CN117605469A - Wellhead content intelligent metering device and method - Google Patents
Wellhead content intelligent metering device and method Download PDFInfo
- Publication number
- CN117605469A CN117605469A CN202311628698.7A CN202311628698A CN117605469A CN 117605469 A CN117605469 A CN 117605469A CN 202311628698 A CN202311628698 A CN 202311628698A CN 117605469 A CN117605469 A CN 117605469A
- Authority
- CN
- China
- Prior art keywords
- frequency
- wellhead
- oil
- signal
- signals
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 239000003129 oil well Substances 0.000 claims abstract description 37
- 239000007788 liquid Substances 0.000 claims abstract description 27
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 26
- 238000005259 measurement Methods 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 15
- 230000005284 excitation Effects 0.000 claims description 12
- 239000012530 fluid Substances 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 238000012795 verification Methods 0.000 claims description 12
- 230000005684 electric field Effects 0.000 claims description 9
- 230000006698 induction Effects 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000004220 aggregation Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 4
- 238000003860 storage Methods 0.000 claims description 4
- 239000002699 waste material Substances 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000002156 mixing Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000003921 oil Substances 0.000 description 31
- 238000013136 deep learning model Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000004931 aggregating effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005514 two-phase flow Effects 0.000 description 2
- 230000033558 biomineral tissue development Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 239000010779 crude oil Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 235000020681 well water Nutrition 0.000 description 1
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B49/00—Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
- E21B49/08—Obtaining fluid samples or testing fluids, in boreholes or wells
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N22/00—Investigating or analysing materials by the use of microwaves or radio waves, i.e. electromagnetic waves with a wavelength of one millimetre or more
- G01N22/04—Investigating moisture content
-
- 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
- G06F18/211—Selection of the most significant subset of features
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Geochemistry & Mineralogy (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Electromagnetism (AREA)
- Fluid Mechanics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Measuring Volume Flow (AREA)
Abstract
The invention relates to an intelligent metering device and method for wellhead yield, which are used for measuring fluctuation signals of oil well produced liquid at a wellhead based on a high-frequency microwave sensor measuring device, constructing a time-frequency linkage network to extract signal multidimensional characteristics and realizing accurate measurement of water content parameters of the oil well wellhead produced liquid.
Description
Technical Field
The invention belongs to the technical field of crude oil production, and particularly relates to an intelligent metering device and method for wellhead inclusion rate.
Background
The water content measurement of the oil well produced fluid is very important in oil field production, and accurate measurement is helpful for optimizing the production process, increasing yield and improving efficiency. Traditional oil well produced fluid water content measurement technology comprises a chemical analysis and physical measurement method of oil well produced fluid, but offline sample collection and laboratory analysis are required, and real-time monitoring cannot be achieved. With the continuous and intensive research, more emerging technologies are applied to oil well water content measurement, such as microwave and radio frequency technologies, which can be used for non-invasively measuring wellhead water content without sampling and real-time metering.
In recent years, deep learning models are increasingly widely used due to their powerful data processing capability, and based on the deep learning models, we can mine complex and subtle patterns and information from sensor data. The deep learning models also exhibit excellent performance in terms of oil-water two-phase flow water cut measurement, and they can identify and utilize multi-level data features with high automation, thereby realizing more accurate and real-time water cut measurement than conventional methods.
Disclosure of Invention
The invention aims to provide an intelligent metering device and method for wellhead yield, which are used for measuring fluctuation signals of oil well produced liquid at a wellhead based on a high-frequency microwave sensor measuring device, constructing a time-frequency linkage network to extract signal multidimensional characteristics and realizing accurate measurement of water content parameters of the oil well wellhead produced liquid.
The technical scheme adopted by the invention is as follows: an intelligent metering device and method for wellhead content comprises the following steps:
(1) And constructing a high-frequency microwave sensor measuring device for measuring the fluctuation signal of the oil well produced fluid, and installing the measuring device on the horizontal section of the oil well wellhead conveying pipeline.
