CN116181287A - Shale gas well production abnormal condition early warning system and method - Google Patents
Shale gas well production abnormal condition early warning system and method Download PDFInfo
- Publication number
- CN116181287A CN116181287A CN202310090698.XA CN202310090698A CN116181287A CN 116181287 A CN116181287 A CN 116181287A CN 202310090698 A CN202310090698 A CN 202310090698A CN 116181287 A CN116181287 A CN 116181287A
- Authority
- CN
- China
- Prior art keywords
- shale gas
- sound
- early warning
- abnormal
- wellhead
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 94
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000013145 classification model Methods 0.000 claims abstract description 37
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 230000005236 sound signal Effects 0.000 claims description 20
- 239000004576 sand Substances 0.000 claims description 17
- 238000001514 detection method Methods 0.000 claims description 14
- 230000009467 reduction Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000009432 framing Methods 0.000 claims description 8
- 239000007788 liquid Substances 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000013473 artificial intelligence Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 5
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 4
- 238000010248 power generation Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 239000013598 vector Substances 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 206010030113 Oedema Diseases 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B43/00—Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells
- E21B43/006—Production of coal-bed methane
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B41/00—Equipment or details not covered by groups E21B15/00 - E21B40/00
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B47/00—Survey of boreholes or wells
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH DRILLING; MINING
- E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B47/00—Survey of boreholes or wells
- E21B47/12—Means for transmitting measuring-signals or control signals from the well to the surface, or from the surface to the well, e.g. for logging while drilling
- E21B47/14—Means for transmitting measuring-signals or control signals from the well to the surface, or from the surface to the well, e.g. for logging while drilling using acoustic waves
Abstract
The invention discloses a shale gas well production abnormal condition early warning system and method, which mainly comprise a preprocessing module, a feature extraction module, a classification model building module, a model matching module and an early warning module. The method comprises the steps of collecting a large amount of shale gas well production sound information, preprocessing the shale gas well production sound information, extracting features to form feature vectors, and finally establishing a model such as a shaft effusion model, a shaft sand-out model, a pipeline leakage model and a pipeline blockage model. The method comprises the steps of placing a model into a wellhead abnormal working condition early warning system, collecting sound data of shale gas wellhead production in real time through a sound receiver, preprocessing the sound data through an abnormal working condition module placed in a microcomputer, extracting characteristic information, matching the characteristic information with a built-in abnormal working condition classification model, analyzing a matching result, and sending abnormal working condition early warning information to a remote production control decision module through an information receiving transmitter.
Description
Technical Field
The invention relates to the technical field of shale gas exploitation, in particular to a shale gas well production abnormal condition early warning system and method.
Background
At present, in the development process of shale gas, a large-scale hydraulic sand fracturing transformation technology is adopted in the early stage, and in the production process of a gas well, a gas-liquid two-phase flowing state is formed in a shaft, and certain fracturing sand and shaft impurities sometimes return. The abnormal phenomena of shaft liquid accumulation, shaft sand discharge, pipeline thorn leakage, pipeline blockage and the like gradually occur in the gas well, so that the gas well cannot be produced normally. Therefore, a set of shale gas well production process abnormality diagnosis method is established, accurate gas well abnormality diagnosis is a basis for gas well production management, and the method has guiding significance for efficient development of shale gas wells.
Aiming at abnormal working condition early warning, the gas well multi-parameter combined early warning model research and application calculates corresponding parameter values through corresponding algorithms to early warn the abnormal working conditions, and pushes abnormal conditions and treatment comments according to preset values; according to a plunger gas lift dynamics model in the exploration of a shale gas well production early warning method in the Chuan nan area, judging whether a well bottom is filled with liquid or not and whether a pipeline is blocked or not through theoretical models such as the oil jacket pressure and the like; predicting bottom hole effusion according to a Turner critical carrier fluid flow model by a low-pressure low-yield shale gas well intelligent production optimization method; the method for judging and managing sand production of the Fuling shale gas well judges whether the gas well produces sand or not according to the instantaneous flow of the gas well. The existing abnormal working condition early warning technology is mainly based on a theoretical model, the model has more assumption conditions, the truest working condition cannot be restored, the early warning accuracy is low, meanwhile, the predicted abnormal working condition is single, and working conditions such as accumulated liquid in a shaft, sand discharge in the shaft, pipeline leakage and pipeline blockage cannot be identified at the same time.
