CN117665009A - Oil-water two-phase flow parameter measurement method of double-flow differential network - Google Patents
Oil-water two-phase flow parameter measurement method of double-flow differential network Download PDFInfo
- Publication number
- CN117665009A CN117665009A CN202311639487.3A CN202311639487A CN117665009A CN 117665009 A CN117665009 A CN 117665009A CN 202311639487 A CN202311639487 A CN 202311639487A CN 117665009 A CN117665009 A CN 117665009A
- Authority
- CN
- China
- Prior art keywords
- oil
- water
- flow
- network
- phase flow
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 64
- 230000005514 two-phase flow Effects 0.000 title claims abstract description 61
- 238000000691 measurement method Methods 0.000 title description 3
- 238000000034 method Methods 0.000 claims abstract description 43
- 238000005259 measurement Methods 0.000 claims description 20
- 230000004927 fusion Effects 0.000 claims description 14
- 239000000463 material Substances 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 230000006698 induction Effects 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 230000009977 dual effect Effects 0.000 claims description 6
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 5
- 229910052802 copper Inorganic materials 0.000 claims description 5
- 239000010949 copper Substances 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims description 3
- 230000007935 neutral effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims 1
- 238000013135 deep learning Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 239000004696 Poly ether ether ketone Substances 0.000 description 2
- JUPQTSLXMOCDHR-UHFFFAOYSA-N benzene-1,4-diol;bis(4-fluorophenyl)methanone Chemical compound OC1=CC=C(O)C=C1.C1=CC(F)=CC=C1C(=O)C1=CC=C(F)C=C1 JUPQTSLXMOCDHR-UHFFFAOYSA-N 0.000 description 2
- 239000007772 electrode material Substances 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 229920002530 polyetherether ketone Polymers 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000036632 reaction speed Effects 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Landscapes
- Measuring Volume Flow (AREA)
Abstract
The invention relates to a measuring method for measuring parameters of an oil-water two-phase flow, in particular to a measuring method for measuring parameters of an oil-water two-phase flow based on a double-flow differential network, which consists of a high-precision oil-water two-phase flow sensor and a soft measuring model. The designed oil-water two-phase flow sensor is used for acquiring parameter fluctuation signals in the flowing process of the oil-water two-phase flow, then a structured deep learning network is adopted to extract and analyze the fluctuation signals acquired by the sensor, and finally corrected flowing parameters of the oil-water two-phase flow can be obtained.
Description
Technical Field
A measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a double-flow differential network.
Technical Field
The oil-water two-phase flow refers to the phenomenon that oil and water flow simultaneously in the same pipeline, and is widely applied to industries such as petroleum, chemical industry, electric power and the like. The accurate measurement of the oil-water two-phase flow parameter has important significance for controlling and optimizing the production process. The traditional oil-water two-phase flow measuring method mainly comprises a venturi flowmeter, a turbine flowmeter and the like, but the problems of low measuring precision, easiness in being influenced by fluid properties and the like exist in the methods. Therefore, the development of the oil-water two-phase flow parameter measuring method with high precision and strong anti-interference capability has important practical significance.
The deep learning soft measurement method widens the application range of the multiphase flow soft measurement model. The deep learning technology is a new theory which is raised in recent years, the characteristics of the tested object are extracted layer by layer in an unsupervised mode or a supervised mode, the objectivity of the characteristics is strong, and the essence of the tested object can be accurately and fully reflected. The data measured by the sensor can be fused through a soft measurement means, and the characteristic information of the multiphase flow can be extracted accurately and efficiently by adopting an intelligent and deep learning method.
Disclosure of Invention
A measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a double-flow differential network.
The spiral electrode sensor is used for measuring the oil-water two-phase flow water content, and then the flow information characteristics in the measurement signals are extracted based on the double-flow differential network and are subjected to characteristic fusion, so that the accurate measurement of the oil-water two-phase flow content is realized.
