CN102269684A - Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method - Google Patents
Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method Download PDFInfo
- Publication number
- CN102269684A CN102269684A CN2011101680812A CN201110168081A CN102269684A CN 102269684 A CN102269684 A CN 102269684A CN 2011101680812 A CN2011101680812 A CN 2011101680812A CN 201110168081 A CN201110168081 A CN 201110168081A CN 102269684 A CN102269684 A CN 102269684A
- Authority
- CN
- China
- Prior art keywords
- liquid
- flow
- capacitance
- flow pattern
- 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
- 230000005514 two-phase flow Effects 0.000 title claims abstract description 73
- 239000007788 liquid Substances 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 20
- 230000003287 optical effect Effects 0.000 claims abstract description 61
- 238000005259 measurement Methods 0.000 claims abstract description 55
- 238000012706 support-vector machine Methods 0.000 claims abstract description 5
- 230000008859 change Effects 0.000 claims abstract description 4
- 238000001514 detection method Methods 0.000 claims description 11
- 238000012843 least square support vector machine Methods 0.000 claims description 10
- 230000005284 excitation Effects 0.000 claims description 9
- 239000002184 metal Substances 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000003909 pattern recognition Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 8
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Measuring Volume Flow (AREA)
Abstract
The invention discloses a small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and a method. The system comprises a transparent insulated pipeline, a capacitance sensor, a capacitance/voltage converter, a capacitance data acquisition unit, an optical sensor, a photocurrent/voltage converter, an optical data acquisition unit and a computer. Firstly the capacitance sensor and the optical sensor are used to simultaneously obtain a measuring signal which reflects the flow state change of liquid-liquid two-phase flow; the time domain characteristic quantity (mean value and standard deviation), the amplitude domain characteristic quantity (kurtosis and skewness of a probability density function curve) and the frequency domain characteristic quantity (energy percentages of the signals at different frequency bands) of the measuring signal are extracted; the characteristic quantity is inputted into a least squares support vector machine flow pattern classifier so as to finally realize the flow pattern identification of the small-diameter pipeline liquid-liquid two-phase flow. The invention has the advantages of no contact, high precision and the like, not only can effectively identify various flow patterns of small-diameter pipeline liquid-liquid two-phase flow, but also can provide beneficial reference for the measurement of other parameters of liquid-liquid two-phase flow.
Description
Technical Field
The invention relates to the field of multiphase flow measurement, in particular to a flow pattern identification system and method for small-pipeline liquid-liquid two-phase flow.
Background
The liquid-liquid two-phase flow system is widely used in the industrial fields of petroleum, daily chemicals, food, pharmacy and the like. In recent years, with the development of the miniaturization trend of industrial equipment, liquid-liquid two-phase flow systems of millimeter-scale small pipelines are researched and applied more and more, and the online detection of parameters of the millimeter-scale small pipelines is paid more and more attention. However, compared with the liquid-liquid two-phase flow at other scales, the mechanism research and the engineering application of the millimeter-scale small-pipeline liquid-liquid two-phase flow are relatively weak, and the corresponding detection technology is extremely lacking and needs to be strengthened urgently.
The flow pattern as an important parameter of the liquid-liquid two-phase flow not only affects the fluid dynamic characteristics and the heat and mass transfer performance of the two-phase flow, but also affects the accurate measurement of other parameters of the two-phase flow to a great extent. Therefore, the accurate identification of the flow pattern has important significance for the research and engineering application of the liquid-liquid two-phase flow. However, the flow characteristics of the liquid-liquid two-phase flow are very complex, and due to different dimensions, the liquid-liquid two-phase flow of the small pipe often shows different characteristics from the liquid-liquid two-phase flow of the conventional pipe diameter, which all pose higher requirements and challenges for flow pattern identification of the liquid-liquid two-phase flow of the small pipe. At present, the literature reports on the identification of the flow pattern of the liquid-liquid two-phase flow of the small pipeline are very limited, and the research and development are urgently needed.
Disclosure of Invention
The invention aims to overcome the defects of the existing research and technology and provide a flow pattern identification system and method for small-pipeline liquid-liquid two-phase flow.