The high-frequency microwave sensor measuring device comprises an oil inlet inner pipe, a double-helix high-frequency microwave sensor, an oil outlet pipe and an oil return outer pipe, wherein the double-helix high-frequency microwave sensor consists of a measuring pipeline, an exciting electrode, an induction electrode, a protection electrode, a shielding layer and a sensor shell, the exciting electrode and the induction electrode always keep an electrode-to-wall structure when rotating at 360 degrees of the outer wall of the measuring pipeline, two protection electrodes which are synchronous for 360 degrees are arranged between the exciting electrode and a receiving electrode, and a shielding layer is arranged outside the electrodes to prevent an electric field from diffusing to the edge, so that the effects of protecting the electric field and enhancing the intensity of the electric field inside the pipeline are achieved;
the measuring device is arranged on the horizontal section of the oil well produced fluid conveying pipeline, the oil inlet inner pipe is fixedly connected with the sensor shell through the flange plate, the sensor shell is fixedly connected with the oil outlet pipe through the flange plate, the oil outlet pipe is fixedly connected with the oil return outer pipe, the oil return outer pipe is arranged on the outer side of the oil inlet inner pipe, and the oil inlet inner pipe and the oil outlet inner pipe form a sleeve.
(2) And measuring a fluctuation signal of the wellhead produced liquid by adopting a high-frequency microwave sensor measuring device, converting the fluctuation signal into a microwave difference frequency signal, and transmitting the signal to an upper computer for subsequent processing. The method specifically comprises the following steps: in the measuring process, the oil well produced liquid enters from the oil inlet inner pipe to pass through the measuring pipeline, a sine excitation signal source generates high-frequency excitation signals which are respectively sent to the excitation electrode and an external monitor, the excitation electrode emits high-frequency electromagnetic waves into the oil well produced liquid in the pipeline, the absorbed wave energy is different due to the fact that the content of polar water molecules in the oil well produced liquid is different, the induction electrode sends different signals to the monitor due to the fact that the content of the polar water molecules is different, the two paths of signals are subjected to signal mixing at the monitor, one path of microwave difference frequency signals are obtained after the signals are processed by the adder, the microwave difference frequency signals are transmitted to the upper computer for subsequent processing, and the oil well produced liquid continues to flow through the oil outlet pipe and the oil return outer pipe to return to the oil well produced liquid conveying pipeline for transportation and storage, so that waste is avoided.
(3) The method comprises the steps of constructing a data set, specifically preprocessing a microwave difference frequency signal, then intercepting samples from the data by using a sliding window, adding corresponding labels for the samples, taking 80% of the samples as a training set, 10% of the samples as a verification set, and taking the remaining 10% of the samples as a test set. The method specifically comprises the following steps:
the microwave difference frequency signal is preprocessed, and the formula is as follows:
wherein I is o Represents the o data, I mean And I std The mean and standard deviation of the data are respectively,is the o data after pretreatment; dividing the preprocessed microwave difference frequency signal by adopting a sliding window method, and acquiring data with the window length of H and the length of P>Samples of (a) wherein->Representative is rounded down, the data label is tested by the water content test value, and the data label is sharedObtaining N samples with tag values; randomly dividing N fluctuation samples into data sets with training sets, verification sets and test sets, wherein the specific proportion is [ training sets: verification sets: test sets ]]=[8∶1∶1]。
(4) Constructing a time-frequency linkage network, firstly performing fast Fourier transform on the obtained sample signals to obtain frequency distribution, dividing and stacking one-dimensional sample signals into two-dimensional signals based on the most obvious period, extracting time-frequency linkage characteristics of the signals, and using the time-frequency linkage characteristics to realize accurate measurement of the water content of wellhead produced fluid. The method specifically comprises the following steps: the time-frequency linkage network comprises a splitting and stacking module, a characteristic extracting and aggregating module and a predicting module.
The splitting and stacking module firstly performs fast Fourier transform on an input signal to obtain the strongest n frequency components and corresponding periods, and the calculation formula is as follows:
Q=Amp(FFT(X))
f 1 ,f 2 ,...f n =argTOPn(Q)
where X is the one-dimensional input signal, FFT () is the fast fourier transform, amp () is the amplitude calculation, and argTOPn is the first n frequency components with the largest amplitude. Next, we split-stack the original signal based on the most significant n periods, whose calculation formula is:
where Padding (-) is to expand X with 0 to make it cycle compatible, re is split stack operation, P i ,f i Representing the number of rows and columns after transformation.
The feature extraction and aggregation module firstly performs feature extraction on the sequence output by the splitting and stacking module, and convolutionally extracts time-frequency linkage features on the sequence by DenseNet, wherein the calculation formula is as follows:
and then the obtainedThe sequence is converted into a one-dimensional sequence, and the calculation formula is as follows:
where trunk is the operation of truncating the generated sequence to the original one-dimensional input signal length.