The technical scheme adopted for solving the technical problems is as follows:
1. a measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a double-flow differential network. The method is characterized by comprising the following steps of:
(1) Measuring fluctuation signals of oil-water two-phase flow parameters by adopting a spiral electrode sensor;
(2) And constructing a double-flow differential network model for realizing characteristic fusion and content measurement of the oil-water two-phase flow parameter fluctuation signal.
2. A measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a double-flow differential network. The method is characterized in that the step (1) comprises the following steps:
when materials enter the measuring pipe section from the pipeline, the spiral electrode sensor arranged on the pipeline is used for measuring oil-water two-phase flow parameter fluctuation signals. The designed electrostatic sensor consists of a spiral electrode, an insulating layer and a material pipeline. Firstly, a spiral induction electrode is designed according to the working principle of a microwave sensor, the spiral induction electrode is positioned in the middle of a pipeline, then the spiral induction electrode on one side is excited by a high-frequency excitation source, and the other side receives signals by an AD8302 module. Meanwhile, the whole sensor is covered with a shielding layer made of copper sheets. The measuring signal of the oil-water two-phase flow can be effectively collected through the spiral electrode sensor. The sensor can capture mass flow information of the oil-water two-phase flow from a microscopic angle by utilizing an electrostatic induction principle.
In the actual material flowing process, due to factors such as gas pressure, material pressure and the like, the material pipeline material is selected to be PEEK so as to meet the pressure-resistant requirement, meanwhile, in order to meet the requirement of large-flow flowing, the inner diameter of the pipeline is set to be DN50, and the electrode material is selected to be copper. The model diagram is shown in figure 1. The length of the two electrodes is 7cm and the height is 15cm. Experiments show that the sensor can effectively detect dynamic changes flowing through the spiral electrode, and accurate measurement of dynamic parameters of the oil-water two-phase flow is achieved.
3. A measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a depth self-adaptive network.
The method is characterized in that the step (2) comprises the following steps:
(1) the constructed double-flow differential network model is characterized in that firstly, oil-water two-phase flow signals measured by a spiral electrode sensor are preprocessed. Dividing the fluctuation signal by adopting a non-overlapping window mode, wherein the window length is H, and the sample signal with the length of L can be obtainedSamples of (a) wherein->Representing a rounding down. And windowing and intercepting the fluctuation signal acquired by the sensor, and obtaining N parameter fluctuation samples with label values by using the actual value as a data label. And carrying out discrete wavelet decomposition on the signals subjected to the sliding window to obtain N preprocessing signals.
(2) The data set is divided by 2N samples after the original signal and the discrete wavelet are decomposed at random, and the specific proportion is [ training set: verification set: test set ] = [8:1:1].
4. A measuring method for oil-water two-phase flow parameter measurement, in particular to an oil-water two-phase flow parameter measuring method based on a depth self-adaptive network.
The method is characterized in that the step (2) comprises the following steps:
(1) the constructed double-flow differential network model is characterized in that the model comprises a double-flow self-coding network and a multi-scale differential fusion network. A dual stream self-encoding network includes an original data stream and a discrete wavelet data stream. The coding features are extracted using a self-coding network. The multi-scale differential fusion network decouples characteristic attributes under different working conditions by fusing multi-scale differential characteristics of different levels. Meanwhile, the multi-layer perceptron model fuses the characteristic difference of the multi-scale intermediate layer to obtain a final content prediction result. The structure is shown in figure 2.
(2) The built double-flow differential network model is characterized in that the self-coding model adopts a depth residual neutral network and mainly comprises five convolution modules, signals are coded to high-order characteristic representation, and then the high-dimensional characteristic is sampled to the corresponding characteristic size in the encoder model through a decoder formed by the deconvolution modules. Wherein the five convolution modules are all 3 x 3 in size and the five deconvolution modules are 3 x 3 in size. Wherein the self-encoding process is
Wherein f (x) is an encoderG (h) is the number of bits in the decoder,is a reconstructed signal by a de-encoding function. In the self-coding model pre-training process, signals are normalized and input to an encoder model f (x), high-dimensional characteristic representations of the signals are extracted, and then reconstructed signals are obtained through a decoder model g (h). Model training the model is iterated by means of a Mean Square Error (MSE) loss function, the loss function formula is
Wherein x is i,j Andfor the values of the original signal and the generated signal, n and m are the dimensions of the signal. After the pre-training model training is completed, the encoder is used to extract a high-dimensional feature representation of the signal. Firstly, inputting signals into an encoder model to construct an original signal flow network, obtaining high-dimensional characteristic distribution through an encoder, and inputting wavelet decomposition signals into the encoder model to construct a wavelet decomposition network to obtain mixed characteristic distribution.