The small pipeline liquid-liquid two-phase flow pattern identification system comprises a transparent insulating pipeline, a capacitance sensor, a capacitance/voltage converter, a capacitance data acquisition unit, an optical sensor, a photocurrent/voltage converter, an optical data acquisition unit and a computer; the capacitance sensor and the optical sensor are respectively arranged on the transparent insulating pipeline, the capacitance sensor is connected with the capacitance/voltage converter, the capacitance/voltage converter is connected with a computer through a capacitance data acquisition unit, the optical sensor is connected with the photocurrent/voltage converter, and the photocurrent/voltage converter is connected with the computer through the optical data acquisition unit.
The capacitance sensor is as follows: two metal electrodes with the same size, namely an excitation electrode and a detection electrode, are symmetrically adhered to the outer wall of the insulating pipeline relative to the central axis of the insulating pipeline, the excitation electrode is connected with a lead, the detection electrode is also connected with the lead, a protection electrode is arranged between the excitation electrode and the detection electrode and clings to the outer wall of the pipeline, the outer side of the whole measuring pipeline is wrapped by a metal shielding cover, and the protection electrode is connected with the metal shielding cover.
The optical sensor comprises a transparent pipeline, a laser, a cylindrical lens and a photocell, wherein the laser and the cylindrical lens are sequentially arranged on one side of the transparent pipeline, the optical axes of the laser and the cylindrical lens are vertical to the central axis of the transparent pipeline, and the photocell is arranged on the other side of the transparent pipeline on a plane vertical to the optical axes of the laser and the cylindrical lens.
The method for identifying the flow pattern of the liquid-liquid two-phase flow of the small pipeline comprises the following steps:
1) respectively and simultaneously obtaining measurement signals reflecting the flow state change of the liquid-liquid two-phase flow of the small pipeline by a capacitance sensor and an optical sensor, converting the capacitance measurement signals into voltage signals by a capacitance/voltage converter, sending the voltage signals to a computer by a capacitance data acquisition unit, and sending the voltage signals to the computer by an optical data acquisition unit after converting the optical measurement signals into the voltage signals by a photocurrent/voltage converter;
2) extracting time domain, amplitude domain and frequency domain characteristic quantities of a small pipeline liquid-liquid two-phase flow capacitance measurement signal and an optical measurement signal;
3) and inputting the extracted characteristic quantities of the capacitance measurement signal and the optical measurement signal into a flow pattern classifier established by a least square support vector machine, and identifying the flow pattern of the small-pipeline liquid-liquid two-phase flow on line.
The step 2) of extracting the time domain, amplitude domain and frequency domain characteristic quantities of the small pipeline liquid-liquid two-phase flow capacitance measurement signal and the optical measurement signal comprises the following steps:
(1) time domain feature quantity extraction
The extracted time domain characteristic quantity is the mean value and standard deviation of the small pipeline liquid-liquid two-phase flow measurement signal,
mean value:
standard deviation:
wherein
In order to measure the time series of the signal,is a serial number of a time series,measuring the signal time sequence length;
(2) amplitude domain feature quantity extraction
The extracted amplitude domain characteristic quantity is the kurtosis of the probability density function curve of the small-pipeline liquid-liquid two-phase flow measurement signal
Degree of convergence
，
Kurtosis:
skewness:
wherein
，
Are respectively probability density function
The mean value and the standard deviation of (a),is the sequence number of the sequence of probability density functions,as a function of probability density
Length of (d);
(3) empirical mode decomposition is carried out on the capacitance measurement signal and the optical measurement signal of the small pipeline liquid-liquid two-phase flow to obtain multilayer inherent modal function components, the percentage of the energy of the first 12 layers of inherent modal function components in the total energy of all the inherent modal function components is calculated,
wherein,
the percentage of the energy of the ith layer natural mode function component to the total energy of all the natural mode function components,
is the energy of the i-th layer natural mode function component,
when the number of layers of the inherent mode function components obtained by the empirical mode decomposition of the signal is less than 12, the vacant energy percentage is replaced by zero;
(4) obtaining 32 characteristic quantities representing flow pattern characteristics of small pipeline liquid-liquid two-phase flow
Wherein,
the signal characteristic quantities are measured for 16 capacitances,
the signal characteristic quantities are measured for 16 optical measurements.
