay Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 241000282412 Homo Species 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Classifications
-
- 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
-
- 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
- E21B44/00—Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application relates to the field of intelligent monitoring, and particularly discloses a real-time monitoring system and a real-time monitoring method for coal bed methane drilling engineering. Thus, a real-time monitoring scheme for the coalbed methane drilling engineering is constructed to synthesize each parameter value in the drilling process, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
Description
Technical Field
The application relates to the field of intelligent detection, in particular to a real-time monitoring system and a real-time monitoring method for coal bed methane drilling engineering.
Background
The coal bed gas drilling engineering refers to an engineering process of exploiting coal bed gas in a coal bed. Coalbed methane is a natural gas consisting essentially of methane and is stored in pores in the coal bed. The coalbed methane drilling engineering comprises links of exploration, drilling, well completion, production and the like, and aims to exploit and utilize coalbed methane through a drilling technology. The underground condition needs to be detected in the coalbed methane drilling engineering, and the underground condition is mastered: 1. and the drilling parameters can be adjusted in time by operators so as to optimize the drilling process and improve the drilling efficiency. 2. Risk early warning: monitoring can help discover potential risks and anomalies early, such as elevated wellhead temperatures, abnormal fluctuations in drilling fluid pressure, etc. Corresponding measures can be taken through timely risk early warning, accidents are avoided, and the safety of staff is guaranteed. The common monitoring method for coal bed gas drilling engineering may have the following disadvantages: 1. delay performance: common monitoring methods may not be able to acquire downhole data in real time, with some delay. Such delays may lead to reduced timeliness of decisions and adjustments and failure to timely cope with changes in downhole conditions. 2. Relying on manual operations: some common monitoring methods may require manual operations for data collection and monitoring, which increases labor costs and may present a risk of human error. 3. Technology update hysteresis: common monitoring methods may not keep pace with the development of new technologies in time, resulting in relative hysteresis in the performance and functionality of the monitoring methods.
Thus, an optimized real-time monitoring scheme for coal bed methane drilling projects is desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a real-time monitoring system and a real-time monitoring method for coal bed methane drilling engineering, which use an artificial intelligence technology based on a deep neural network model to intelligently perform feature coding and extraction on each parameter value in the drilling process so as to obtain a more accurate classification label for representing whether an early warning of an abnormal situation is given out. Thus, a real-time monitoring scheme for the coalbed methane drilling engineering is constructed to synthesize each parameter value in the drilling process, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
According to one aspect of the present application, there is provided a real-time monitoring system for a coal bed methane drilling project, comprising:
the data acquisition module is used for acquiring each parameter value in the drilling process in a preset time;
the matrixing module is used for arranging all parameter values in the drilling process in the preset time into a parameter full-time input matrix according to the time dimension and the sample dimension;
the multi-scale feature extraction module is used for enabling the parameter full-time sequence input matrix to pass through the multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence association matrix;
The segmentation module is used for carrying out feature matrix segmentation on the multi-scale parameter full-time sequence incidence matrix into a plurality of parameter time sequence incidence submatrices;
the context coding module is used for expanding the plurality of parameter time sequence incidence submatrices based on row vectors or column vectors and then obtaining classification feature vectors through a context coder based on a converter;
and the detection result generation module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out or not.
In the above real-time monitoring system for coal bed gas drilling engineering, the multi-scale feature extraction module includes:
the first scale feature extraction unit is used for enabling the parameter full-time sequence input matrix to pass through a first convolution layer of the multi-scale feature extraction module to obtain a first scale parameter full-time sequence incidence matrix;
the second scale feature extraction unit is used for enabling the parameter full-time sequence input matrix to pass through a second convolution layer of the multi-scale feature extraction module to obtain a second scale parameter full-time sequence incidence matrix;
and the probability density space association unit is used for carrying out probability density space association between feature matrixes on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix so as to obtain a multi-scale parameter full-time sequence association matrix.
In the above real-time monitoring system for coal bed gas drilling engineering, the first scale feature extraction unit is configured to:
carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram;
carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix;
and performing nonlinear activation on the pooling feature matrix to obtain the first scale parameter full-time-sequence correlation matrix.
In the above real-time monitoring system for coal bed gas drilling engineering, the second scale feature extraction unit is configured to:
carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram;
carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix;
and performing nonlinear activation on the pooling feature matrix to obtain the second scale parameter full-time sequence correlation matrix.
In the above real-time monitoring system for coal bed gas drilling engineering, the probability density space association unit includes:
the line-by-line expansion subunit is used for carrying out line-by-line expansion on the first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix so as to obtain a first characteristic vector and a second characteristic vector;
A probability density calculating subunit, configured to calculate probability density functions of the first feature vector and the second feature vector to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector;
a mutual information calculation subunit, configured to calculate mutual information between probability density distribution of the first feature vector and probability density distribution of the second feature vector;
a fusion subunit, configured to select an adapted fusion policy to fuse the first feature vector and the second feature vector to obtain a fused feature vector based on a comparison between the mutual information and a predetermined threshold;
and the restoring subunit is used for restoring the fusion feature vector to obtain the multi-scale parameter full-time sequence correlation matrix.
In the above real-time monitoring system for coal bed gas drilling engineering, the context coding module includes:
a spreading unit, configured to spread the plurality of parameter timing correlation submatrices into a plurality of parameter feature vectors based on row vectors or column vectors;
a query vector construction unit, configured to arrange the plurality of parameter feature vectors into an input vector;
the vector conversion unit is used for respectively converting the input vector into a query vector and a key vector through a learning embedding matrix;
A self-attention unit, configured to calculate a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix;
the normalization unit is used for performing normalization processing on the self-attention association matrix to obtain a normalized self-attention association matrix;
the attention calculating unit is used for inputting the standardized self-attention association matrix into a Softmax activation function to activate so as to obtain a self-attention feature matrix;
an attention applying unit, configured to multiply the self-attention feature matrix with each parameter feature vector in the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors;
and the cascading unit is used for cascading the plurality of context parameter feature vectors to obtain the classification feature vector.