Then n are based on different periodsAggregation is carried out, and the calculation formula is as follows:
the prediction module uses a Sigmoid activation function to realize the numerical measurement of the water content.
The super-parameters of the time-frequency linkage network are that AMSGrad optimization algorithm is used, the weights of the time-frequency linkage network are updated reversely through gradients based on a training set and a verification set, mean Square Error (MSE), mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used as loss functions, the difference between the output of the time-frequency linkage network and a real label is measured, and the network training and optimization are guided to be carried out in the correct direction by taking the minimized difference as a criterion.
Due to the adoption of the technical scheme, the invention has the following advantages:
the invention provides an intelligent metering device and method for wellhead yield, which are used for collecting signals of oil well produced liquid based on a double-helix high-frequency microwave sensor, wherein the sensor adopts a non-invasive structure, and is used for measuring flow information in real time based on high-frequency microwave signals, so that the sensitivity is high and the influence of mineralization is small. And the produced liquid of the oil well flows through the oil outlet pipe and the oil return outer pipe for transportation and storage after the content rate of the produced liquid of the oil well is measured by the measuring system, so that the resource waste is avoided.
The method comprises the steps of constructing a time-frequency linkage network, firstly carrying out fast Fourier transform on an obtained one-dimensional microwave difference frequency signal to obtain frequency distribution, dividing and stacking the one-dimensional signal into two-dimensional signals based on the most obvious period, simultaneously focusing two time sequence change characteristics in the time sequence period and between periods, and carrying out characteristic extraction on the obtained sensor signal by combining time-frequency domain information to realize accurate measurement of the water content of the oil-water two-phase flow.
Drawings
FIG. 1 is a schematic diagram of a high frequency microwave sensor and its measurement system;
fig. 2 is a diagram of a time-frequency linked network framework.
Detailed Description
The present invention will be described in detail with reference to the following examples and the accompanying drawings.
According to the wellhead yield intelligent metering device and method, the fluctuation signal of the oil well produced liquid at the wellhead is measured based on the high-frequency microwave sensor measuring device, the time-frequency linkage network is constructed to extract the signal multidimensional characteristic, and the accurate measurement of the water content parameter of the oil well wellhead produced liquid is realized.
(1) A high-frequency microwave sensor measuring device for measuring the fluctuation signal of the oil well produced fluid is constructed as shown in fig. 1, and the measuring device is arranged on the horizontal section of the oil well wellhead conveying pipeline.
The high-frequency microwave sensor measuring device comprises an oil inlet inner pipe, a double-helix high-frequency microwave sensor, an oil outlet pipe and an oil return outer pipe, wherein the double-helix high-frequency microwave sensor consists of a measuring pipeline, an exciting electrode, an induction electrode, a protection electrode, a shielding layer and a sensor shell, the exciting electrode and the induction electrode always keep an electrode-to-wall structure when rotating at 360 degrees of the outer wall of the measuring pipeline, two protection electrodes which are synchronous for 360 degrees are arranged between the exciting electrode and a receiving electrode, and a shielding layer is arranged outside the electrodes to prevent an electric field from diffusing to the edge, so that the effects of protecting the electric field and enhancing the intensity of the electric field inside the pipeline are achieved;
the measuring device is arranged on the horizontal section of the oil well produced fluid conveying pipeline, the oil inlet inner pipe is fixedly connected with the sensor shell through the flange plate, the sensor shell is fixedly connected with the oil outlet pipe through the flange plate, the oil outlet pipe is fixedly connected with the oil return outer pipe, the oil return outer pipe is arranged on the outer side of the oil inlet inner pipe, and the oil inlet inner pipe and the oil outlet inner pipe form a sleeve.