Inputting the extracted characteristic quantities of the capacitance measurement signal and the optical measurement signal into a flow pattern classifier established by a least square support vector machine, and performing online identification on the flow pattern of the small-pipeline liquid-liquid two-phase flow, wherein the step 3) is as follows:
for the identified multiple small pipeline liquid-liquid two-phase flow patterns, two types of classifiers based on a least square support vector machine are designed between every two flow patterns, and the decision function of each flow pattern classifier has the following form:
wherein,
in order to train the sample set,
in order to be a characteristic quantity of the image,
are labeled for the purpose of a category,
in order to train the sample numbers of the set,
in order to train the number of samples in the set,
for the characteristic quantity of the flow pattern sample to be tested,
and
for the trained determined least squares support vector machine parameters,
in order to be a kernel function, the kernel function,
is a parameter of the kernel function;
for the flow pattern sample to be tested, firstly, the characteristic quantities of the signals in a time domain, an amplitude domain and a frequency domain are extracted, then the extracted characteristic quantities are input into each classifier for testing, a ticket is cast to the corresponding flow pattern according to the test result, and finally the flow pattern with the most tickets is used as the current flow pattern identification result.
The invention provides a flow pattern identification system and method for small-pipeline liquid-liquid two-phase flow. The method has the advantages of non-contact, high identification precision and the like, not only provides a feasible and effective way for identifying the flow pattern of the small-pipeline liquid-liquid two-phase flow, but also provides a beneficial reference for measuring other parameters of the small-pipeline liquid-liquid two-phase flow.
Drawings
FIG. 1 is a block diagram of a flow pattern identification system for a small pipe liquid-liquid two-phase flow;
FIG. 2 is a schematic diagram of a capacitive sensor construction of the present invention;
FIG. 3 is a schematic diagram of the optical sensor configuration of the present invention;
FIG. 4 is a flow chart of a flow pattern identification method for a small pipe liquid-liquid two-phase flow.
Detailed Description
Aiming at the current situation that a flow pattern identification method of liquid-liquid two-phase flow of a small pipeline is lacked, the invention provides a system and a method for identifying the flow pattern of the liquid-liquid two-phase flow of the small pipeline by comprehensively utilizing advanced information processing technologies such as a capacitance sensor, an optical sensor, a Probability Density Function (PDF), Empirical Mode Decomposition (EMD) and Least square Support Vector machine (LS-SVM). The method has the advantages of non-contact, high precision and the like, and can effectively identify various flow patterns of the small-pipeline liquid-liquid two-phase flow.
As shown in fig. 1, the small pipe liquid-liquid two-phase flow pattern identification system includes a transparent insulated pipe, a capacitance sensor, a capacitance/voltage converter, a capacitance data acquisition unit, an optical sensor, a photocurrent/voltage converter, an optical data acquisition unit, and a computer; the capacitance sensor and the optical sensor are respectively arranged on the transparent insulating pipeline, the capacitance sensor is connected with the capacitance/voltage converter, the capacitance/voltage converter is connected with a computer through a capacitance data acquisition unit, the optical sensor is connected with the photocurrent/voltage converter, and the photocurrent/voltage converter is connected with the computer through the optical data acquisition unit.
As shown in fig. 2, the capacitive sensor is: two metal electrodes with the same size, namely an excitation electrode 1 and a detection electrode 2, are symmetrically adhered to the outer wall of an insulating pipeline 3 relative to the central axis of the insulating pipeline, the excitation electrode is connected with a lead 4, the detection electrode is connected with a lead 5, a protection electrode 6 is arranged between the excitation electrode and the detection electrode and clings to the outer wall of the pipeline, the outer side of the whole measuring pipeline is wrapped by a metal shielding cover 7, and the protection electrode is connected with the metal shielding cover.