Disclosure of Invention
The invention provides a shale gas well production abnormal condition early warning system and method, which aim to solve the problems that the shale gas well production abnormal condition early warning method is complex and low in accuracy.
According to the first aspect, the shale gas well production abnormal condition early warning method is based on artificial intelligence and comprises the following steps:
step S1: collecting a large amount of shale gas production well wellhead sound information, enhancing sound through preprocessing, detecting end points, framing and windowing, and extracting characteristics by combining a mel cepstrum coefficient, short-time energy and short-time zero crossing rate;
step S2: training and identifying the model by taking the extracted characteristic parameters as the input of the classification model, and finally forming a classification model containing abnormal working conditions of accumulated liquid of a shaft, sand discharge of the shaft, thorny leakage of a pipeline and blockage of the pipeline;
step S3: placing the trained abnormal condition classification model into a wellhead abnormal condition early warning system, and collecting sound data of shale gas wellhead production in real time through a sound receiver;
step S4: the sound data is preprocessed through an abnormal working condition module arranged in the microcomputer, characteristic information is extracted and matched with a built-in abnormal working condition classification model, a matching result is analyzed, and abnormal working condition early warning information is sent to a remote production control decision module through an information receiving transmitter.
Further, the method further comprises the following steps: the photovoltaic wind power generation system is used for charging the battery, and the battery provides power for the well site production control system, so that the normal operation of the whole system is ensured.
Further, the collection of the acoustic data includes collecting an effective shale gas production acoustic signal using a directional acoustic collector.
Further, the voice information preprocessing further comprises the step of screening effective voice signals of the wellhead by detecting the voice signal energy acquired from the wellhead of the shale gas, wherein the calculation formula of the voice signal energy of the wellhead of the shale gas is as follows:
wherein x (n) represents an acquired wellhead sound signal sequence; n represents the sound sequence length; e represents wellhead acoustic signal energy.
Further, the voice information preprocessing further comprises noise reduction processing of collected voice through Fourier transformation, and the noise reduction processing formula is as follows:
where Yi (w) is a fourier transformed value of Yi (m), D (w) is a fourier transformed value of di (m), xi (w) is a fourier transformed value of Xi (m), α is an over-subtraction factor, and β is a gain compensation factor.
Further, the voice information preprocessing further includes endpoint detection of the collected voice, and detection of abnormal voice by a single-parameter double-threshold endpoint detection method of short-time energy, wherein a short-time energy calculation formula is as follows:
wherein N is the frame length, the length of each frame after the sound is divided into frames, m is the sampling point number of the abnormal sound, and N is the frame number.
Furthermore, the abnormal working condition classification model utilizes a convolutional neural network to build a classification model, one or more pairs of convolutional layers and sampling layers replace fully-connected hidden layers, one or more fully-connected layers are added to the top layer, and the output layer calculates posterior probability of each state according to the input characteristic information, so that the aim of identification is finally achieved.
On the other hand, the shale gas well production abnormal condition early warning system is used for realizing an artificial intelligence-based shale gas well production abnormal condition early warning method, and comprises the following steps:
and a pretreatment module: preprocessing collected shale gas production well wellhead sound information;
and the feature extraction module is used for: enhancing the preprocessed sound information, detecting end points, framing and windowing, and extracting features by combining a Mel cepstrum coefficient, short-time energy and a short-time zero crossing rate;
a classification model module: the method comprises the steps of establishing a wellbore effusion, wellbore sand production, pipeline puncture and leakage and pipeline blockage classification model;
model matching module: matching the sound information characteristics with the classification model, and analyzing the matching result;
abnormal condition early warning module: and early warning is carried out on the abnormal working condition according to the matching result, and an adjustment instruction is sent to the wellhead.