In the above real-time monitoring system for coal bed gas drilling engineering, the detection result generating module includes:
the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors;
and the classification result generating unit is used for inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
According to another aspect of the present application, there is also provided a real-time monitoring method for coal bed methane drilling engineering, including:
acquiring each parameter value in the drilling process in a preset time;
arranging all parameter values in the drilling process within the preset time into a parameter full-time input matrix according to a time dimension and a sample dimension;
the parameter full-time sequence input matrix is processed through a multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence association matrix;
the multi-scale parameter full-time sequence incidence matrix is subjected to feature matrix segmentation to form a plurality of parameter time sequence incidence submatrices;
the plurality of parameter time sequence incidence submatrices are unfolded based on row vectors or column vectors and then pass through a context encoder based on a converter to obtain classification feature vectors;
and the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out or not.
Compared with the prior art, the real-time monitoring system and the real-time monitoring method for the coal bed methane drilling engineering provided by the application use an artificial intelligence technology based on a deep neural network model to intelligently perform feature coding and extraction on each parameter value in the drilling process so as to obtain a more accurate classification label for representing whether an early warning of an abnormal situation is given out. Thus, a real-time monitoring scheme for the coalbed methane drilling engineering is constructed to synthesize each parameter value in the drilling process, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
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 illustrates a block diagram of a real-time monitoring system for coal-bed methane drilling projects in accordance with an embodiment of the present application.
FIG. 2 illustrates a system architecture diagram of a real-time monitoring system for coal-bed methane drilling projects in accordance with an embodiment of the present application.
FIG. 3 illustrates a block diagram of a multi-scale feature extraction module in a real-time monitoring system for coal-bed methane drilling engineering in accordance with an embodiment of the application.
FIG. 4 illustrates a block diagram of a context encoding module in a real-time monitoring system for coal-bed methane drilling engineering, in accordance with an embodiment of the present application.
FIG. 5 illustrates a flow chart of a real-time monitoring method for a coalbed methane drilling project in accordance with an embodiment of the application.
Fig. 6 illustrates a block diagram of an electronic device according to an embodiment of the application.
Detailed Description
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 described above in the background, the following drawbacks may exist in the conventional monitoring method for coal bed methane drilling engineering: 1. delay performance: common monitoring methods may not be able to acquire downhole data in real time, with some delay. Such delays may lead to reduced timeliness of decisions and adjustments and failure to timely cope with changes in downhole conditions. 2. Relying on manual operations: some common monitoring methods may require manual operations for data collection and monitoring, which increases labor costs and may present a risk of human error. 3. Technology update hysteresis: common monitoring methods may not keep pace with the development of new technologies in time, resulting in relative hysteresis in the performance and functionality of the monitoring methods. Thus, an optimized real-time monitoring scheme for coal bed methane drilling projects is desired.
Aiming at the technical problems, a real-time monitoring scheme for coal bed methane drilling engineering is provided, which uses an artificial intelligence technology based on a deep neural network model to intelligently perform feature coding and extraction on each parameter value in the drilling process so as to obtain a more accurate classification label for representing whether abnormal early warning is given out. Therefore, by constructing a real-time monitoring scheme for the coalbed methane drilling engineering, various parameter values in the drilling process are synthesized, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech 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.
In recent years, the development of deep learning and neural networks has provided new solutions and solutions for real-time monitoring systems for coal bed methane drilling engineering.
Specifically, values of various parameters during drilling are obtained for a predetermined time. Many parameters are involved in the coalbed methane drilling process, such as the flow rate, pressure, temperature of the drilling fluid, the rotational speed and the down force of the drill bit, etc. Variations in these parameters may reflect conditions and performance during drilling. By acquiring the values of the parameters, various conditions in the drilling process can be monitored in real time, including the circulation condition of drilling fluid, the working state of a drill bit, the stability of a well wall and the like.
And then, arranging each parameter value in the drilling process in the preset time into a parameter full-time input matrix according to the time dimension and the sample dimension. By arranging the parameters according to the time dimension, the real-time monitoring of the parameter values in the drilling process can be realized. In this way, abnormal conditions or potential problems can be found in time, and corresponding measures are adopted to adjust and optimize so as to ensure the safe and efficient implementation of drilling engineering. The parameters are arranged according to the dimension of the samples, so that the comparison and analysis of the parameter values of different samples can be realized. By comparing the parameter values of different samples, the rule and trend in the drilling process can be found out, and the drilling process and operation mode are further optimized. The arrangement mode of the parameter full-time sequence input matrix can provide convenience for subsequent data analysis and modeling. By carrying out data analysis on the parameter full-time input matrix, useful information and characteristics can be extracted, and the work such as anomaly detection, prediction, optimization analysis and the like can be carried out, so that the efficiency and economic benefit of drilling engineering are improved.
And then, the parameter full-time sequence input matrix passes through a multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence incidence matrix. During drilling, complex correlations may exist between different parameters, which may exist on different time scales. By using a convolutional neural network model, features between parameters can be extracted on different scales, thereby capturing more associated information. The multi-scale feature extraction module may process the input matrix through convolution kernels of different sizes to extract features on different time scales. In this way, the correlation between parameters at different time scales can be captured by taking the short-term and long-term time sequence characteristics into consideration. By obtaining the multi-scale parameter full-time sequence incidence matrix, the interaction and influence among the parameters can be better understood, and further data analysis and modeling can be performed. This helps to find rules and trends hidden in the data, providing more accurate predictive and optimization analysis.