(3) The constructed double-flow differential network model is characterized in that the network structure of the double-flow differential network model is characterized in that the multi-scale differential fusion network mainly comprises seven local differential networks. The differential characteristics of the input layer and the five intermediate layers and the coding layers in the dual-stream encoder network are input to the corresponding local differential network model. And finally, splicing the results of each differential network, passing through a multi-layer perceptron module, and outputting a prediction result. The difference characteristic calculation method of each scale is shown as follows
y＝f mlp (z d )
Wherein y is the predicted result,and->For the ith convolutional layer feature in dual stream, < +.>As a difference feature of the i-th layer,representing the difference characteristics of the original signal, < >>For the difference of the coding features, the dimension of each difference feature corresponds to the dimension of the convolution layer, and the range of the parameter θ is [0,1]The influence of the Euclidean distance and the absolute value distance on the differential feature is determined by adjusting the change of the theta parameter.
5. An intelligent metering method for oil-water two-phase flow, in particular to an oil-water two-phase flow metering method based on a double-flow differential network model, which has the advantages that:
(1) The spiral electrode sensor has high reaction speed, can perform real-time monitoring and dynamic processing, and can rapidly and accurately obtain the content sequence fluctuation signal.
(2) The double-flow differential network adopted by the invention has objectivity in predicting the content, and can have stronger prediction performance on oil-water two-phase flow data.
Drawings
FIG. 1 is a schematic diagram of the structure of a spiral electrode sensor of the present invention;
fig. 2 is a dual stream differential network model of the present invention.
Detailed Description
The invention is further illustrated by the following examples: the following examples are illustrative, not limiting, and are not intended to limit the scope of the invention.
The invention relates to a method for predicting oil-water two-phase flow, which adopts a spiral electrode sensor to collect fluctuation information of oil-water two-phase flow content in a pipeline, uses the collected flow fluctuation signal as input of a double-flow differential network model to perform feature extraction and fusion, adopts a supervised learning mode, and uses actual water content as a label to predict the oil-water two-phase flow content.
The invention relates to an oil-water two-phase flow metering method, which adopts a spiral electrode sensor to measure a multi-element flow fluctuation signal in a pipeline, and the internal structure of an instrument for measuring the oil-water two-phase flow is shown in figure 1. When materials enter the measuring pipe section from the pipeline, the spiral electrode sensor arranged on the pipeline is used for measuring oil-water two-phase flow parameter fluctuation signals. The designed electrostatic sensor consists of a spiral electrode, an insulating layer and a material pipeline. Firstly, a spiral induction electrode is designed according to the working principle of a microwave sensor, the spiral induction electrode is positioned in the middle of a pipeline, then the spiral induction electrode on one side is excited by a high-frequency excitation source, and the other side receives signals by an AD8302 module. Meanwhile, the whole sensor is covered with a shielding layer made of copper sheets. The measuring signal of the oil-water two-phase flow can be effectively collected through the spiral electrode sensor. The sensor can capture mass flow information of the oil-water two-phase flow from a microscopic angle by utilizing an electrostatic induction principle.
In the actual material flowing process, due to factors such as gas pressure, material pressure and the like, the material pipeline material is selected to be PEEK so as to meet the pressure-resistant requirement, meanwhile, in order to meet the requirement of large-flow flowing, the inner diameter of the pipeline is set to be DN50, and the electrode material is selected to be copper. The model diagram is shown in figure 1. The length of the two electrodes is 7cm and the height is 15cm. Experiments show that the sensor can effectively detect dynamic changes flowing through the spiral electrode, and accurate measurement of dynamic parameters of the oil-water two-phase flow is achieved.
