阅读说明：本技术 一种应用少电极ect的气液两相含率检测方法 (Gas-liquid two-phase content rate detection method applying ECT with few electrodes ) 是由 孙江涛 李效霖 田文斌 白旭 徐立军 于 2021-09-10 设计创作，主要内容包括：本发明公开了一种应用少电极ECT的气液两相含率检测方法,属于电磁测量领域。首先,将6电极ECT传感器安装于待测气液两相流管道外壁,逐一对电极施加电压,得到所有电极对之间的电容,并进行归一化；管道内的待成像区域利用归一化电容通过像素插值图像重建算法得到重建图像。同理,利用满场电容值C-(H)和空场电容值C-(L)重建满场重建图像G-(H)和空场重建图像G-(L),并分别作为最大值和最小值对重建图像进行归一化,得到归一化重建图像G'。根据归一化重建图像G'的各像素面积,获得像素的加权平均灰度值。最后,将归一化重建图像像素的加权平均灰度值代入灰度值-相含率映射模型,得到待测管道内的气液两相含率值。本发明减少了电极数量,提高两相含率测量精度。(The invention discloses a gas-liquid two-phase content detection method applying ECT with few electrodes, and belongs to the field of electromagnetic measurement. Firstly, mounting a 6-electrode ECT sensor on the outer wall of a gas-liquid two-phase flow pipeline to be measured, applying voltage to electrodes one by one to obtain capacitance between all electrode pairs, and performing normalization; and obtaining a reconstructed image in the region to be imaged in the pipeline by utilizing the normalized capacitance through a pixel interpolation image reconstruction algorithm. Similarly, the full field capacitance C is used H And null field capacitance value C L Reconstructing a full field reconstructed image G H And null field reconstructed image G L And normalizing the reconstructed image respectively as a maximum value and a minimum value to obtain a normalized reconstructed image G'. And obtaining the weighted average gray value of the pixels according to the area of each pixel of the normalized reconstructed image G'. And finally, substituting the weighted average gray value of the pixel of the normalized reconstructed image into a gray value-phase content mapping model to obtain a gas-liquid two-phase content value in the pipeline to be measured. The invention reduces the number of electrodes and improves the measurement precision of the two-phase content rate.)

1. A gas-liquid two-phase content rate detection method applying less electrodes ECT is characterized by comprising the following specific processes:

firstly, mounting a 6-electrode ECT sensor on the outer wall of a gas-liquid two-phase flow pipeline to be measured, applying voltage to 6 electrodes one by one to obtain capacitance between all electrode pairs, and performing normalized calculation to obtain normalized capacitance; obtaining a reconstructed image for a region to be imaged in the pipeline by using a pixel interpolation image reconstruction algorithm through the normalized capacitance; similarly, the full field capacitance C is used_HAnd null field capacitance value C_LRespectively obtaining full field reconstruction images G of the regions to be imaged by adopting a pixel interpolation image reconstruction algorithm_HAnd null field reconstructed image G_L；

Then, the full field reconstructed image G_HAnd null field reconstructed image G_LNormalizing the reconstructed image respectively as a maximum value and a minimum value to obtain a normalized reconstructed image G'; obtaining a weighted average gray value of the pixels according to the area of each pixel of the normalized reconstructed image G';

the weighted average gray scale value of a pixel is:

wherein m is the number of pixels, G' [ k ]]Is the grey value, A 'of the kth pixel of the normalized reconstructed image G'_kIs the area of the kth pixel, A' is the total area of the image;

finally, substituting the weighted average gray value of the pixel of the normalized reconstructed image into a gray value-phase content mapping model to obtain a gas-liquid two-phase content value in the pipeline to be measured;

according to the gray value-phase content mapping model, the liquid phase content in the pipeline to be tested is beta, and the gas phase content is 1-beta.

2. The gas-liquid two-phase content rate detection method using ECT with less electrodes as claimed in claim 1, wherein the normalized capacitance λ:

in the formula, C_MIs a capacitance matrix of all electrode pairs, C_HIs a high-scale capacitance matrix, C_LIs a low calibration capacitance matrix.