As shown in fig. 3, the optical sensor includes a transparent pipe, a laser, a cylindrical lens and a photocell, the laser and the cylindrical lens are sequentially installed at one side of the transparent pipe, optical axes of the laser and the cylindrical lens are perpendicular to a central axis of the transparent pipe, and the photocell is installed at a plane perpendicular to the optical axes of the laser and the cylindrical lens at the other side of the transparent pipe. When the sensor works, the point light source generated by the laser is converted into sheet light after passing through the cylindrical lens, the sheet light is orthogonally projected onto the transparent pipeline as incident light, passes through the liquid-liquid two-phase flow in the pipeline, is emitted from the other side of the transparent pipeline and is projected onto the photocell, and the photocell converts the detected light intensity into a current signal to be output.
In the actual online flow pattern identification process, liquid-liquid two-phase flow enters a measuring pipeline and flows through a capacitance sensor and an optical sensor. The capacitance measurement signal obtained by the capacitance sensor is converted into a voltage signal by a capacitance/voltage converter, and then is sent to a computer by a capacitance data acquisition unit. The optical measurement signal obtained by the optical sensor is converted into a voltage signal by a photocurrent/voltage converter, and then is sent to a computer by an optical data acquisition unit. The computer stores, analyzes and calculates the obtained capacitance measurement signal and optical measurement signal.
As shown in fig. 4, the method for identifying the flow pattern of the liquid-liquid two-phase flow in the small pipeline comprises the following steps:
1) respectively and simultaneously obtaining measurement signals reflecting the flow state change of the liquid-liquid two-phase flow of the small pipeline by a capacitance sensor and an optical sensor, converting the capacitance measurement signals into voltage signals by a capacitance/voltage converter, sending the voltage signals to a computer by a capacitance data acquisition unit, and sending the voltage signals to the computer by an optical data acquisition unit after converting the optical measurement signals into the voltage signals by a photocurrent/voltage converter;
2) extracting time domain, amplitude domain and frequency domain characteristic quantities of a small pipeline liquid-liquid two-phase flow capacitance measurement signal and an optical measurement signal;
3) and inputting the extracted characteristic quantities of the capacitance measurement signal and the optical measurement signal into a flow pattern classifier established by a least square support vector machine, and identifying the flow pattern of the small-pipeline liquid-liquid two-phase flow on line.
The step 2) of extracting the time domain, amplitude domain and frequency domain characteristic quantities of the small pipeline liquid-liquid two-phase flow capacitance measurement signal and the optical measurement signal comprises the following steps:
(1) time domain feature quantity extraction
The extracted time domain characteristic quantity is the mean value and standard deviation of the small pipeline liquid-liquid two-phase flow measurement signal,
mean value:
standard deviation:
wherein
In order to measure the time series of the signal,
is a serial number of a time series,
measuring the signal time sequence length;
(2) amplitude domain feature quantity extraction
Extracted amplitude feature quantityPeak degree of signal probability density function curve for small pipeline liquid-liquid two-phase flow measurement
Degree of convergence
，
Kurtosis:
skewness:
wherein
，
Are respectively probability density function
The mean value and the standard deviation of (a),
is the sequence number of the sequence of probability density functions,
as a function of probability density
Length of (d);
(3) and performing empirical mode decomposition on the capacitance measurement signal and the optical measurement signal of the small-pipeline liquid-liquid two-phase flow to obtain a multilayer inherent modal function component.
According to the characteristics of the signal, the EMD decomposes the original signal into the sum of finite layer natural mode function components according to the sequence of the frequency from high to low, and each layer of natural mode function component contains the local characteristic information of the original signal. The empirical mode function component of each layer needs to satisfy the following two conditions: 1) in the whole empirical mode function component data sequence section, the number of zero points is equal to the number of poles or the difference is 1, 2) for any data point, the envelope of the local maximum value and the envelope mean value of the local minimum value are zero. According to this rule, the eigenmode function components are "filtered" one by one until the time series becomes a monotonic function. Finally, the raw measurement signal is decomposed into the sum of the multilayer eigenmode function components and the trend term, i.e.:
wherein,
in order to obtain a mean-removed measurement signal,is composed of
The characteristic frequency of the signal represented by the layer natural mode function component is gradually reduced,
the trend term represents the monotonous trend of the signal after all the natural modal function components are extracted.