In the invention, the acquired flow fluctuation signal is used as the input of a deep learning model, and the flow of the measured oil-water two-phase flow can be obtained through model calculation. During the measurement, the sampling period was set to 0.001 seconds.
The invention relates to an intelligent metering method for oil-water two-phase flow, in particular to an intelligent metering method for oil-water two-phase flow based on a multi-network feature fusion model, which comprises the following steps:
(1) And collecting measurement data, preprocessing the data, and constructing a data set.
(1) The constructed double-flow differential network model is characterized in that firstly, oil-water two-phase flow signals measured by a spiral electrode sensor are preprocessed. Dividing the fluctuation signal by adopting a non-overlapping window mode, wherein the window length is H, and the sample signal with the length of L can be obtainedSamples of (a) wherein->Representing a rounding down. And windowing and intercepting the fluctuation signal acquired by the sensor, and obtaining N parameter fluctuation samples with label values by using the actual value as a data label. And carrying out discrete wavelet decomposition on the signals subjected to the sliding window to obtain N preprocessing signals.
(2) The data set is divided by 2N samples after the original signal and the discrete wavelet are decomposed at random, and the specific proportion is [ training set: verification set: test set ] = [8:1:1].
(2) Building a double-flow differential network model
(1) The constructed double-flow differential network model is characterized in that the model comprises a double-flow self-coding network and a multi-scale differential fusion network. A dual stream self-encoding network includes an original data stream and a discrete wavelet data stream. The coding features are extracted using a self-coding network. The multi-scale differential fusion network decouples characteristic attributes under different working conditions by fusing multi-scale differential characteristics of different levels. Meanwhile, the multi-layer perceptron model fuses the characteristic difference of the multi-scale intermediate layer to obtain a final content prediction result. The structure is shown in figure 2.
(2) The built double-flow differential network model is characterized in that the self-coding model adopts a depth residual neutral network and mainly comprises five convolution modules, signals are coded to high-order characteristic representation, and then the high-dimensional characteristic is sampled to the corresponding characteristic size in the encoder model through a decoder formed by the deconvolution modules. Wherein the five convolution modules are all 3 x 3 in size and the five deconvolution modules are 3 x 3 in size. Wherein the self-encoding process is
Where f (x) is the encoder, g (h) is the decoder,is a reconstructed signal by a de-encoding function. In the self-coding model pre-training process, signals are normalized and input to an encoder model f (x), high-dimensional characteristic representations of the signals are extracted, and then reconstructed signals are obtained through a decoder model g (h). Model training the model is iterated by means of a Mean Square Error (MSE) loss function, the loss function formula is
Wherein x is i,j Andfor the values of the original signal and the generated signal, n and m are the dimensions of the signal. After the pre-training model training is completed, the encoder is used to extract a high-dimensional feature representation of the signal. Firstly, inputting signals into an encoder model to construct an original signal flow network, obtaining high-dimensional characteristic distribution through an encoder, and inputting wavelet decomposition signals into the encoder model to construct a wavelet decomposition network to obtain mixed characteristic distribution.
(3) The constructed double-flow differential network model is characterized in that the network structure of the double-flow differential network model is characterized in that the multi-scale differential fusion network mainly comprises seven local differential networks. The differential characteristics of the input layer and the five intermediate layers and the coding layers in the dual-stream encoder network are input to the corresponding local differential network model. And finally, splicing the results of each differential network, passing through a multi-layer perceptron module, and outputting a prediction result. The difference characteristic calculation method of each scale is shown as follows
y＝f mlp (z d )
Wherein y is the predicted result,and->For the ith convolutional layer feature in dual stream, < +.>As a difference feature of the i-th layer,representing the difference characteristics of the original signal, < >>For the difference of the coding features, the dimension of each difference feature corresponds to the dimension of the convolution layer, and the range of the parameter θ is [0,1]The influence of the Euclidean distance and the absolute value distance on the differential feature is determined by adjusting the change of the theta parameter.
The oil-water two-phase flow content measuring method designed by the invention has the advantages that:
(1) The spiral electrode sensor has high reaction speed, can perform real-time monitoring and dynamic processing, and can rapidly and accurately obtain the content sequence fluctuation signal.