(2) And measuring a fluctuation signal of the wellhead produced liquid by adopting a high-frequency microwave sensor measuring device, converting the fluctuation signal into a microwave difference frequency signal, and transmitting the signal to an upper computer for subsequent processing. The method specifically comprises the following steps: in the measuring process, the oil well produced liquid enters from the oil inlet inner pipe to pass through the measuring pipeline, a sine excitation signal source generates high-frequency excitation signals which are respectively sent to the excitation electrode and an external monitor, the excitation electrode emits high-frequency electromagnetic waves into the oil well produced liquid in the pipeline, the absorbed wave energy is different due to the fact that the content of polar water molecules in the oil well produced liquid is different, the induction electrode sends different signals to the monitor due to the fact that the content of the polar water molecules is different, the two paths of signals are subjected to signal mixing at the monitor, one path of microwave difference frequency signals are obtained after the signals are processed by the adder, the microwave difference frequency signals are transmitted to the upper computer for subsequent processing, and the oil well produced liquid continues to flow through the oil outlet pipe and the oil return outer pipe to return to the oil well produced liquid conveying pipeline for transportation and storage, so that waste is avoided.
(3) The method comprises the steps of constructing a data set, specifically preprocessing a microwave difference frequency signal, then intercepting samples from the data by using a sliding window, adding corresponding labels for the samples, taking 80% of the samples as a training set, 10% of the samples as a verification set, and taking the remaining 10% of the samples as a test set. The method specifically comprises the following steps:
the microwave difference frequency signal is preprocessed, and the formula is as follows:
wherein I is o Represents the o data, I mean And I std The mean and standard deviation of the data are respectively,is the o data after pretreatment; dividing the preprocessed microwave difference frequency signal by adopting a sliding window method, and acquiring data with the window length of H and the length of P>Samples of (a) wherein->The representative is rounded downwards, the water content test value is used for data labels, and N samples with label values are obtained; randomly dividing N fluctuation samples into data sets with training sets, verification sets and test sets, wherein the specific proportion is [ training sets: verification sets: test sets ]]=[8∶1∶1]。
(4) Constructing a time-frequency linkage network shown in fig. 2, performing fast Fourier transform on the obtained sample signals to obtain frequency distribution, dividing and stacking one-dimensional sample signals into two-dimensional signals based on the most remarkable period, and extracting time-frequency linkage characteristics of the signals for realizing accurate measurement of water content of wellhead produced fluid. The method specifically comprises the following steps: the time-frequency linkage network comprises a splitting and stacking module, a characteristic extracting and aggregating module and a predicting module.
The splitting and stacking module firstly performs fast Fourier transform on an input signal to obtain the strongest n frequency components and corresponding periods, and the calculation formula is as follows:
Q=Amp(FFT(X))
f 1 ,f 2 ,...f n =argTOPn(Q)
where X is the one-dimensional input signal, FFT () is the fast fourier transform, amp () is the amplitude calculation, and argTOPn is the first n frequency components with the largest amplitude. Next, we split-stack the original signal based on the most significant n periods, whose calculation formula is:
where Padding (-) is to expand X with 0 to make it cycle compatible, re is split stack operation, P i ,f i Representing the number of rows and columns after transformation.
The feature extraction and aggregation module firstly performs feature extraction on the sequence output by the splitting and stacking module, and convolutionally extracts time-frequency linkage features on the sequence by DenseNet, wherein the calculation formula is as follows:
and then the obtainedThe sequence is converted into a one-dimensional sequence, and the calculation formula is as follows:
where trunk is the operation of truncating the generated sequence to the original one-dimensional input signal length.
Then n are based on different periodsAggregation is carried out, and the calculation formula is as follows:
the prediction module uses a Sigmoid activation function to realize the numerical measurement of the water content.
The super-parameters of the time-frequency linkage network are that AMSGrad optimization algorithm is used, the weights of the time-frequency linkage network are updated reversely through gradients based on a training set and a verification set, mean Square Error (MSE), mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used as loss functions, the difference between the output of the time-frequency linkage network and a real label is measured, and the network training and optimization are guided to be carried out in the correct direction by taking the minimized difference as a criterion.
Claims (6)
1. An intelligent metering device and method for wellhead content is characterized by comprising the following steps:
(1) Constructing a high-frequency microwave sensor measuring device for measuring an oil well produced fluid fluctuation signal, and installing the measuring device on the horizontal section of an oil well wellhead conveying pipeline;
(2) Measuring a fluctuation signal of the wellhead produced liquid by adopting a high-frequency microwave sensor measuring device, converting the fluctuation signal into a microwave difference frequency signal, and transmitting the signal to an upper computer for subsequent processing;
(3) Constructing a data set, namely preprocessing a microwave difference frequency signal, intercepting samples from the data by using a sliding window, adding corresponding labels for the samples, taking 80% of the samples as a training set, 10% of the samples as a verification set, and taking the remaining 10% of the samples as a test set;
(4) Constructing a time-frequency linkage network, firstly performing fast Fourier transform on the obtained sample signals to obtain frequency distribution, dividing and stacking one-dimensional sample signals into two-dimensional signals based on the most obvious period, extracting time-frequency linkage characteristics of the signals, and using the time-frequency linkage characteristics to realize accurate measurement of the water content of wellhead produced fluid.