And then, carrying out feature matrix segmentation on the multi-scale parameter full-time sequence incidence matrix into a plurality of parameter time sequence incidence submatrices. The multi-scale parameter full-time sequence incidence matrix is divided into a plurality of parameter time sequence incidence submatrices, so that the time sequence incidence of different parameters can be separated and independently analyzed. This allows a better view of the degree of correlation between each parameter and the other parameters, finding interactions and effects between them. Splitting into multiple parameter timing correlation sub-matrices may also provide finer granularity analysis. The correlations between different parameters may be different in different time periods or at specific times, and by splitting the submatrices, the changes in these correlations may be more accurately located and analyzed. The split sub-matrix also helps to reduce computational complexity and improve computational efficiency. For a large-scale parameter full-time sequence incidence matrix, the full-time sequence incidence matrix is segmented into a plurality of submatrices, the calculation load is reduced, different submatrices can be processed in parallel, and the analysis speed is increased.
Then, the plurality of parameter time sequence association submatrices are expanded based on row vectors or column vectors and then pass through a context encoder based on a converter to obtain classification feature vectors. After the time sequence correlation submatrices of the parameters are unfolded according to row vectors or column vectors, the time sequence correlation information among the parameters can be converted into a spatial relation of the matrix. These expanded matrices can be encoded and modeled by a context encoder based on the converter, thereby extracting classification feature vectors. A converter-based context encoder is a powerful sequence modeling tool that can learn global dependencies and context information in an input sequence. It weights the different positions in the sequence by a self-attention mechanism so that each position can obtain information of other positions. Thus, by encoding the expanded matrix, timing correlation information between parameters can be captured while taking into account their context in the overall sequence. The classification feature vectors obtained by the context encoder based on the converter may be used for subsequent classification, prediction or decision tasks. The method comprises the steps of encoding and modeling, wherein the method comprises the time sequence association information between parameters, and has higher expression capability and semantic information. Therefore, after the time sequence correlation submatrices of the parameters are unfolded, the context encoder based on the converter is used for obtaining the classification feature vector, so that the correlation information among the parameters can be better utilized, and the performance and accuracy of the model are improved.
Specifically, first, a plurality of parameter timing correlation submatrices are expanded into a plurality of parameter feature vectors based on row vectors or column vectors. The plurality of parametric feature vectors are arranged as input vectors. The input vectors are then converted into query vectors and key vectors, respectively, by a learnable embedding matrix. Then, a product between the query vector and a transpose vector of the key vector is calculated to obtain a self-attention correlation matrix. And then, carrying out standardization processing on the self-attention association matrix to obtain a standardized self-attention association matrix. The normalized self-attention correlation matrix is then activated by inputting a Softmax activation function to obtain a self-attention feature matrix. And then multiplying the self-attention feature matrix with each parameter feature vector in the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors. The plurality of context parameter feature vectors are then concatenated to obtain the classification feature vector.
And then, the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out. Therefore, by constructing a real-time monitoring scheme for the coalbed methane drilling engineering, various parameter values in the drilling process are synthesized, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
In particular, the first scale parameter full-time-sequence correlation matrix and the second scale parameter full-time-sequence correlation matrix are considered to be the correlation parameter features extracted through the convolutional neural network model. Since both feature matrices are obtained by processing the same set of input vectors, they may capture similar feature information. Thus, there may be some degree of duplication of information when they are fused. The first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix are extracted by convolution kernels with different scales respectively. The convolution kernels of different scales can capture characteristic information of different scales. However, these feature information may be relevant to some extent because they all come from the same set of input vectors. Thus, a degree of redundancy of information may occur when the first scale and the second scale associated parameter feature matrices are fused. And in order to avoid information redundancy, carrying out probability density space association between feature matrices on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix. Therefore, for the fused feature matrix, further feature analysis and processing can be performed to improve the classification robustness of the multi-scale parameter full-time sequence correlation matrix.
Performing probability density space association between feature matrices on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix, wherein the method comprises the following steps: performing row-wise expansion on the first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix to obtain a first feature vector and a second feature vector; calculating probability density functions of the first feature vector and the second feature vector to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector; calculating mutual information between probability density distribution of the first feature vector and probability density distribution of the second feature vector; selecting an adapted fusion strategy to fuse the first feature vector and the second feature vector to obtain a fused feature vector based on a comparison between the mutual information and a predetermined threshold; and restoring the fusion feature vector to obtain the multi-scale parameter full-time sequence association matrix.
And carrying out probability density space association between feature vectors on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix, wherein an optimal direction is essentially searched in a high-dimensional feature space, so that the projection in the direction can reflect the similarity or the difference of the two feature matrices to the greatest extent, and the direction is the multi-scale parameter full-time sequence association matrix. In this way, redundant information between two feature matrices is eliminated to improve the robustness of the multi-scale parametric full-time-series correlation matrix.
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 System
FIG. 1 illustrates a block diagram of a real-time monitoring system for coal-bed methane drilling projects in accordance with an embodiment of the present application. As shown in fig. 1, a real-time monitoring system 100 for coal-bed methane drilling engineering according to an embodiment of the present application includes: a data acquisition module 110 for acquiring values of various parameters during drilling within a predetermined time; a matrixing module 120, configured to arrange each parameter value in the drilling process in the predetermined time into a parameter full-time input matrix according to a time dimension and a sample dimension; the multi-scale feature extraction module 130 is configured to pass the parameter full-time input matrix through the multi-scale feature extraction module to obtain a multi-scale parameter full-time correlation matrix; the segmentation module 140 is configured to segment the feature matrix of the multi-scale parameter full-time-sequence correlation matrix into a plurality of parameter time-sequence correlation sub-matrices; a context encoding module 150, configured to obtain a classification feature vector by using a context encoder based on a converter after expanding the plurality of parameter timing correlation sub-matrices based on row vectors or column vectors; and a detection result generating module 160, configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether to send out an early warning of an abnormal situation.