The invention has the beneficial effects that: the invention provides a shale gas well production abnormal condition early warning system and method, which mainly comprise a preprocessing module, a feature extraction module, a classification model building module, a model matching module and an early warning module. The method comprises the steps of collecting a large amount of shale gas well production sound information, preprocessing the shale gas well production sound information, extracting features to form feature vectors, and finally establishing a model such as a shaft effusion model, a shaft sand-out model, a pipeline leakage model and a pipeline blockage model. The method is characterized in that a model is placed into a wellhead abnormal working condition early warning system, sound data during shale gas wellhead production is collected in real time through a sound receiver, the sound data is preprocessed through an abnormal working condition module placed in a microcomputer, characteristic information is extracted and matched with a built-in abnormal working condition classification model, a matching result is analyzed, and abnormal working condition early warning information is sent to a remote production control decision module through an information receiving transmitter, so that the method has the following advantages:
(1) The technology can predict abnormal working conditions such as accumulated liquid of a shaft, sand discharge of the shaft, pipeline penetration, pipeline blockage and the like, and predicted parameters are more comprehensive and can be more fit with the actual field;
(2) The sound collector is used for collecting sound in the shale gas production process, the sound recognition technology is used for early warning of abnormal working conditions, and compared with the traditional theoretical model, the method has the advantages that the accuracy is greatly improved;
(3) The abnormal working condition module is placed in the well site microcomputer, so that all analysis work is directly completed in the well site, and only specific abnormal working condition types are finally uploaded, so that a large amount of data is not required to be uploaded to the remote production control decision module for identification, the transmission quantity of the data is greatly reduced, and meanwhile, the early warning advance is improved.
Drawings
FIG. 1 is a schematic diagram of a shale gas well production control system in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram of an abnormal sound recognition technique according to an embodiment of the present invention;
FIG. 3 is a schematic illustration of wellbore fluid sound noise reduction in an embodiment of the present invention;
FIG. 4 is a schematic illustration of wellbore sand production sound noise reduction in an embodiment of the present invention;
FIG. 5 is a schematic diagram of noise reduction of pipeline penetration sound in an embodiment of the invention;
FIG. 6 is a schematic diagram of noise reduction of a pipeline blockage sound in an embodiment of the invention.
Detailed Description
For a clearer understanding of technical features, objects, and effects of the present invention, a specific embodiment of the present invention will be described with reference to the accompanying drawings.
The invention provides a shale gas well production abnormal condition early warning system and method, and in the first aspect, the shale gas well production abnormal condition early warning method is based on artificial intelligence and comprises the following steps:
step S1: collecting a large amount of shale gas production well wellhead sound information, enhancing sound through preprocessing, detecting end points, framing and windowing, and extracting characteristics by combining a mel cepstrum coefficient, short-time energy and short-time zero crossing rate;
step S2: training and identifying the model by taking the extracted characteristic parameters as the input of the classification model, and finally forming a classification model containing abnormal working conditions of accumulated liquid of a shaft, sand discharge of the shaft, thorny leakage of a pipeline and blockage of the pipeline;
step S3: placing the trained abnormal condition classification model into a wellhead abnormal condition early warning system, and collecting sound data of shale gas wellhead production in real time through a sound receiver;
step S4: the sound data is preprocessed through an abnormal working condition module arranged in the microcomputer, characteristic information is extracted and matched with a built-in abnormal working condition classification model, a matching result is analyzed, and abnormal working condition early warning information is sent to a remote production control decision module through an information receiving transmitter.
Further, the method further comprises the following steps: the photovoltaic wind power generation system is used for charging the battery, and the battery provides power for the well site production control system, so that the normal operation of the whole system is ensured.
Further, the collection of the acoustic data includes collecting an effective shale gas production acoustic signal using a directional acoustic collector.
Further, the voice information preprocessing further comprises the step of screening effective voice signals of the wellhead by detecting the voice signal energy acquired from the wellhead of the shale gas, wherein the calculation formula of the voice signal energy of the wellhead of the shale gas is as follows:
wherein x (n) represents an acquired wellhead sound signal sequence; n represents the sound sequence length; e represents wellhead acoustic signal energy.