The percentage of the energy of the first 12 layers of natural modal function components in the total energy of all the natural modal function components is calculated,
wherein,
the percentage of the energy of the ith layer natural mode function component to the total energy of all the natural mode function components,
is the energy of the i-th layer natural mode function component,
when the number of layers of the inherent mode function components obtained by the empirical mode decomposition of the signal is less than 12, the vacant energy percentage is replaced by zero;
(4) obtaining 32 characteristic quantities representing flow pattern characteristics of small pipeline liquid-liquid two-phase flow
Wherein,
the signal characteristic quantities are measured for 16 capacitances,
the signal characteristic quantities are measured for 16 optical measurements.
Inputting the extracted characteristic quantities of the capacitance signal and the optical signal into a flow pattern classifier established by a least square support vector machine, and performing online identification on the flow pattern of the small-pipeline liquid-liquid two-phase flow, wherein the step 3) is as follows:
for a variety of identified small pipe liquid-liquid two-phase flow patterns, such as: the method comprises the following steps of (1) plug flow, laminar flow, oil-based bubble flow, water-based bubble flow, oil core annular flow and water core annular flow, wherein two types of classifiers based on a least square support vector machine are designed between every two flow types, the six typical flow types are totally 15 types of classifiers, and a decision function of each flow type classifier has the following form:
wherein,
in order to train the sample set,
in order to be a characteristic quantity of the image,
is a flow pattern class designation,
in order to train the sample numbers of the set,in order to train the number of samples in the set,
for the characteristic quantity of the flow pattern sample to be tested,
and
for the trained determined least squares support vector machine parameters,
in order to be a kernel function, the kernel function,
is a parameter of the kernel function;
for the flow pattern sample to be tested, firstly, the characteristic quantities of the signals in a time domain, an amplitude domain and a frequency domain are extracted, then the extracted characteristic quantities are input into each classifier for testing, a ticket is cast to the corresponding flow pattern according to the test result, and finally the flow pattern with the most tickets is used as the current flow pattern identification result.
Two-phase oil-water flow experiments have been performed on small horizontal pipes with an internal diameter of 1.8mm, 3.1mm and 4.0mm, respectively. Six typical flow patterns of oil-water two-phase flow are realized by using the system and the method provided by the invention: the online identification of plug flow, laminar flow, oil-based bubble flow, water-based bubble flow, oil core annular flow and water core annular flow, and the identification result of the oil-water two-phase flow pattern under each pipe diameter is shown in table 1. Experimental results show that the flow pattern identification system and method for the liquid-liquid two-phase flow of the small pipe, provided by the invention, are effective, satisfactory flow pattern identification accuracy is obtained, and the identification accuracy of each typical flow pattern under each pipe diameter is over 96%.
TABLE 1 identification result of oil-water two-phase flow pattern under each pipe diameter
| Number of samples in training set | Number of samples in test set | Number of misclassified samples | Flow pattern identification accuracy (%) | |
| ID=1.8mm | 198 | 99 | 2 | 97.98% |
| ID=3.1mm | 206 | 103 | 0 | 100% |
| ID=4.0mm | 190 | 95 | 3 | 96.84% |
Claims (6)
1. A flow pattern identification system for liquid-liquid two-phase flow of a small pipeline is characterized by comprising a transparent insulating pipeline, a capacitance sensor, a capacitance/voltage converter, a capacitance data acquisition unit, an optical sensor, a photocurrent/voltage converter, an optical data acquisition unit and a computer; the capacitance sensor and the optical sensor are respectively arranged on the transparent insulating pipeline, the capacitance sensor is connected with the capacitance/voltage converter, the capacitance/voltage converter is connected with a computer through a capacitance data acquisition unit, the optical sensor is connected with the photocurrent/voltage converter, and the photocurrent/voltage converter is connected with the computer through the optical data acquisition unit.