FIG. 2 illustrates a system architecture diagram of a real-time monitoring system for coal-bed methane drilling projects in accordance with an embodiment of the present application. In the system architecture, as shown in fig. 2, first, values of various parameters during drilling are acquired for a predetermined time. And then, arranging each parameter value in the drilling process in the preset time into a parameter full-time input matrix according to the time dimension and the sample dimension. And then, the parameter full-time sequence input matrix passes through a multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence incidence matrix. And then, carrying out feature matrix segmentation on the multi-scale parameter full-time sequence incidence matrix into a plurality of parameter time sequence incidence submatrices. Then, the plurality of parameter time sequence association submatrices are expanded based on row vectors or column vectors and then pass through a context encoder based on a converter to obtain classification feature vectors. And then, the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out.
In the above real-time monitoring system 100 for coal-bed methane drilling engineering, the data acquisition module 110 is configured to acquire values of parameters during drilling in a predetermined time. As described above in the background, the following drawbacks may exist in the conventional monitoring method for coal bed methane drilling engineering: 1. delay performance: common monitoring methods may not be able to acquire downhole data in real time, with some delay. Such delays may lead to reduced timeliness of decisions and adjustments and failure to timely cope with changes in downhole conditions. 2. Relying on manual operations: some common monitoring methods may require manual operations for data collection and monitoring, which increases labor costs and may present a risk of human error. 3. Technology update hysteresis: common monitoring methods may not keep pace with the development of new technologies in time, resulting in relative hysteresis in the performance and functionality of the monitoring methods. Thus, an optimized real-time monitoring scheme for coal bed methane drilling projects is desired.
Aiming at the technical problems, a real-time monitoring scheme for coal bed methane drilling engineering is provided, which uses an artificial intelligence technology based on a deep neural network model to intelligently perform feature coding and extraction on each parameter value in the drilling process so as to obtain a more accurate classification label for representing whether abnormal early warning is given out. Therefore, by constructing a real-time monitoring scheme for the coalbed methane drilling engineering, various parameter values in the drilling process are synthesized, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech 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.
In recent years, the development of deep learning and neural networks has provided new solutions and solutions for real-time monitoring systems for coal bed methane drilling engineering.
Specifically, values of various parameters during drilling are obtained for a predetermined time. Many parameters are involved in the coalbed methane drilling process, such as the flow rate, pressure, temperature of the drilling fluid, the rotational speed and the down force of the drill bit, etc. Variations in these parameters may reflect conditions and performance during drilling. By acquiring the values of the parameters, various conditions in the drilling process can be monitored in real time, including the circulation condition of drilling fluid, the working state of a drill bit, the stability of a well wall and the like.
In the above real-time monitoring system 100 for coal bed methane drilling engineering, the matrixing module 120 is configured to arrange each parameter value in the drilling process in the predetermined time into a parameter full-time input matrix according to a time dimension and a sample dimension. By arranging the parameters according to the time dimension, the real-time monitoring of the parameter values in the drilling process can be realized. In this way, abnormal conditions or potential problems can be found in time, and corresponding measures are adopted to adjust and optimize so as to ensure the safe and efficient implementation of drilling engineering. The parameters are arranged according to the dimension of the samples, so that the comparison and analysis of the parameter values of different samples can be realized. By comparing the parameter values of different samples, the rule and trend in the drilling process can be found out, and the drilling process and operation mode are further optimized. The arrangement mode of the parameter full-time sequence input matrix can provide convenience for subsequent data analysis and modeling. By carrying out data analysis on the parameter full-time input matrix, useful information and characteristics can be extracted, and the work such as anomaly detection, prediction, optimization analysis and the like can be carried out, so that the efficiency and economic benefit of drilling engineering are improved.
In the above real-time monitoring system 100 for coal bed methane drilling engineering, the multi-scale feature extraction module 130 is configured to obtain the multi-scale parameter full-time correlation matrix by passing the parameter full-time input matrix through the multi-scale feature extraction module. During drilling, complex correlations may exist between different parameters, which may exist on different time scales. By using a convolutional neural network model, features between parameters can be extracted on different scales, thereby capturing more associated information. The multi-scale feature extraction module may process the input matrix through convolution kernels of different sizes to extract features on different time scales. In this way, the correlation between parameters at different time scales can be captured by taking the short-term and long-term time sequence characteristics into consideration. By obtaining the multi-scale parameter full-time sequence incidence matrix, the interaction and influence among the parameters can be better understood, and further data analysis and modeling can be performed. This helps to find rules and trends hidden in the data, providing more accurate predictive and optimization analysis.
In particular, the first scale parameter full-time-sequence correlation matrix and the second scale parameter full-time-sequence correlation matrix are considered to be the correlation parameter features extracted through the convolutional neural network model. Since both feature matrices are obtained by processing the same set of input vectors, they may capture similar feature information. Thus, there may be some degree of duplication of information when they are fused. The first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix are extracted by convolution kernels with different scales respectively. The convolution kernels of different scales can capture characteristic information of different scales. However, these feature information may be relevant to some extent because they all come from the same set of input vectors. Thus, a degree of redundancy of information may occur when the first scale and the second scale associated parameter feature matrices are fused. And in order to avoid information redundancy, carrying out probability density space association between feature matrices on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix. Therefore, for the fused feature matrix, further feature analysis and processing can be performed to improve the classification robustness of the multi-scale parameter full-time sequence correlation matrix.
Performing probability density space association between feature matrices on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix, wherein the method comprises the following steps: performing row-wise expansion on the first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix to obtain a first feature vector and a second feature vector; calculating probability density functions of the first feature vector and the second feature vector to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector; calculating mutual information between probability density distribution of the first feature vector and probability density distribution of the second feature vector; selecting an adapted fusion strategy to fuse the first feature vector and the second feature vector to obtain a fused feature vector based on a comparison between the mutual information and a predetermined threshold; and restoring the fusion feature vector to obtain the multi-scale parameter full-time sequence association matrix.
And carrying out probability density space association between feature vectors on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix, wherein an optimal direction is essentially searched in a high-dimensional feature space, so that the projection in the direction can reflect the similarity or the difference of the two feature matrices to the greatest extent, and the direction is the multi-scale parameter full-time sequence association matrix. In this way, redundant information between two feature matrices is eliminated to improve the classification robustness of the multi-scale parameter full-time sequence correlation matrix.