3. The gas-liquid two-phase content rate detection method using ECT with less electrodes as claimed in claim 1, wherein the specific process of the pixel interpolation image reconstruction algorithm is as follows:

step 201, periodically selecting boundary points from the region to be imaged, dividing the region to be imaged into P × Q regions, which are expressed as:

in the formula, X is the number of transverse pixels in the region to be imaged, Y is the number of longitudinal pixels in the region to be imaged, Δ X is the period of the transverse direction, and Δ Y is the period of the longitudinal direction;

step 202, setting elements of the region to be imaged to form a vector G according to the sequence of arranging from left to right and from top to bottom, and utilizing the vector G to perform alignment on each pixel point I [ x, y ] in each region]Combining the coordinates of four adjacent points to carry out normalization expression to obtain a point (r) to be interpolated_x,r_y)；

The normalization of the points to be interpolated of the divided regions is expressed as (r)_x,r_y)：

Step 203, according to the point (r) to be interpolated_x,r_y) To obtain a point I [ x, y ]]According to the bilinear interpolation formula, XY- (1-P) (1-Q) interpolation equations are established according to the bilinear interpolation formula, and the interpolation equations are rewritten into a matrix form to obtain a constraint relation matrix H;

the bilinear interpolation formula is:

I[r_x,r_y]≈I[0,0](1-r_x)(1-r_y)+I[0,1](1-r_x)r_y+I(1,0)r_x(1-r_y)+I[1,1]r_xr_y

the interpolation equation is:

namely, it is

-G[x+yX]+G[x-xmodΔx+(y-ymodΔy)X](1-r_x)(1-r_y)+G[x+Δx-xmodΔx+(y-ymodΔy)X](1-r_x)r_y+G[x-xmodΔx+(y+Δy-ymodΔy)X]r_x(1-r_y)+G[x+Δx-xmodΔx+(y+Δy-ymodΔy)X]r_xr_y＝0

Merging and writing the interpolation equations into a homogeneous linear equation set, and converting the homogeneous linear equation set into a matrix form:

HG＝O

obtaining a constraint relation matrix H from the matrix form;

step 204, merging the constraint relation matrix H into the original ECT sensitive field S to obtain a new sensitive field A;

step 205, bringing the new sensitive field A and the normalized capacitance lambda into a fusion image reconstruction method to obtain a reconstructed image;

and combining Tikhonov regularization and linear back projection to form a fused image reconstruction method.

4. The method for detecting gas-liquid two-phase content rate by using ECT with less electrodes as claimed in claim 1, wherein said normalized reconstructed image G' is:

Technical Field

The invention belongs to the field of electromagnetic measurement, and particularly relates to a gas-liquid two-phase content rate detection method applying less electrodes ECT.

Background

In recent years, the demand of gas-liquid two-phase flow parameter measurement in the fields of scientific research, biomedicine, energy exploitation, environmental monitoring and the like is increasing. For example, in a ventilator, reliable detection of possible air embolism in blood is required, otherwise the life safety of a patient is affected; in the natural gas moisture acquisition system, the accuracy of gas-liquid two-phase flow content measurement also has great influence on the acquisition efficiency. It is therefore desirable to visualize and accurately measure the two-phase flow in a pipeline. Based on the method, reference can be provided for the research of the gas-liquid two-phase flow mechanism and the system control in the pipeline.

The gas-liquid two-phase flow has complex mechanism and high measurement difficulty, and is a difficult problem in the field of two-phase flow measurement. Electrical Capacitance Tomography (ECT) techniques are often applied to detect two-phase flow. However, the signal-to-noise ratio of the traditional 12-electrode and 16-electrode ECT technology is lower, gas-liquid two-phase flow measurement is more complex compared with other two-phase flow, and the requirement on the signal-to-noise ratio of the system is higher. If the number of electrodes is simply reduced and the signal-to-noise ratio is improved, the imaging quality is seriously influenced, and the problem of measuring the two-phase content of the gas-liquid two-phase flow cannot be solved.

Disclosure of Invention

The invention provides a gas-liquid two-phase content detection method applying less electrodes ECT based on an ECT technology in order to realize visualization of gas-liquid two-phase flow and measurement of two-phase content.