2. The wellhead content intelligent metering device and method according to claim 1, characterized in that the high-frequency microwave sensor measuring device in step (1) comprises an oil inlet inner pipe, a double-helix high-frequency microwave sensor, an oil outlet pipe and an oil return outer pipe, wherein the double-helix high-frequency microwave sensor consists of a measuring pipeline, an exciting electrode, an induction electrode, a protection electrode, a shielding layer and a sensor shell, the exciting electrode and the induction electrode always keep an electrode-to-wall structure when rotating for 360 degrees, two protection electrodes which are synchronous for 360 degrees are arranged between the exciting electrode and a receiving electrode, and a shielding layer is arranged outside the electrodes to prevent an electric field from diffusing to the edge, so that the effects of protecting the electric field and enhancing the intensity of the electric field inside the pipeline are achieved;
the measuring device is arranged on the horizontal section of the oil well produced fluid conveying pipeline, the oil inlet inner pipe is fixedly connected with the sensor shell through the flange plate, the sensor shell is fixedly connected with the oil outlet pipe through the flange plate, the oil outlet pipe is fixedly connected with the oil return outer pipe, the oil return outer pipe is arranged on the outer side of the oil inlet inner pipe, and the oil inlet inner pipe and the oil outlet inner pipe form a sleeve.
3. The wellhead content intelligent metering device and method according to claim 1, wherein step (2) comprises: in the measuring process, the oil well produced liquid enters from the oil inlet inner pipe to pass through the measuring pipeline, a sine excitation signal source generates high-frequency excitation signals which are respectively sent to the excitation electrode and an external monitor, the excitation electrode emits high-frequency electromagnetic waves into the oil well produced liquid in the pipeline, the absorbed wave energy is different due to the fact that the content of polar water molecules in the oil well produced liquid is different, the induction electrode sends different signals to the monitor due to the fact that the content of the polar water molecules is different, the two paths of signals are subjected to signal mixing at the monitor, one path of microwave difference frequency signals are obtained after the signals are processed by the adder, the microwave difference frequency signals are transmitted to the upper computer for subsequent processing, and the oil well produced liquid continues to flow through the oil outlet pipe and the oil return outer pipe to return to the oil well produced liquid conveying pipeline for transportation and storage, so that waste is avoided.
4. The wellhead content intelligent metering device and method according to claim 1, wherein step (3) comprises: the microwave difference frequency signal is preprocessed, and the formula is as follows:
wherein I is o Represents the o data, I mean And I std The mean and standard deviation of the data are respectively,is the o data after pretreatment; dividing the preprocessed microwave difference frequency signal by adopting a sliding window method, and acquiring data with the window length of H and the length of P>Samples of (a) wherein->The representative is rounded downwards, the water content test value is used for data labels, and N samples with label values are obtained; the N fluctuation samples are randomly divided into data sets with training sets, verification sets and test sets, and the specific proportion is [ training set: verification set: test set]=[8:1:1]。
5. The wellhead content intelligent metering device and method according to claim 1, wherein step (4) comprises: the time-frequency linkage network comprises a splitting and stacking module, a feature extraction and aggregation module and a prediction module;
the splitting and stacking module firstly performs fast Fourier transform on an input signal to obtain the strongest n frequency components and corresponding periods, and the calculation formula is as follows:
Q=Amp(FFT(X))
f 1 ,f 2 ,...f n =argTOPn(Q)
wherein X is a one-dimensional input signal, FFT is fast Fourier transform, amp is amplitude calculation, argTOPN is the first n frequency components with the largest amplitude, then the original signal is split and stacked based on the most obvious n periods, and the calculation formula is as follows:
where Padding (-) is to expand X with 0 to make it cycle compatible, re is split stack operation, P i ,f i Representing the number of rows and columns after transformation;
the feature extraction and aggregation module firstly performs feature extraction on the sequence output by the splitting and stacking module, and convolutionally extracts time-frequency linkage features on the sequence by DenseNet, wherein the calculation formula is as follows:
and then the obtainedThe sequence is converted into a one-dimensional sequence, and the calculation formula is as follows:
wherein trunk is an operation of truncating the generated sequence to the original one-dimensional input signal length, followed by n-th based on different periodsAggregation is carried out, and the calculation formula is as follows:
the prediction module uses a Sigmoid activation function to realize the numerical measurement of the water content.