FIG. 3 illustrates a block diagram of a multi-scale feature extraction module in a real-time monitoring system for coal-bed methane drilling engineering in accordance with an embodiment of the application. As shown in fig. 3, the multi-scale feature extraction module includes: a first scale feature extraction unit 131, configured to pass the parameter full-time sequence input matrix through a first convolution layer of the multi-scale feature extraction module to obtain a first scale parameter full-time sequence correlation matrix; a second scale feature extraction unit 132, configured to pass the parameter full-time-sequence input matrix through a second convolution layer of the multi-scale feature extraction module to obtain a second scale parameter full-time-sequence correlation matrix; and a probability density space association unit 133, configured to perform probability density space association between feature matrices on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix to obtain a multi-scale parameter full-time sequence association matrix.
Specifically, in the embodiment of the present application, the first scale feature extraction unit 131 is configured to: carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram; carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix; and performing nonlinear activation on the pooled feature matrix to obtain the first scale parameter full-time-sequence correlation matrix.
Specifically, in the embodiment of the present application, the second scale feature extraction unit 132 is configured to: carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram; carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix; and performing nonlinear activation on the pooling feature matrix to obtain the second scale parameter full-time sequence correlation matrix.
Specifically, in the embodiment of the present application, the probability density space association unit 133 includes: the line-by-line expansion subunit is used for carrying out line-by-line expansion on the first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix so as to obtain a first characteristic vector and a second characteristic vector; a probability density calculating subunit, configured to calculate probability density functions of the first feature vector and the second feature vector to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector; a mutual information calculation subunit, configured to calculate mutual information between probability density distribution of the first feature vector and probability density distribution of the second feature vector; a fusion subunit, configured to select an adapted fusion policy to fuse the first feature vector and the second feature vector to obtain the fused feature vector based on a comparison between the mutual information and a predetermined threshold; and the restoring subunit is used for restoring the fusion feature vector to obtain the multi-scale parameter full-time sequence correlation matrix.
Specifically, in the embodiment of the present application, probability density functions of the first feature vector and the second feature vector are calculated to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector: for each eigenvalue, the frequency of its occurrence in the eigenvector is calculated, and the probability density function can be estimated by counting the frequency of the eigenvalues.
Specifically, in an embodiment of the present application, selecting an adapted fusion policy to fuse the first feature vector and the second feature vector to obtain the fused feature vector based on a comparison between the mutual information and a predetermined threshold value, includes: different fusion strategies can be selected according to the association degree of the feature vectors. The following are some common fusion strategies: 1. weighted average: when the degree of association between the first feature vector and the second feature vector is high, they may be fused using a weighted average. The weights can be distributed according to the association degree, and the feature vector weight with higher association degree is larger. This strategy is applicable where there is similarity or correlation between feature vectors. 2. Characteristic splicing: when the degree of association between the first feature vector and the second feature vector is low, it may be selected to splice them together to form a longer feature vector. Such a strategy may preserve independent information for each feature vector, but may increase the dimension of the feature space. 3. Linear combination: for feature vectors that are highly correlated, linear combinations may be used to fuse them. By adjusting the coefficients of the linear combination, the degree of contribution of each feature vector in the fusion result can be controlled. The strategy can flexibly adjust the weight of the feature vector to adapt to different association degrees. 4. Feature selection: when the degree of association between the first feature vector and the second feature vector is large, it may be selected to retain only the feature vector having a high degree of association, and discard the feature vector having a low degree of association. Such a strategy may simplify feature space and reduce feature dimensions, but may lose a portion of the information. 5. Nonlinear transformation: in some cases, the degree of association between feature vectors may not be linear. It is contemplated that non-linear transformations may be used to fuse feature vectors, such as feature fusion using kernel methods or neural network models.
In the above real-time monitoring system 100 for coal bed methane drilling engineering, the splitting module 140 is configured to split the feature matrix of the multi-scale parameter full-time-sequence correlation matrix into a plurality of parameter time-sequence correlation submatrices. The multi-scale parameter full-time sequence incidence matrix is divided into a plurality of parameter time sequence incidence submatrices, so that the time sequence incidence of different parameters can be separated and independently analyzed. This allows a better view of the degree of correlation between each parameter and the other parameters, finding interactions and effects between them. Splitting into multiple parameter timing correlation sub-matrices may also provide finer granularity analysis. The correlations between different parameters may be different in different time periods or at specific times, and by splitting the submatrices, the changes in these correlations may be more accurately located and analyzed. The split sub-matrix also helps to reduce computational complexity and improve computational efficiency. For a large-scale parameter full-time sequence incidence matrix, the full-time sequence incidence matrix is segmented into a plurality of submatrices, the calculation load is reduced, different submatrices can be processed in parallel, and the analysis speed is increased.
In the above-mentioned real-time monitoring system 100 for coal-bed methane drilling engineering, the context encoding module 150 is configured to obtain the classification feature vector by using a context encoder based on a converter after expanding the plurality of parameter time-sequence correlation sub-matrices based on row vectors or column vectors. After the time sequence correlation submatrices of the parameters are unfolded according to row vectors or column vectors, the time sequence correlation information among the parameters can be converted into a spatial relation of the matrix. These expanded matrices can be encoded and modeled by a context encoder based on the converter, thereby extracting classification feature vectors. A converter-based context encoder is a powerful sequence modeling tool that can learn global dependencies and context information in an input sequence. It weights the different positions in the sequence by a self-attention mechanism so that each position can obtain information of other positions. Thus, by encoding the expanded matrix, timing correlation information between parameters can be captured while taking into account their context in the overall sequence. The classification feature vectors obtained by the context encoder based on the converter may be used for subsequent classification, prediction or decision tasks. The method comprises the steps of encoding and modeling, wherein the method comprises the time sequence association information between parameters, and has higher expression capability and semantic information. Therefore, after the time sequence correlation submatrices of the parameters are unfolded, the context encoder based on the converter is used for obtaining the classification feature vector, so that the correlation information among the parameters can be better utilized, and the performance and accuracy of the model are improved.