Further, the voice information preprocessing further comprises noise reduction processing of collected voice through Fourier transformation, and the noise reduction processing formula is as follows:
where Yi (w) is a fourier transformed value of Yi (m), D (w) is a fourier transformed value of di (m), xi (w) is a fourier transformed value of Xi (m), α is an over-subtraction factor, and β is a gain compensation factor.
Further, the voice information preprocessing further includes endpoint detection of the collected voice, and detection of abnormal voice by a single-parameter double-threshold endpoint detection method of short-time energy, wherein a short-time energy calculation formula is as follows:
wherein N is the frame length, the length of each frame after the sound is divided into frames, m is the sampling point number of the abnormal sound, and N is the frame number.
Furthermore, the abnormal working condition classification model utilizes a convolutional neural network to build a classification model, one or more pairs of convolutional layers and sampling layers replace fully-connected hidden layers, one or more fully-connected layers are added to the top layer, and the output layer calculates posterior probability of each state according to the input characteristic information, so that the aim of identification is finally achieved.
On the other hand, the shale gas well production abnormal condition early warning system is used for realizing an artificial intelligence-based shale gas well production abnormal condition early warning method, and comprises the following steps:
and a pretreatment module: preprocessing collected shale gas production well wellhead sound information;
and the feature extraction module is used for: enhancing the preprocessed sound information, detecting end points, framing and windowing, and extracting features by combining a Mel cepstrum coefficient, short-time energy and a short-time zero crossing rate;
a classification model module: the method comprises the steps of establishing a wellbore effusion, wellbore sand production, pipeline puncture and leakage and pipeline blockage classification model;
model matching module: matching the sound information characteristics with the classification model, and analyzing the matching result;
abnormal condition early warning module: and early warning is carried out on the abnormal working condition according to the matching result, and an adjustment instruction is sent to the wellhead.
In this embodiment, in the first step, as shown in fig. 1, in order to ensure power supply of wellhead equipment under various weather conditions, a photovoltaic wind power generation system is installed near a well site to charge a battery, and the battery provides a power supply for a well site production control system to ensure normal operation of the whole system;
secondly, collecting a large amount of shale gas production well wellhead sound information, carrying out enhancement, endpoint detection, framing and windowing on sound through preprocessing, extracting features by combining a mel cepstrum coefficient, short-time energy and short-time zero crossing rate, training and identifying feature parameters as input of a classification model, and finally forming a classification model comprising a shaft dropsy, a shaft sand outlet, a pipeline thorny, a pipeline blockage and the like, as shown in fig. 2;
(1) Shale gas production sound signal acquisition
Because the environment of shale gas production is severe and the condition is complex, in order to better collect effective shale gas production sound signals, a directional sound collector is adopted, and the device has the characteristics of water resistance, explosion resistance, directional pickup, long pickup distance and the like;
(2) Shale gas production abnormal sound type selection
The shale gas wellhead can produce different sound signals under different production working conditions, and the patent selects shaft effusion, shaft sand production, pipeline thorny leakage and pipeline blockage as a research object.
(3) Sound signal validity detection
The screening of the effective acoustic signals at the wellhead is realized by detecting the acoustic signal energy acquired from the shale gas wellhead. The energy of the sound signal acquired in real time is equal to the sum of the energy of all the acquisition points, and the energy of the sound signal obtained through calculation is compared with a set threshold value, so that an effective sound signal is selected for the next processing. The shale gas wellhead sound signal energy calculation formula is as follows:
wherein x (n) represents an acquired wellhead sound signal sequence; n represents the sound sequence length; e represents wellhead acoustic signal energy.
(4) Sound signal noise reduction processing
The sound signals collected by the pickup are used in the wellhead practical environment, so that the sound signals produced by shale gas are contained, and a large amount of noise signals are inevitably doped. There is a lot of noise in the wellhead environment, such as equipment noise, environmental bird sounds, wellhead maintenance knocks, etc., as the purity of the shale gas produced acoustic signals directly affects the identification of the later acoustic signals. Therefore, it is important to reduce noise of the sound collected in the wellhead. The corresponding noise reduction results under the four working conditions are shown in fig. 3, fig. 4, fig. 5 and fig. 6.