2. The system according to claim 1, wherein the capacitive sensor is: two metal electrodes with the same size, namely an excitation electrode (1) and a detection electrode (2), are symmetrically adhered to the outer wall of the insulating pipeline (3) relative to the central axis of the insulating pipeline (3), the excitation electrode is connected with a lead (4), the detection electrode is connected with a lead (5), a protection electrode (6) is arranged between the excitation electrode and the detection electrode and clings to the outer wall of the pipeline, the outer side of the whole measuring pipeline is wrapped by a metal shielding cover (7), and the protection electrode is connected with the metal shielding cover.
3. The two-phase flow pattern recognition system according to claim 1, wherein the optical sensor comprises a transparent pipe, a laser, a cylindrical lens and a photocell, the laser and the cylindrical lens are sequentially mounted on one side of the transparent pipe, optical axes of the laser and the cylindrical lens are perpendicular to a central axis of the transparent pipe, and the photocell is mounted on the other side of the transparent pipe on a plane perpendicular to the optical axes of the laser and the cylindrical lens.
4. A method for identifying the flow pattern of a small pipe liquid-liquid two-phase flow by using the system of claim 1, comprising the steps of:
1) respectively and simultaneously obtaining measurement signals reflecting the flow state change of the liquid-liquid two-phase flow by a capacitance sensor and an optical sensor, converting the capacitance measurement signals into voltage signals by a capacitance/voltage converter, sending the voltage signals to a computer by a capacitance data acquisition unit, and sending the voltage signals to the computer by an optical data acquisition unit after converting the optical measurement signals into the voltage signals by a photocurrent/voltage converter;
2) extracting time domain, amplitude domain and frequency domain characteristic quantities of a small pipeline liquid-liquid two-phase flow capacitance measurement signal and an optical measurement signal;
3) and inputting the extracted characteristic quantities of the capacitance measurement signal and the optical measurement signal into a flow pattern classifier established by a least square support vector machine, and identifying the flow pattern of the small-pipeline liquid-liquid two-phase flow on line.
5. The method for identifying the flow pattern of the small pipe liquid-liquid two-phase flow according to claim 4, wherein the step of extracting the time domain, amplitude domain and frequency domain characteristic quantities of the small pipe liquid-liquid two-phase flow capacitance measurement signal and the optical measurement signal in step 2) comprises the steps of:
(1) time domain feature quantity extraction
The extracted time domain characteristic quantity is the mean value and standard deviation of the small pipeline liquid-liquid two-phase flow measurement signal,
mean value:
standard deviation:
wherein
In order to measure the time series of the signal,
is a serial number of a time series,
to measure the length of the signal time series;
(2) amplitude domain feature quantity extraction
The extracted amplitude domain characteristic quantity is the kurtosis of the probability density function curve of the small-pipeline liquid-liquid two-phase flow measurement signal
Degree of convergence
，
Kurtosis:
skewness:
wherein
，
Are respectively probability density function
The mean value and the standard deviation of (a),
is the sequence number of the sequence of probability density functions,
as a function of probability density
Length of (d);
(3) empirical mode decomposition is carried out on the capacitance measurement signal and the optical measurement signal of the small pipeline liquid-liquid two-phase flow to obtain multilayer inherent modal function components, the percentage of the energy of the first 12 layers of inherent modal function components in the total energy of all the inherent modal function components is calculated,
wherein,
the percentage of the energy of the ith layer natural mode function component to the total energy of all the natural mode function components,
is the energy of the i-th layer natural mode function component,
when the number of layers of the natural mode function components obtained by the empirical mode decomposition of the signal is less than 12, the vacant energy percentage is replaced by zero;
(4) obtaining 32 characteristic quantities representing flow pattern characteristics of small pipeline liquid-liquid two-phase flow
Wherein,
the signal characteristic quantities are measured for 16 capacitances,
the signal characteristic quantities are measured for 16 optical measurements.