Specifically, first, a plurality of parameter timing correlation submatrices are expanded into a plurality of parameter feature vectors based on row vectors or column vectors. The plurality of parametric feature vectors are arranged as input vectors. The input vectors are then converted into query vectors and key vectors, respectively, by a learnable embedding matrix. Then, a product between the query vector and a transpose vector of the key vector is calculated to obtain a self-attention correlation matrix. And then, carrying out standardization processing on the self-attention association matrix to obtain a standardized self-attention association matrix. The normalized self-attention correlation matrix is then activated by inputting a Softmax activation function to obtain a self-attention feature matrix. And then multiplying the self-attention feature matrix with each parameter feature vector in the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors. The plurality of context parameter feature vectors are then concatenated to obtain the classification feature vector.
FIG. 4 illustrates a block diagram of a context encoding module in a real-time monitoring system for coal-bed methane drilling engineering, in accordance with an embodiment of the present application. As shown in fig. 4, the context encoding module includes: a spreading unit 151 for spreading the plurality of parameter timing correlation submatrices into a plurality of parameter feature vectors based on row vectors or column vectors; a query vector construction unit 152, configured to arrange the plurality of parameter feature vectors into an input vector; a vector conversion unit 153, configured to convert the input vector into a query vector and a key vector through a learning embedding matrix, respectively; a self-attention unit 154 for calculating a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix; a normalization unit 155, configured to perform normalization processing on the self-attention correlation matrix to obtain a normalized self-attention correlation matrix; a attention calculating unit 156, configured to activate the normalized self-attention correlation matrix input Softmax activation function to obtain a self-attention feature matrix; an attention applying unit 157, configured to multiply the self-attention feature matrix with each of the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors; and a concatenation unit 158, configured to concatenate the plurality of context parameter feature vectors to obtain the classification feature vector.
In the above real-time monitoring system 100 for coal bed methane drilling engineering, the detection result generating module 160 is configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether to send out an early warning of an abnormal situation. Therefore, by constructing a real-time monitoring scheme for the coalbed methane drilling engineering, various parameter values in the drilling process are synthesized, and the real-time performance and accuracy of the coalbed methane drilling engineering are improved based on the classification result.
Specifically, in the embodiment of the present application, the detection result generating module 160 includes: the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors; and the classification result generating unit is used for inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
Exemplary method
FIG. 5 illustrates a flow chart of a real-time monitoring method for a coalbed methane drilling project in accordance with an embodiment of the application. As shown in fig. 5, the real-time monitoring method for coal bed methane drilling engineering according to the embodiment of the application includes the following steps: s110, acquiring each parameter value in the drilling process in a preset time; s120, arranging all parameter values in the drilling process within the preset time into a parameter full-time input matrix according to a time dimension and a sample dimension; s130, the parameter full-time sequence input matrix passes through a multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence incidence matrix; s140, feature matrix segmentation is carried out on the multi-scale parameter full-time sequence incidence matrix to obtain a plurality of parameter time sequence incidence submatrices; s150, expanding the plurality of parameter time sequence association submatrices based on row vectors or column vectors, and then obtaining classification feature vectors through a context encoder based on a converter; and S160, the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal situation is sent out or not.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described real-time monitoring method for a coal-bed methane drilling process have been described in detail in the above description of the real-time monitoring system for a coal-bed methane drilling process with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.
As described above, the real-time monitoring system 100 for a coalbed methane drilling project according to an embodiment of the present application may be implemented in various terminal devices, such as a real-time monitoring server for a coalbed methane drilling project, and the like. In one example, the real-time monitoring system 100 for coal-bed methane drilling engineering according to embodiments of the present application may be integrated into the terminal equipment as a software module and/or hardware module. For example, the real-time monitoring system 100 for coal-bed gas drilling engineering may be a software module in the operating system of the terminal equipment, or may be an application developed for the terminal equipment; of course, the real-time monitoring system 100 for coal-bed methane drilling engineering may also be one of the numerous hardware modules of the terminal equipment.
Alternatively, in another example, the real-time monitoring system 100 for coal-bed methane drilling engineering and the terminal equipment may be separate devices, and the real-time monitoring system 100 for coal-bed methane drilling engineering may be connected to the terminal equipment through a wired and/or wireless network, and transmit interactive information according to 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 perform the functions in the real-time monitoring method 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 various parameter values during drilling for a predetermined 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 whether or not an abnormal situation is given, and the like. 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.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.
The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (10)
1. A real-time monitoring system for coal bed gas drilling engineering, comprising:
the data acquisition module is used for acquiring each parameter value in the drilling process in a preset time;
the matrixing module is used for arranging all parameter values in the drilling process in the preset time into a parameter full-time input matrix according to the time dimension and the sample dimension;
the multi-scale feature extraction module is used for enabling the parameter full-time sequence input matrix to pass through the multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence association matrix;
the segmentation module is used for carrying out feature matrix segmentation on the multi-scale parameter full-time sequence incidence matrix into a plurality of parameter time sequence incidence submatrices;
the context coding module is used for expanding the plurality of parameter time sequence incidence submatrices based on row vectors or column vectors and then obtaining classification feature vectors through a context coder based on a converter;
And the detection result generation module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out or not.
2. The real-time monitoring system for coal bed methane drilling engineering of claim 1, wherein the multi-scale feature extraction module comprises:
the first scale feature extraction unit is used for enabling the parameter full-time sequence input matrix to pass through a first convolution layer of the multi-scale feature extraction module to obtain a first scale parameter full-time sequence incidence matrix;
the second scale feature extraction unit is used for enabling the parameter full-time sequence input matrix to pass through a second convolution layer of the multi-scale feature extraction module to obtain a second scale parameter full-time sequence incidence matrix;
and the probability density space association unit is used for carrying out probability density space association between feature matrixes on the first scale parameter full-time sequence association matrix and the second scale parameter full-time sequence association matrix so as to obtain a multi-scale parameter full-time sequence association matrix.