Wherein Y is i (w) is y i (m) Fourier transformThe value after the conversion, D (w) is D i (m) Fourier transformed values, X i (w) is x i (m) fourier transformed values, α being an over-subtraction factor, β being a gain compensation factor.
(5) Endpoint detection
The sound signal still has a silent section after the noise of the sound under the abnormal working condition is removed, and in order to prevent the silent section from interfering with the recognition effect, the sound signal needs to be subjected to end point detection to determine the starting point and the end point of the effective sound. The method for detecting the abnormal sound by using the single-parameter double-threshold endpoint detection method based on short-time energy is selected, wherein the short-time energy calculation formula is as follows:
wherein N is the frame length, the length of each frame after the sound is divided into frames, m is the sampling point number of the abnormal sound, and N is the frame number.
(6) Extracting features of mel-pattern
Through analysis, various sound samples under abnormal working conditions are found to be within 3s, the length of the intercepted sound is selected to be 3s, each section of sound is guaranteed to contain one period of abnormal sound pronunciation, a Mel filter is designed to filter the preprocessed abnormal working condition sounds, and finally a Mel spectrogram of a sound signal is obtained.
(7) Classification model selection
In this embodiment, a classification model is built by using a convolutional neural network, which is characterized by a mel spectrogram of an abnormal-condition sound signal. Convolutional neural networks can be regarded as a variant of standard neural networks in that a fully connected hidden layer is replaced by one or more pairs of convolutional layers and sampling layers. In addition, one or more full connection layers are added on the top layer of the convolutional neural network, so that the characteristics of all frequency bands are connected and integrated into one-dimensional characteristic information before reaching the output layer, and direct classification is facilitated. And the output layer calculates the posterior probability of each state according to the input characteristic information, and finally achieves the aim of identification.
Thirdly, an abnormal working condition early warning method for shale gas well production based on artificial intelligence is characterized in that the abnormal working condition model trained in the second step is put into a wellhead microcomputer in advance;
the fourth step, the abnormal working condition early warning method of shale gas well production based on artificial intelligence is characterized in that sound information during shale gas production is collected through a wellhead sound receiver, sound is enhanced, end point detection and framing and windowing are carried out through a microcomputer built-in program, characteristics are extracted through combination of a mel cepstrum coefficient, short-time energy and short-time zero crossing rate, and the characteristic value is matched with a built-in shaft dropsy, shaft sand, pipeline puncture and leakage, pipeline blockage and other similar models;
and fifthly, the sound receiver can collect sound information during wellhead production in real time, and the collected sound information is processed by the built-in algorithm program. Firstly, denoising original data, wherein a built-in noise reduction spectrum subtraction formula is as follows, and secondly, removing background noise well through Fourier transformation. Finally, matching the extracted feature vector with a built-in abnormal working condition model, and judging the type of the abnormal working condition;
step six, the wellhead directly uploads the abnormal working condition type to a remote production control decision-making module through an information receiving and sending module, an operator can send an adjustment instruction to the wellhead according to the abnormal type, and the wellhead control module can adjust the opening of an electric valve or an electric oil nozzle to control whether the shale gas is produced or not or adjust the shale gas yield;
seventh, when the abnormal working condition can not be solved through remote control, the remote operator can send early warning information to the well site operator to perform manual adjustment.
Therefore, the invention adopts the voice recognition technology to early warn the abnormal working condition, compared with the traditional theoretical model for early warning the abnormal working condition, the accuracy is greatly improved, and the early warning advance is greatly improved. Compared with the prior art, the method has outstanding substantive characteristics and remarkable progress.