6. The method for identifying the flow pattern of the two-phase liquid-liquid flow in the small pipe according to claim 4, wherein the step 3) of inputting the extracted characteristic quantities of the capacitance measurement signal and the optical measurement signal into the flow pattern classifier established by the least squares support vector machine comprises the steps of:
for the identified multiple small pipeline liquid-liquid two-phase flow patterns, two types of classifiers based on a least square support vector machine are designed between every two flow patterns, and the decision function of each flow pattern classifier has the following form:
wherein,
in order to train the sample set,
in order to be a characteristic quantity of the image,
is a flow pattern class designation,in order to train the sample numbers of the set,in order to train the number of samples in the set,
for the characteristic quantity of the flow pattern sample to be tested,and
for the trained determined least squares support vector machine parameters,
in order to be a kernel function, the kernel function,
is a parameter of the kernel function;
for the flow pattern sample to be tested, firstly, the characteristic quantities of the signals in a time domain, an amplitude domain and a frequency domain are extracted, then the extracted characteristic quantities are input into each classifier for testing, a ticket is cast to the corresponding flow pattern according to the test result, and finally the flow pattern with the most tickets is used as the current flow pattern identification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101680812A CN102269684A (en) | 2011-06-22 | 2011-06-22 | Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101680812A CN102269684A (en) | 2011-06-22 | 2011-06-22 | Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102269684A true CN102269684A (en) | 2011-12-07 |
Family
ID=45052049
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101680812A Pending CN102269684A (en) | 2011-06-22 | 2011-06-22 | Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102269684A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590030A (en) * | 2012-01-18 | 2012-07-18 | 浙江大学 | Small-passage gas-liquid phase flow pattern identification device and method based on photovoltaic array sensor |
| CN103674478A (en) * | 2013-12-11 | 2014-03-26 | 中国石油大学(华东) | Low-gas-content gas-liquid two phase flow pattern identifying method |
| CN105527226A (en) * | 2016-01-20 | 2016-04-27 | 浙江大学 | Photoelectric diode array sensor-based ductule gas-liquid two-phase parameter measurement device and method |
| CN105842273A (en) * | 2016-03-16 | 2016-08-10 | 华北电力大学（保定） | Acquiring method and system for compression factor |
| CN109557146A (en) * | 2019-01-11 | 2019-04-02 | 南京工程学院 | A kind of apparatus for detecting disconnection of diamond wire and its prediction technique that breaks |
| CN110595948A (en) * | 2019-08-27 | 2019-12-20 | 杭州电子科技大学 | Small-channel two-phase flow parameter measuring device and method based on annular optical array |
| CN111191398A (en) * | 2019-12-06 | 2020-05-22 | 云南电网有限责任公司玉溪供电局 | SVR-based method for predicting degradation trend of storage battery of direct-current system of transformer substation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100507509C (en) * | 2006-08-01 | 2009-07-01 | 东北电力大学 | Oil gas water multiphase flow type identification method based on main component analysis and supporting vector machine |
| CN101603974B (en) * | 2009-07-21 | 2010-09-29 | 浙江大学 | Device and method for optical measurement for two-phase flow parameters of small-caliber pipeline |
-
2011
- 2011-06-22 CN CN2011101680812A patent/CN102269684A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100507509C (en) * | 2006-08-01 | 2009-07-01 | 东北电力大学 | Oil gas water multiphase flow type identification method based on main component analysis and supporting vector machine |
| CN101603974B (en) * | 2009-07-21 | 2010-09-29 | 浙江大学 | Device and method for optical measurement for two-phase flow parameters of small-caliber pipeline |
Non-Patent Citations (2)
| Title |
|---|
| ZHIYAO HUANG ET AL.: "Oil-Water Two-Phase Flow Measurement Based on a Hybrid Flowmeter and Dominant Phase Identification", 《12MTC 2009 - INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE》, 7 May 2009 (2009-05-07) * |
| 贺贞贞: "基于电容法的小通道气液两相流参数检测系统研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》, no. 8, 15 August 2010 (2010-08-15) * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102590030A (en) * | 2012-01-18 | 2012-07-18 | 浙江大学 | Small-passage gas-liquid phase flow pattern identification device and method based on photovoltaic array sensor |