3. The real-time monitoring system for coal bed methane drilling engineering according to claim 2, wherein the first scale feature extraction unit is configured to:
Carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram;
carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix;
and performing nonlinear activation on the pooling feature matrix to obtain the first scale parameter full-time-sequence correlation matrix.
4. A real time monitoring system for coal bed methane drilling engineering according to claim 3, wherein the second scale feature extraction unit is configured to:
carrying out convolution processing on the parameter full-time sequence input matrix to obtain a convolution characteristic diagram;
carrying out global average pooling treatment along the channel dimension on the convolution feature map to obtain a pooled feature matrix;
and performing nonlinear activation on the pooling feature matrix to obtain the second scale parameter full-time sequence correlation matrix.
5. The real-time monitoring system for coal bed methane drilling engineering according to claim 4, wherein the probability density space correlation unit comprises:
the line-by-line expansion subunit is used for carrying out line-by-line expansion on the first scale parameter full-time sequence incidence matrix and the second scale parameter full-time sequence incidence matrix so as to obtain a first characteristic vector and a second characteristic vector;
A probability density calculating subunit, configured to calculate probability density functions of the first feature vector and the second feature vector to obtain probability density distribution of the first feature vector and probability density distribution of the second feature vector;
a mutual information calculation subunit, configured to calculate mutual information between probability density distribution of the first feature vector and probability density distribution of the second feature vector;
a fusion subunit, configured to select an adapted fusion policy to fuse the first feature vector and the second feature vector to obtain a fused feature vector based on a comparison between the mutual information and a predetermined threshold;
and the restoring subunit is used for restoring the fusion feature vector to obtain the multi-scale parameter full-time sequence correlation matrix.
6. The real-time monitoring system for coal bed methane drilling engineering according to claim 5, wherein the context encoding module comprises:
a spreading unit, configured to spread the plurality of parameter timing correlation submatrices into a plurality of parameter feature vectors based on row vectors or column vectors;
a query vector construction unit, configured to arrange the plurality of parameter feature vectors into an input vector;
The vector conversion unit is used for respectively converting the input vector into a query vector and a key vector through a learning embedding matrix;
a self-attention unit, configured to calculate a product between the query vector and a transpose vector of the key vector to obtain a self-attention correlation matrix;
the normalization unit is used for performing normalization processing on the self-attention association matrix to obtain a normalized self-attention association matrix;
the attention calculating unit is used for inputting the standardized self-attention association matrix into a Softmax activation function to activate so as to obtain a self-attention feature matrix;
an attention applying unit, configured to multiply the self-attention feature matrix with each parameter feature vector in the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors;
and the cascading unit is used for cascading the plurality of context parameter feature vectors to obtain the classification feature vector.
7. The real-time monitoring system for coal bed methane drilling engineering according to claim 6, wherein the detection result generating module comprises:
the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors;
And the classification result generating unit is used for inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
8. A real-time monitoring method for coal bed methane drilling engineering, comprising:
acquiring each parameter value in the drilling process in a preset time;
arranging all parameter values in the drilling process within the preset time into a parameter full-time input matrix according to a time dimension and a sample dimension;
the parameter full-time sequence input matrix is processed through a multi-scale feature extraction module to obtain a multi-scale parameter full-time sequence association matrix;
the multi-scale parameter full-time sequence incidence matrix is subjected to feature matrix segmentation to form a plurality of parameter time sequence incidence submatrices;
the plurality of parameter time sequence incidence submatrices are unfolded based on row vectors or column vectors and then pass through a context encoder based on a converter to obtain classification feature vectors;
and the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether an early warning of an abnormal condition is sent out or not.
9. The method of claim 8, wherein developing the plurality of parameter timing correlation sub-matrices based on row vectors or column vectors and then passing through a transducer-based context encoder to obtain classification feature vectors comprises:
Expanding the plurality of parameter time sequence association submatrices into a plurality of parameter feature vectors based on row vectors or column vectors;
arranging the plurality of parameter feature vectors into an input vector;
respectively converting the input vector into a query vector and a key vector through a learning embedding matrix;
calculating the product between the query vector and the transpose vector of the key vector to obtain a self-attention correlation matrix;
carrying out standardization processing on the self-attention association matrix to obtain a standardized self-attention association matrix;
inputting the standardized self-attention association matrix into a Softmax activation function to activate so as to obtain a self-attention feature matrix;
multiplying the self-attention feature matrix with each parameter feature vector in the plurality of parameter feature vectors to obtain the plurality of context parameter feature vectors;
and cascading the plurality of context parameter feature vectors to obtain the classification feature vector.