The invention provides a shale gas well production abnormal condition early warning system and method, which mainly comprise a preprocessing module, a feature extraction module, a classification model building module, a model matching module and an early warning module. The method comprises the steps of collecting a large amount of shale gas well production sound information, preprocessing the shale gas well production sound information, extracting features to form feature vectors, and finally establishing a model such as a shaft effusion model, a shaft sand-out model, a pipeline leakage model and a pipeline blockage model. The method comprises the steps of placing a model into a wellhead abnormal working condition early warning system, collecting sound data of shale gas wellhead production in real time through a sound receiver, preprocessing the sound data through an abnormal working condition module placed in a microcomputer, extracting characteristic information, matching the characteristic information with a built-in abnormal working condition classification model, analyzing a matching result, and sending abnormal working condition early warning information to a remote production control decision module through an information receiving transmitter.
The foregoing has shown and described the basic principles and features of the invention and the advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (8)
1. The shale gas well production abnormal condition early warning method is based on artificial intelligence and is characterized by comprising the following steps of:
step S1: collecting a large amount of shale gas production well wellhead sound information, enhancing sound through preprocessing, detecting end points, framing and windowing, and extracting characteristics by combining a mel cepstrum coefficient, short-time energy and short-time zero crossing rate;
step S2: training and identifying the model by taking the extracted characteristic parameters as the input of the classification model, and finally forming a classification model containing abnormal working conditions of accumulated liquid of a shaft, sand discharge of the shaft, thorny leakage of a pipeline and blockage of the pipeline;
step S3: placing the trained abnormal condition classification model into a wellhead abnormal condition early warning system, and collecting sound data of shale gas wellhead production in real time through a sound receiver;
step S4: the sound data is preprocessed through an abnormal working condition module arranged in the microcomputer, characteristic information is extracted and matched with a built-in abnormal working condition classification model, a matching result is analyzed, and abnormal working condition early warning information is sent to a remote production control decision module through an information receiving transmitter.
2. The shale gas well production anomaly condition early warning method of claim 1, further comprising: the photovoltaic wind power generation system is used for charging the battery, and the battery provides power for the well site production control system, so that the normal operation of the whole system is ensured.
3. The method for pre-warning of abnormal shale gas well production conditions according to claim 1, wherein the collection of sound data comprises the step of collecting effective shale gas production sound signals by using a directional sound collector.
4. The method for early warning of abnormal production conditions of a shale gas well according to claim 1, wherein the voice information preprocessing further comprises the steps of detecting the energy of a voice signal obtained from a shale gas wellhead to screen effective voice signals of the wellhead, and the calculation formula of the voice signal energy of the shale gas wellhead is as follows:
wherein x (n) represents an acquired wellhead sound signal sequence; n represents the sound sequence length; e represents wellhead acoustic signal energy.
5. The method for early warning of abnormal production conditions of a shale gas well according to claim 1, wherein the voice information preprocessing further comprises noise reduction processing of collected voice through fourier transformation, and a noise reduction processing formula is as follows:
where Yi (w) is a fourier transformed value of Yi (m), D (w) is a fourier transformed value of di (m), xi (w) is a fourier transformed value of Xi (m), α is an over-subtraction factor, and β is a gain compensation factor.
6. The method for early warning of abnormal production conditions of a shale gas well according to claim 1, wherein the voice information preprocessing further comprises endpoint detection of collected voice, and the abnormal voice is detected by a single-parameter double-threshold endpoint detection method of short-time energy, wherein a short-time energy calculation formula is as follows:
wherein N is the frame length, the length of each frame after the sound is divided into frames, m is the sampling point number of the abnormal sound, and N is the frame number.
7. The method for early warning of abnormal production conditions of shale gas wells according to claim 1, wherein the abnormal production conditions classification model is characterized in that a convolutional neural network is used for building a classification model, one or more pairs of convolutional layers and sampling layers replace fully-connected hidden layers, one or more fully-connected layers are added to the top layer, the output layer calculates posterior probability of each state according to input characteristic information, and finally the purpose of identification is achieved.