The gas-liquid two-phase content rate detection method based on the ECT with less electrodes comprises the following specific steps:

step one, installing 6 electrodes ECT sensors on the outer wall of a gas-liquid two-phase flow pipeline to be measured, applying voltage to the 6 electrodes one by one, obtaining and normalizing the capacitance between all electrode pairs;

the normalized capacitance λ is calculated as:

in the formula, C_MIs a capacitance matrix of all electrode pairs, C_HIs a high-scale capacitance matrix, C_LIs a low calibration capacitance matrix.

Obtaining a visual reconstruction image of the gas-liquid two-phase fluid in the pipeline through a pixel interpolation image reconstruction algorithm for an area to be imaged in the pipeline by utilizing the normalized capacitor;

the method specifically comprises the following steps:

step 201, periodically selecting boundary points from the region to be imaged, dividing the region to be imaged into P × Q regions, which are expressed as:

in the formula, X is the number of pixels in the lateral direction in the region to be imaged, Y is the number of pixels in the longitudinal direction in the region to be imaged, Δ X is the period in the lateral direction, and Δ Y is the period in the longitudinal direction.

Step 202, setting elements of the region to be imaged to form a vector G according to the sequence of arranging from left to right and from top to bottom, and utilizing the vector G to perform alignment on each pixel point I [ x, y ] in each region]Combining the coordinate normalization of four adjacent points to obtain the pixel point Ix, y]Corresponding point (r) to be interpolated_x,r_y)；

Point I [ x, y]The corresponding elements in vector G are: g [ x + yX ]]The coordinates of four elements are normalized by using the element representation of the elements of four adjacent points of the point in the vector G to obtain the pixel point I [ x, y ]]Corresponding point (r) to be interpolated_x,r_y)：

Step 203, according to the point (r) to be interpolated_x,r_y) To obtain a point I [ x, y ]]According to the bilinear interpolation formula, XY- (1-P) (1-Q) interpolation equations are established according to the bilinear interpolation formula, and the interpolation equations are rewritten into a matrix form to obtain a constraint relation matrix H;

the bilinear interpolation formula is:

I[r_x,r_y]≈I[0,0](1-r_x)(1-r_y)+I[0,1](1-r_x)r_y+I(1,0)r_x(1-r_y)+I[1,1]r_xr_y

the interpolation equation is:

namely, it is

-G[x+yX]+G[x-x modΔx+(y-y modΔy)X](1-r_x)(1-r_y)+G[x+Δx-x modΔx+(y-y modΔy)X](1-r_x)r_y+G[x-x modΔx+(y+Δy-y modΔy)X]r_x(1-r_y)+G[x+Δx-x modΔx+(y+Δy-ymodΔy)X]r_xr_y＝0

Merging and writing the interpolation equations into a homogeneous linear equation set, and converting the homogeneous linear equation set into a matrix form:

HG＝O

from this matrix form, a constraint relation matrix H is derived.

Step 204, merging the constraint relation matrix H into the original ECT sensitive field S to obtain a new sensitive field A;

and step 205, bringing the new sensitive field A and the normalized capacitance lambda into a fusion image reconstruction method to obtain a reconstructed image.

And combining Tikhonov regularization and linear back projection to form a fused image reconstruction method.

Step three, similarly, using the full field capacitance C_HAnd null field capacitance value C_LObtaining a full field reconstruction image G by adopting a pixel interpolation image reconstruction algorithm for a region to be imaged_HAnd null field reconstructed image G_L。

Step four, reconstructing the full field image G_HAnd null field reconstructed image G_LAnd normalizing the reconstructed image respectively as a maximum value and a minimum value to obtain a normalized reconstructed image G'.

The normalized reconstructed image G' is:

step five: obtaining a weighted average gray value of the pixels according to the area of each pixel of the normalized reconstructed image G';

the weighted average gray scale value of a pixel is:

wherein m is the number of pixels, G' [ k ]]Is the grey value, A 'of the kth pixel of the normalized reconstructed image G'_kIs the area of the kth pixel and A' is the total area of the image.