10. The method of claim 9, wherein inputting the encoded classification feature vector into a Softmax classification function of the classifier to obtain the classification result comprises:
Performing full-connection coding on the classification feature vectors by using a full-connection layer of the classifier to obtain coded classification feature vectors;
and inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310971828.0A CN116680557B (en) | 2023-08-03 | 2023-08-03 | Real-time monitoring system and method for coal bed gas drilling engineering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310971828.0A CN116680557B (en) | 2023-08-03 | 2023-08-03 | Real-time monitoring system and method for coal bed gas drilling engineering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116680557A true CN116680557A (en) | 2023-09-01 |
| CN116680557B CN116680557B (en) | 2023-10-27 |
Family
ID=87785942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310971828.0A Active CN116680557B (en) | 2023-08-03 | 2023-08-03 | Real-time monitoring system and method for coal bed gas drilling engineering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116680557B (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 |
| CN117211758A (en) * | 2023-11-07 | 2023-12-12 | 克拉玛依市远山石油科技有限公司 | Intelligent drilling control system and method for shallow hole coring |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6064768A (en) * | 1996-07-29 | 2000-05-16 | Wisconsin Alumni Research Foundation | Multiscale feature detector using filter banks |
| KR101007409B1 (en) * | 2010-05-26 | 2011-01-14 | 삼성탈레스 주식회사 | Apparatus and method for processing image fusion signal for improvement of target detection |
| US11210801B1 (en) * | 2020-10-26 | 2021-12-28 | Tsinghua University | Adaptive multi-sensor data fusion method and system based on mutual information |
| WO2022198218A1 (en) * | 2021-03-16 | 2022-09-22 | The University Of Chicago | Spectral correlation analysis of layered evolutionary signals |
| CN116092216A (en) * | 2023-01-09 | 2023-05-09 | 江西航盛电子科技有限公司 | Vehicle state early warning method and system based on multi-sensor information fusion |
| CN116257406A (en) * | 2023-03-15 | 2023-06-13 | 湖南华彩伟业网络科技有限公司 | Gateway data management method and system for smart city |
| CN116454772A (en) * | 2023-06-14 | 2023-07-18 | 浙江浙能迈领环境科技有限公司 | Decompression device and method for medium-voltage distribution cabinet of container |
| CN116451145A (en) * | 2023-03-06 | 2023-07-18 | 河南茵特赛尔生物技术有限公司 | Cooling device and method for stem cell transportation |
| CN116504382A (en) * | 2023-05-29 | 2023-07-28 | 杭州医锐联科技有限公司 | Remote medical monitoring system and method thereof |
-
2023
- 2023-08-03 CN CN202310971828.0A patent/CN116680557B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6064768A (en) * | 1996-07-29 | 2000-05-16 | Wisconsin Alumni Research Foundation | Multiscale feature detector using filter banks |
| KR101007409B1 (en) * | 2010-05-26 | 2011-01-14 | 삼성탈레스 주식회사 | Apparatus and method for processing image fusion signal for improvement of target detection |
| US11210801B1 (en) * | 2020-10-26 | 2021-12-28 | Tsinghua University | Adaptive multi-sensor data fusion method and system based on mutual information |
| WO2022198218A1 (en) * | 2021-03-16 | 2022-09-22 | The University Of Chicago | Spectral correlation analysis of layered evolutionary signals |
| CN116092216A (en) * | 2023-01-09 | 2023-05-09 | 江西航盛电子科技有限公司 | Vehicle state early warning method and system based on multi-sensor information fusion |
| CN116451145A (en) * | 2023-03-06 | 2023-07-18 | 河南茵特赛尔生物技术有限公司 | Cooling device and method for stem cell transportation |
| CN116257406A (en) * | 2023-03-15 | 2023-06-13 | 湖南华彩伟业网络科技有限公司 | Gateway data management method and system for smart city |
| CN116504382A (en) * | 2023-05-29 | 2023-07-28 | 杭州医锐联科技有限公司 | Remote medical monitoring system and method thereof |
| CN116454772A (en) * | 2023-06-14 | 2023-07-18 | 浙江浙能迈领环境科技有限公司 | Decompression device and method for medium-voltage distribution cabinet of container |
Non-Patent Citations (2)
| Title |
|---|
| XIFENG GUO等: ""A short-term load forecasting model of multi-scale CNN-LSTM hybrid neural network considering the real-time electricity price"", 《ENERGY REPORTS》, vol. 6 * |
| 李刚等: ""能源互联网关联数据融合的互信息方法"", 《电力建设》, vol. 37, no. 9 * |
Cited By (3)
| 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 |
| CN117211758A (en) * | 2023-11-07 | 2023-12-12 | 克拉玛依市远山石油科技有限公司 | Intelligent drilling control system and method for shallow hole coring |
| CN117211758B (en) * | 2023-11-07 | 2024-04-02 | 克拉玛依市远山石油科技有限公司 | Intelligent drilling control system and method for shallow hole coring |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116680557B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116680557B (en) | Real-time monitoring system and method for coal bed gas drilling engineering | |
| Lee et al. | Integrating machine learning in embedded sensor systems for Internet-of-Things applications | |
| CN115693918B (en) | Comprehensive intelligent electricity utilization system and method for building | |
| CN113327279B (en) | Point cloud data processing method and device, computer equipment and storage medium | |
| CN116627708A (en) | Storage fault analysis system and method thereof | |
| CN116245513A (en) | Automatic operation and maintenance system and method based on rule base | |
| CN115859437A (en) | Jacket underwater stress detection system based on distributed optical fiber sensing system | |
| CN116665086A (en) | Teaching method and system based on intelligent analysis of learning behaviors | |
| CN115471216B (en) | Data management method of intelligent laboratory management platform | |
| CN115759658A (en) | Enterprise energy consumption data management system suitable for smart city | |
| CN112764738A (en) | Code automatic generation method and system based on multi-view program characteristics | |
| CN117058622A (en) | Intelligent monitoring system and method for sewage treatment equipment | |
| CN114387567A (en) | Video data processing method and device, electronic equipment and storage medium | |
| CN116596150A (en) | Event prediction method of transform Hoxwell process model based on multi-branch self-attention | |
| CN117787287A (en) | Accident handling method, system and storage medium based on large model | |
| CN116975728B (en) | Safety management method and system for coal bed methane drilling engineering | |
| CN114626426A (en) | Industrial equipment behavior detection method based on K-means optimization algorithm | |
| CN117034123A (en) | Fault monitoring system and method for fitness equipment | |
| CN117011609A (en) | Automatic tracking system and method for textile processing progress | |
| CN116216234A (en) | Intelligent monitoring system and method for coal conveying system based on video | |
| CN112949300B (en) | Typhoon early warning planning model automatic generation method and system based on deep learning | |
| CN113159164B (en) | Industrial Internet equipment collaborative operation method based on distribution type | |
| CN117612071B (en) | Video action recognition method based on transfer learning | |
| Ayodeji et al. | Wavy-attention network for real-time cyber-attack detection in a small modular pressurized water reactor digital control system | |
| CN117073055A (en) | Hydraulic balance control system and method thereof |
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 |