(2) The double-flow differential network adopted by the invention has objectivity in predicting the content, and can have stronger prediction performance on oil-water two-phase flow data.
Claims (4)
1. The measuring method of the oil-water two-phase flow parameter measurement, in particular to the measuring method of the oil-water two-phase flow parameter measurement based on a double-flow differential network, which is characterized by comprising the following steps:
(1) Measuring fluctuation signals of oil-water two-phase flow parameters by adopting a spiral electrode sensor;
(2) And constructing a double-flow differential network model for realizing characteristic fusion and content measurement of the oil-water two-phase flow parameter fluctuation signal.
2. The method for measuring parameters of an oil-water two-phase flow according to claim 1, wherein the step (1) comprises:
when materials enter a measuring pipe section from a pipeline, oil-water two-phase flow parameter fluctuation signal measurement is carried out by a spiral electrode sensor arranged on the pipeline, the designed electrostatic sensor consists of a spiral electrode, an insulating layer and a material pipeline, firstly, a spiral induction electrode is designed according to the working principle of a microwave sensor, the spiral induction electrode is positioned in the middle of the pipeline, then the spiral induction electrode on one side is excited by a high-frequency excitation source, the signal receiving is carried out on the other side by an AD8302 module, meanwhile, a shielding layer made of a copper sheet is covered outside the whole sensor, and the measuring signal of the oil-water two-phase flow can be effectively collected by the spiral electrode sensor.
3. The method for measuring parameters of an oil-water two-phase flow according to claim 1, wherein the step (2) comprises:
(1) the constructed double-flow differential network model is characterized in that firstly, oil-water two-phase flow signals measured by a spiral electrode sensor are preprocessed, fluctuation signals are respectively segmented in a non-overlapping window mode, the window length is H, and sample signals with the length of L can be obtainedSamples of (a) wherein->Representative of rounding down, and measuring fluctuation signals acquired by the sensorThe number is subjected to windowing interception, the actual value is used as a data tag, N parameter fluctuation samples with tag values are obtained, and the signal after sliding window is subjected to discrete wavelet decomposition to obtain N preprocessing signals;
(2) the data set is divided by 2N samples after the original signal and the discrete wavelet are decomposed at random, and the specific proportion is [ training set: verification set: test set ] = [8:1:1].
4. The method for measuring parameters of an oil-water two-phase flow according to claim 1, wherein the step (2) comprises:
(1) the built double-flow differential network model is characterized in that the model comprises a double-flow self-coding network and a multi-scale differential fusion network, the double-flow self-coding network comprises an original data flow and a discrete wavelet data flow, coding features are extracted by utilizing the self-coding network, the multi-scale differential fusion network decouples feature attributes under different working conditions by fusing multi-scale differential features of different levels, and meanwhile, the multi-layer perceptron model fuses feature differences of a multi-scale intermediate layer to obtain a final content prediction result;
(2) the constructed double-flow differential network model is characterized in that the self-coding model adopts a depth residual neutral network and mainly comprises five convolution modules, signals are coded to high-order characteristic representation, then the high-dimensional characteristic is sampled to the corresponding characteristic size in the encoder model through a decoder formed by deconvolution modules, wherein the sizes of the five convolution modules are 3 multiplied by 3, the sizes of the five deconvolution modules are 3 multiplied by 3, and the self-coding process is as follows
Where f (x) is the encoder, g (h) is the decoder,is realized by reconstructing a signal through a decoding function in the pre-training process of a self-coding modelFirstly, normalizing signals and inputting the signals into an encoder model f (x), extracting high-dimensional characteristic representation of the signals, then obtaining a reconstructed signal through a decoder model g (h), carrying out parameter updating iteration of the model through a Mean Square Error (MSE) loss function by model training, wherein the formula of the loss function is as follows
Wherein x is i,j Andfor the original signal and the value of the generated signal, n and n are the dimensions of the signal;