And step six, substituting the weighted average gray value of the pixel of the normalized reconstructed image into a gray value-phase content mapping model to obtain a gas-liquid two-phase content value in the pipeline to be measured.

According to the gray value-phase content mapping model, the liquid phase content in the pipeline to be tested is beta, and the gas phase content is 1-beta.

The invention has the advantages that:

1. according to the invention, the ECT technology is adopted, so that the visualization of gas-liquid two-phase flow in the pipeline is realized, and the gas-liquid two-phase content is accurately measured;

2. the ECT system in the invention adopts the flexible PCB to manufacture the 6-electrode sensor, thereby reducing the number of electrodes, improving the signal to noise ratio of the system and simultaneously reducing the processing and installation difficulty.

3. In the invention, a pixel interpolation image reconstruction algorithm for expanding the sensitive field matrix is adopted, so that the image reconstruction quality of the ECT system under the condition of less electrodes is improved.

4. In the invention, before the gray value-phase content mapping model is brought in, the reconstructed image is normalized by adopting the empty field reconstructed image and the full field reconstructed image, so that the precision of the two-phase content measurement is improved.

Drawings

FIG. 1 is a schematic structural diagram of a 6-electrode ECT sensor for detecting gas-liquid two-phase content in a pipeline to be detected according to the invention;

FIG. 2 is a flow chart of a gas-liquid two-phase content rate detection method using low electrode capacitance tomography according to the present invention;

FIG. 3 is a flow chart of the present invention for obtaining a reconstructed image by a pixel interpolation image reconstruction algorithm.

Detailed Description

The technical solution of the present invention is further illustrated by the following examples and the accompanying drawings.

The invention provides a non-contact electromagnetic measurement method, which is used for visualizing gas-liquid two-phase flow in a pipeline by applying ECT (electron emission tomography) with few electrodes and measuring the two-phase content. By reducing the number of electrodes, the signal-to-noise ratio of capacitance data acquisition around the gas-liquid two-phase flow pipeline is improved, and the difficulty in the processing and installation process is reduced. Aiming at the problem of low resolution of the reconstructed image with few electrodes, the image reconstruction algorithm based on pixel interpolation is provided by combining the characteristics of the gas-liquid two-phase flow process in the pipeline, and the imaging quality of the ECT system with few electrodes is greatly improved. And finally, establishing a reconstructed image gray value-phase content mapping model, thereby realizing accurate measurement of the two-phase content.

The structure of the 6-electrode ECT sensor applied in the gas-liquid two-phase content detection is shown in figure 1, 6 same electrodes are uniformly pasted along the outer wall of a pipeline to be detected, a shielding shell is arranged on the outer periphery of the pipeline, and the distance between the shielding shell and the outer wall of the pipeline is equal to the thickness of the pipeline wall.

In the embodiment of the invention, the pipe wall of the selected to-be-detected pipeline is made of PVC material, the inner diameter of the pipeline is 2mm, the outer diameter of the pipeline is 3mm, and the inner diameter of the shielding shell is 4mm, so that the thickness of the pipe wall shell and the distance between the shielding shell and the outer wall of the pipeline are both 1 mm.

A gas-liquid two-phase content rate detection method applying ECT with few electrodes is shown in figure 2, and comprises the following specific steps:

step one, applying voltage to 6 electrodes of the 6-electrode ECT sensor one by one to obtain capacitance between all electrode pairs, and performing normalization calculation to obtain normalized capacitance lambda.

The method specifically comprises the following steps:

firstly, a voltage signal is applied to one electrode of the 6-electrode ECT sensor, and the capacitance of the electrode to other electrodes is measured;

then, voltage signals are applied to other electrodes one by one, and the capacitance between all electrode pairs is measured to obtain a capacitance matrix C_M；

Finally, a parallel normalization method is used to obtain a normalized capacitance λ:

in the formula, C_HIs a high nominal capacitance, C_LIs a low calibration capacitance.

And step two, acquiring a visual reconstruction image of the gas-liquid two-phase fluid in the pipeline by using a pixel interpolation image reconstruction algorithm for the region to be imaged in the pipeline by using the normalized capacitor.