6. The intelligent wellhead content metering device and method according to claim 5, wherein the super parameter of the time-frequency linkage network is that the weight of the time-frequency linkage network is updated reversely by gradient based on a training set and a verification set by using AMSGrad optimization algorithm, mean Square Error (MSE), mean Absolute Error (MAE) and Mean Absolute Percent Error (MAPE) are used as loss functions, and the gap between the output of the time-frequency linkage network and the real labels is measured by using the minimum gap as a criterion to guide the training and optimization of the network to the correct direction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311628698.7A CN117605469A (en) | 2023-12-01 | 2023-12-01 | Wellhead content intelligent metering device and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311628698.7A CN117605469A (en) | 2023-12-01 | 2023-12-01 | Wellhead content intelligent metering device and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117605469A true CN117605469A (en) | 2024-02-27 |
Family
ID=89949567
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311628698.7A Pending CN117605469A (en) | 2023-12-01 | 2023-12-01 | Wellhead content intelligent metering device and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117605469A (en) |
-
2023
- 2023-12-01 CN CN202311628698.7A patent/CN117605469A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103257182B (en) | A kind of impulse eddy current defect quantitative detection method and detection system | |
CN110630256B (en) | Low-gas-production oil well wellhead water content prediction system and method based on depth time memory network | |
CN106199474B (en) | A kind of low-field nuclear magnetic resonance two-dimensional spectrum inversion algorithm | |
CN104807589B (en) | A kind of ONLINE RECOGNITION method collecting flow pattern of gas-liquid two-phase flow in defeated-riser systems | |
Burriel-Valencia et al. | Fault diagnosis of induction machines in a transient regime using current sensors with an optimized slepian window | |
CN112215405A (en) | Non-invasive type residential electricity load decomposition method based on DANN domain adaptive learning | |
CN113378680B (en) | Intelligent database building method for Raman spectrum data | |
CN104217112A (en) | Multi-type signal-based power system low-frequency oscillation analysis method | |
CN111428585B (en) | Metamaterial terahertz spectroscopy identification method based on deep learning | |
Xie et al. | Sliding-window based scale-frequency map for bird sound classification using 2D-and 3D-CNN | |
CN110632132A (en) | High-yield gas-oil well wellhead liquid water content prediction method based on multi-sensor measurement and deep convolutional neural network | |
CN110243885A (en) | A kind of low yield gas well mouth of oil well hydrated comples ion method based on time-frequency characteristics | |
CN117605469A (en) | Wellhead content intelligent metering device and method | |
CN110243886A (en) | A kind of low yield gas well mouth of oil well hydrated comples ion method based on nonlinear characteristic | |
CN110630244B (en) | High-yield gas-oil well water content prediction system and method based on multi-sensor measurement and long-time and short-time memory network | |
CN108549069A (en) | A kind of warping under imperfect waveguid converts modification method | |
Liu et al. | Intelligent defect identification based on pect signals and an optimized two-dimensional deep convolutional network | |
Gao et al. | Feature extraction method of ball mill load based on adaptive variational mode decomposition and improved power spectrum analysis | |
CN114492146A (en) | Bolt group loosening positioning and quantitative analysis method and system based on transfer learning | |
CN104633461A (en) | PCCP pipe failure forewarning and monitoring signal processing method based on sound signals | |
CN109187498A (en) | The rapid detection method of soil calcium constituent based on conllinear double-pulse laser induced breakdown spectrum | |
Yang et al. | Research on fault identification method based on multi-resolution permutation entropy and ABC-SVM | |
CN103577877A (en) | Ship motion prediction method based on time-frequency analysis and BP neural network | |
CN112788057A (en) | LDoS attack detection method based on FSWT time-frequency distribution | |
Shan et al. | γ-polyglutamic acid fermentation monitoring with ATR-FTIR spectroscopy based on a shallow convolutional neural network combined with data augmentation and attention module |
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 |