(3) the constructed double-flow differential network model is characterized in that the multi-scale differential fusion network mainly comprises seven local differential networks, differential characteristics of an input layer in a double-flow encoder network and five middle layers and an encoding layer in a convolutional neural network are input into the corresponding local differential network model, finally, the result of each differential network is spliced and passes through a multi-layer perceptron module, a prediction result is output, and the differential characteristic calculation method of each scale is shown as the following formula
y＝f mlp (z d )
Wherein y is the predicted result,and->For the ith convolutional layer feature in dual stream, < +.>For the difference feature of the i-th layer, +.>Representing the difference characteristics of the original signal, < >>For the difference of the coding features, the dimension of each difference feature corresponds to the dimension of the convolution layer, and the range of the parameter θ is [0,1]The influence of the Euclidean distance and the absolute value distance on the differential feature is determined by adjusting the change of the theta parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311639487.3A CN117665009A (en) | 2023-12-01 | 2023-12-01 | Oil-water two-phase flow parameter measurement method of double-flow differential network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311639487.3A CN117665009A (en) | 2023-12-01 | 2023-12-01 | Oil-water two-phase flow parameter measurement method of double-flow differential network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117665009A true CN117665009A (en) | 2024-03-08 |
Family
ID=90080213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311639487.3A Pending CN117665009A (en) | 2023-12-01 | 2023-12-01 | Oil-water two-phase flow parameter measurement method of double-flow differential network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117665009A (en) |
-
2023
- 2023-12-01 CN CN202311639487.3A patent/CN117665009A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111638249B (en) | Water content measuring method based on deep learning and application of water content measuring method in oil well exploitation | |
| CN111649779B (en) | Oil well oil content and total flow rate measuring method based on dense neural network and application | |
| CN103712939A (en) | Pollutant concentration fitting method based on ultraviolet-visible spectrum | |
| CN114722883A (en) | Welding quality real-time detection method and system based on high-frequency time sequence data | |
| CN111810115B (en) | Underground real-time monitoring method and device for stratum water pollution rate | |
| CN112814649B (en) | Intelligent metering device and method for water content of oil well produced liquid | |
| CN110630256A (en) | Low-gas-production oil well wellhead water content prediction system and method based on depth time memory network | |
| CN112183590A (en) | Transformer fault diagnosis method based on Oneclass SVM algorithm | |
| CN112926767A (en) | Annular fog flow gas phase apparent flow velocity prediction method based on particle swarm BP neural network | |
| CN110632132A (en) | High-yield gas-oil well wellhead liquid water content prediction method based on multi-sensor measurement and deep convolutional neural network | |
| CN109060771A (en) | A kind of common recognition model building method based on spectrum different characteristic collection | |
| CN117665009A (en) | Oil-water two-phase flow parameter measurement method of double-flow differential network | |
| CN112700051A (en) | Res-TCN neural network-based intelligent prediction method for oil well liquid production associated gas | |
| CN117056865A (en) | Method and device for diagnosing operation faults of machine pump equipment based on feature fusion | |
| CN110630244A (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 | |
| CN110243886A (en) | A kind of low yield gas well mouth of oil well hydrated comples ion method based on nonlinear characteristic | |
| CN110243885A (en) | A kind of low yield gas well mouth of oil well hydrated comples ion method based on time-frequency characteristics | |
| CN115929285A (en) | Geothermal gradient prediction method based on Lagrange support vector machine algorithm | |
| CN115184054A (en) | Mechanical equipment semi-supervised fault detection and analysis method, device, terminal and medium | |
| CN111197471B (en) | Transient electromagnetic detection calculation model and detection method for underground screen pipe | |
| Zhao et al. | Research on Transformer Oil Multi-frequency Ultrasonic Monitoring Technology Based on Convolutional Neural Network | |
| CN117664227A (en) | Multi-sensor integrated gas-liquid two-phase flow content measuring method | |
| CN117489337A (en) | Oil well parameter measurement method based on multi-mode sensor | |
| Wang et al. | Method for Determining Lower Limit of Soil Geochemical Anomaly and Its Advantages and Disadvantages | |
| CN115618182A (en) | Two-phase flow measuring method, device and system based on aggregation coefficient entropy and network |
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 |