| CN103674478A (en) * | 2013-12-11 | 2014-03-26 | 中国石油大学(华东) | Low-gas-content gas-liquid two phase flow pattern identifying method |
| CN103674478B (en) * | 2013-12-11 | 2016-01-06 | 中国石油大学(华东) | The Method for Discriminating Gas-liquid Two Phase Flow of low air void |
| CN105527226A (en) * | 2016-01-20 | 2016-04-27 | 浙江大学 | Photoelectric diode array sensor-based ductule gas-liquid two-phase parameter measurement device and method |
| CN105842273A (en) * | 2016-03-16 | 2016-08-10 | 华北电力大学（保定） | Acquiring method and system for compression factor |
| CN105842273B (en) * | 2016-03-16 | 2018-08-28 | 华北电力大学（保定） | A kind of acquisition methods and system of compressibility factor |
| CN109557146A (en) * | 2019-01-11 | 2019-04-02 | 南京工程学院 | A kind of apparatus for detecting disconnection of diamond wire and its prediction technique that breaks |
| CN109557146B (en) * | 2019-01-11 | 2024-02-13 | 南京工程学院 | Diamond wire breakage detection device and breakage prediction method thereof |
| CN110595948A (en) * | 2019-08-27 | 2019-12-20 | 杭州电子科技大学 | Small-channel two-phase flow parameter measuring device and method based on annular optical array |
| CN111191398A (en) * | 2019-12-06 | 2020-05-22 | 云南电网有限责任公司玉溪供电局 | SVR-based method for predicting degradation trend of storage battery of direct-current system of transformer substation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102269684A (en) | Small-diameter pipeline liquid-liquid two-phase flow flow pattern identification system and method | |
| Shi et al. | Theory and application of magnetic flux leakage pipeline detection | |
| CN108593260B (en) | Optical cable line fault positioning and detecting method and terminal equipment | |
| Yang et al. | A novel methodology for series arc fault detection by temporal domain visualization and convolutional neural network | |
| Yi et al. | EEMD-based steady-state indexes and their applications to condition monitoring and fault diagnosis of railway axle bearings | |
| CN106643965B (en) | Method for accurately identifying liquid level by utilizing template matching | |
| Zhao et al. | Real‐time fabric defect detection based on multi‐scale convolutional neural network | |
| Han et al. | A critical review of on-line oil wear debris particle detection sensors | |
| Yang et al. | Comparison of impulse wave and sweep frequency response analysis methods for diagnosis of transformer winding faults | |
| CN104298994A (en) | Pointer-type meter reading identification method and device | |
| Abdalla et al. | A novel eddy current testing error compensation technique based on mamdani-type fuzzy coupled differential and absolute probes | |
| CN111639795A (en) | Intelligent inspection task planning method for transformer substation robot | |
| CN102590030A (en) | Small-passage gas-liquid phase flow pattern identification device and method based on photovoltaic array sensor | |
| Demachi et al. | Crack depth estimation of non-magnetic material by convolutional neural network analysis of eddy current testing signal | |
| Thango et al. | A novel approach to assess power transformer winding conditions using regression analysis and frequency response measurements | |
| Xu et al. | Image recognition model of pipeline magnetic flux leakage detection based on deep learning | |
| Yan et al. | Partial discharge pattern-recognition method based on embedded artificial intelligence | |
| Kumar et al. | A Review on the Classification of Partial Discharges in Medium-Voltage Cables: Detection, Feature Extraction, Artificial Intelligence-Based Classification, and Optimization Techniques | |
| CN105527226A (en) | Photoelectric diode array sensor-based ductule gas-liquid two-phase parameter measurement device and method | |
| Kumar et al. | Intelligent Vibration Monitoring System for Smart Industry Utilizing Optical Fiber Sensor Combined with Machine Learning | |
| Fu et al. | A Partial Discharge Signal Separation Method Applicable for Various Sensors Based on Time-Frequency Feature Extraction of t-SNE | |
| Sun et al. | 3D imaging with single‐plane electrical capacitance tomography sensor | |
| Ardila-Rey et al. | A validation of the spectral power clustering technique (SPCT) by using a Rogowski coil in partial discharge measurements | |
| Riba et al. | Using CMOS image sensors to determine the intensity of electrical discharges for aircraft applications | |
| Meng et al. | Real-time automatic thickness recognition using pulse eddy current with deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20111207 |