The pixel interpolation image reconstruction algorithm is a new algorithm which firstly expands an ECT sensitive field matrix by utilizing bilinear interpolation, introduces priori knowledge and then reconstructs an image by using a fusion algorithm combining linear back projection and Tikhonov regularization.

As shown in fig. 3, specifically:

step 201, periodically selecting boundary points from the region to be imaged, dividing the region to be imaged into P × Q regions, which are expressed as:

in the formula, X is the number of pixels in the lateral direction in the region to be imaged, Y is the number of pixels in the longitudinal direction in the region to be imaged, Δ X is the period in the lateral direction, and Δ Y is the period in the longitudinal direction.

Step 202, setting elements in the region to be imaged to form a vector G according to the sequence of arranging from left to right and from top to bottom, and utilizing the vector G to perform alignment on each pixel point I [ x, y ] in each region]And performing element representation on four adjacent points of the upper, lower, left and right sides of the point to be interpolated (r), and normalizing the coordinates of the four elements to obtain the point to be interpolated (r)_x,r_y)；

The corresponding elements of I [ x, y ] in vector G are:

G[x+yX]

within the partitioned area, the element located at the top left corner of I [ x, y ] is:

where mod is the remainder symbol,

at I [ x, y]Upper left corner elementThe corresponding elements in vector G are:

similarly, lie in I [ x, y ]]Lower left corner elementThe elements in vector G that correspond to each other are:

elements of the upper right cornerIn the vector GWherein the corresponding elements are:

elements of the lower right cornerThe elements in vector G that correspond to each other are:

normalizing the coordinates of the four elements to obtain a normalized point (r) to be interpolated_x,r_y)：

Step 203, according to the point (r) to be interpolated_x,r_y) To obtain a point I [ x, y ]]According to the bilinear interpolation formula, XY- (1-P) (1-Q) interpolation equations are established according to the bilinear interpolation formula, and the interpolation equations are rewritten into a matrix form to obtain a constraint relation matrix H;

the bilinear interpolation formula is:

I[r_x,r_y]≈I[0,0](1-r_x)(1-r_y)+I[0,1](1-r_x)r_y+I(1,0)r_x(1-r_y)+I[1,1]r_xr_y

the interpolation equation is:

namely, it is

-G[x+yX]+G[x-x modΔx+(y-y modΔy)X](1-r_x)(1-r_y)+G[x+Δx-x modΔx+(y-y modΔy)X](1-r_x)r_y+G[x-x modΔx+(y+Δy-y modΔy)X]r_x(1-r_y)+G[x+Δx-x modΔx+(y+Δy-ymodΔy)X]r_xr_y＝0

Merging and writing the interpolation equations into a homogeneous linear equation set, and converting the homogeneous linear equation set into a matrix form:

HG＝O

from this matrix form, a constraint relation matrix H is derived.

Step 204, merging the constraint relation matrix H into the original ECT sensitive field S to obtain a new sensitive field

And step 205, bringing the new sensitive field A and the normalized capacitance lambda into a fusion image reconstruction method to obtain a reconstructed image.

Step three, similarly, using the full field capacitance C_HAnd null field capacitance value C_LApplying a pixel interpolation image reconstruction algorithm to the region to be imaged to obtain a full-field reconstructed image G_HAnd null field reconstructed image G_L。

Step four, reconstructing the full field image G_HAnd null field reconstructed image G_LAnd normalizing the reconstructed image respectively as a maximum value and a minimum value to obtain a normalized reconstructed image G'.

Step five: obtaining a weighted average gray value of the pixels according to the area of each pixel of the normalized reconstructed image G';

the weighted average gray scale value of a pixel is:

wherein m is the number of pixels, G' [ k ]]Is the grey value, A 'of the kth pixel of the normalized reconstructed image G'_kIs the area of the kth pixel and A' is the total area of the image.

And step six, substituting the weighted average gray value of the pixel of the normalized reconstructed image into a gray value-phase content mapping model to obtain a gas-liquid two-phase content value in the pipeline to be measured.

According to the gray value-phase content mapping model, the liquid phase content in the pipeline to be tested is beta, and the gas phase content is 1-beta.