8. A shale gas well production anomaly condition early warning system for implementing the shale gas well production anomaly condition early warning method according to any one of claims 1 to 7, characterized by comprising:
and a pretreatment module: preprocessing collected shale gas production well wellhead sound information;
and the feature extraction module is used for: enhancing the preprocessed sound information, detecting end points, framing and windowing, and extracting features by combining a Mel cepstrum coefficient, short-time energy and a short-time zero crossing rate;
a classification model module: the method comprises the steps of establishing a wellbore effusion, wellbore sand production, pipeline puncture and leakage and pipeline blockage classification model;
model matching module: matching the sound information characteristics with the classification model, and analyzing the matching result;
abnormal condition early warning module: and early warning is carried out on the abnormal working condition according to the matching result, and an adjustment instruction is sent to the wellhead.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310090698.XA CN116181287A (en) | 2023-02-09 | 2023-02-09 | Shale gas well production abnormal condition early warning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310090698.XA CN116181287A (en) | 2023-02-09 | 2023-02-09 | Shale gas well production abnormal condition early warning system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116181287A true CN116181287A (en) | 2023-05-30 |
Family
ID=86439824
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310090698.XA Pending CN116181287A (en) | 2023-02-09 | 2023-02-09 | Shale gas well production abnormal condition early warning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116181287A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117454294A (en) * | 2023-11-02 | 2024-01-26 | 西南石油大学 | Shale gas well plunger process well abnormal condition judging method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
| CN108460144A (en) * | 2018-03-14 | 2018-08-28 | 西安华光信息技术有限责任公司 | A kind of coal equipment fault early-warning system and method based on machine learning |
| CN110191318A (en) * | 2019-05-24 | 2019-08-30 | 大连金盛义电子科技有限公司 | A kind of mine hoisting equipment wireless telecommunications video surveillance security early warning system |
| CN111341334A (en) * | 2020-03-06 | 2020-06-26 | 东莞理工学院 | Noise reduction and abnormal sound detection system and method applied to rail transit |
| CN112164408A (en) * | 2020-10-26 | 2021-01-01 | 南京农业大学 | Pig coughing sound monitoring and early warning system based on deep learning |
| CN112483076A (en) * | 2019-09-11 | 2021-03-12 | 中国石油化工股份有限公司 | System for be used for discerning drilling construction complex situation |
| CN112472065A (en) * | 2020-11-18 | 2021-03-12 | 天机医用机器人技术(清远)有限公司 | Disease detection method based on cough sound recognition and related equipment thereof |
| CN112820275A (en) * | 2021-01-15 | 2021-05-18 | 华中农业大学 | Automatic monitoring method for analyzing abnormality of suckling piglets based on sound signals |
| CN113553724A (en) * | 2021-07-30 | 2021-10-26 | 中国石油化工股份有限公司 | Shale gas well production process pit shaft abnormity analysis and processing system |
| CN115048984A (en) * | 2022-05-17 | 2022-09-13 | 天津农学院 | Sow oestrus recognition method based on deep learning |
-
2023
- 2023-02-09 CN CN202310090698.XA patent/CN116181287A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194068A (en) * | 2017-05-22 | 2017-09-22 | 中国石油大学(北京) | Shale gas fracturing process underground unusual service condition real-time estimate method for early warning and device |
| CN108460144A (en) * | 2018-03-14 | 2018-08-28 | 西安华光信息技术有限责任公司 | A kind of coal equipment fault early-warning system and method based on machine learning |
| CN110191318A (en) * | 2019-05-24 | 2019-08-30 | 大连金盛义电子科技有限公司 | A kind of mine hoisting equipment wireless telecommunications video surveillance security early warning system |
| CN112483076A (en) * | 2019-09-11 | 2021-03-12 | 中国石油化工股份有限公司 | System for be used for discerning drilling construction complex situation |
| CN111341334A (en) * | 2020-03-06 | 2020-06-26 | 东莞理工学院 | Noise reduction and abnormal sound detection system and method applied to rail transit |
| CN112164408A (en) * | 2020-10-26 | 2021-01-01 | 南京农业大学 | Pig coughing sound monitoring and early warning system based on deep learning |
| CN112472065A (en) * | 2020-11-18 | 2021-03-12 | 天机医用机器人技术(清远)有限公司 | Disease detection method based on cough sound recognition and related equipment thereof |
| CN112820275A (en) * | 2021-01-15 | 2021-05-18 | 华中农业大学 | Automatic monitoring method for analyzing abnormality of suckling piglets based on sound signals |
| CN113553724A (en) * | 2021-07-30 | 2021-10-26 | 中国石油化工股份有限公司 | Shale gas well production process pit shaft abnormity analysis and processing system |
| CN115048984A (en) * | 2022-05-17 | 2022-09-13 | 天津农学院 | Sow oestrus recognition method based on deep learning |
Non-Patent Citations (5)
| Title |
|---|
| 刘均荣;史伟新;李博宇;周峰;: "分布式光纤声音传感技术在油田中的应用及发展前景", 地质科技情报, no. 05, 15 September 2017 (2017-09-15), pages 262 - 266 * |
| 机械工业技术发展基金会等: "国家工业节能技术应用指南", vol. 1, 30 April 2022, 机械工业出版社, pages: 96 * |
| 肖强宏;周强;王莹;薛敏涛;高乐乐;: "一种基于敲击声时-频分析的陶瓷结构缺陷检测方法研究", 中国陶瓷, no. 09, 5 September 2017 (2017-09-05) * |
| 胡涛等: "卷积神经网络在异常声音识别中的研究", 信号处理, vol. 34, no. 3, 25 March 2018 (2018-03-25), pages 357 - 367 * |
| 陈华华;徐萍;: "基于谱减法和短时能量的声音端点检测方法", 机械制造与自动化, no. 03, 20 June 2016 (2016-06-20) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117454294A (en) * | 2023-11-02 | 2024-01-26 | 西南石油大学 | Shale gas well plunger process well abnormal condition judging method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111488694B (en) | Slurry shield tunneling key parameter prediction method based on multiple linear autoregressive | |
| CN107305601B (en) | Method for analyzing efficiency factors of oil pumping well system | |
| CN103996077B (en) | A kind of electrical equipment malfunction Forecasting Methodology based on multidimensional time-series | |
| CN113723838A (en) | While-drilling safety risk intelligent identification method based on convolutional neural network | |
| CN116181287A (en) | Shale gas well production abnormal condition early warning system and method | |
| CN112036086A (en) | Dynamic risk early warning system for gas pipeline | |
| CN109405924A (en) | A kind of online remote supervision system of drainage pipeline networks liquid level and method | |
| CN206220950U (en) | Afflux drawing type producing profile testing tubing string in a kind of oil pipe of horizontal well | |
| CN111507155A (en) | U-Net + + and UDA combined microseism effective signal first-arrival pickup method and device | |
| CN110288257A (en) | A kind of depth transfinites indicator card learning method | |
| CN115874993A (en) | Shale gas well production control method, equipment and system based on artificial intelligence | |
| CN113011089A (en) | Method and device for optimizing efficiency of mechanical production well system based on deep learning | |
| CN115860197A (en) | Data-driven coal bed gas yield prediction method and system | |
| CN116291271A (en) | Rock-soil body parameter inversion method and system based on while-drilling test | |
| CN114661006A (en) | Optimized control system and method for cabin pumping and bank blowing process of trailing suction hopper dredger | |
| CN114234061A (en) | Neural network-based intelligent judgment method for water leakage sound of pressurized operation water supply pipeline | |
| CN114021620A (en) | Electrical submersible pump fault diagnosis method based on BP neural network feature extraction | |
| CN112836856A (en) | Polynomial regression algorithm-based pumping unit water drive well pump detection period prediction method | |
| CN111460737A (en) | Intelligent settlement prediction method and system for slurry air pressure balance shield | |
| CN115729910A (en) | Implementation method and system for analyzing drilling speed of drilling machine in real time | |
| CN109657404B (en) | Automatic fault diagnosis system for coal mining machine based on chaos correction group intelligent optimization | |
| CN114033377A (en) | Top coal caving control method and system based on voiceprint detection technology | |
| CN112727433A (en) | Drilling parameter optimization method | |
| CN113627640A (en) | Productivity well testing prediction method and system for fracture-cavity type oil reservoir oil and gas well | |
| CN112926839A (en) | Overflow risk cooperative monitoring method and system for oil and gas well drilling process |
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 |