Optimized Kriging interpolation horizontal gas-liquid two-phase flow elastic flow visualization method
Through the optimization of Kriging interpolation method, the interpolation calculation is performed using the optimized distance factor and spatial correlation coefficient, the problem of monitoring gas phase irregular distribution in horizontal gas-liquid two-phase flowing bullet flow is solved, and more accurate visualization of fluid distribution is achieved.
Patent Information
- Application Number
- CN202411061249.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-19
AI Technical Summary
The existing interpolation method is difficult to accurately monitor and visualize the irregular distribution of the gas phase in horizontal gas-liquid two-phase flowing bullet flow, resulting in difficulty in obtaining fluid distribution information.
The optimized Krigin interpolation method is adopted to obtain flow field information through array sensors, and the optimization distance factor and spatial correlation coefficient suitable for gas-liquid two-phase flow are designed, and the weight coefficient is calculated for interpolation calculation is finally realized to visually image the flow field distribution.
The monitoring accuracy and visualization effect of the two-phase gas-liquid flow in the horizontal tube is improved, and the poor visualization effect of traditional methods when the gas phase is continuous and irregularly distributed is overcome.
Smart Images

Figure CN118981982A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of horizontal gas-liquid two-phase flow visualization, in particular to an optimized Kriging interpolation horizontal gas-liquid two-phase flow slug flow visualization method. Background Art
[0002] Slug flow is a common flow pattern in horizontal gas-liquid two-phase flow, in which the gas and liquid phases appear alternately in the form of large slugs and liquid slugs within a horizontal pipe. However, due to factors such as flow velocity, pressure, and gravity, the gas phase distribution often exhibits irregular shapes, random sizes, and distributions, posing a significant challenge to obtaining and visualizing fluid distribution information within the wellbore.
[0003] The randomness and complexity of slug flow make accurately obtaining the fluid distribution of a pipe diameter cross section a complex task. Currently, the commonly used method is to use acoustic, optical, and electrical sensor technologies to obtain fluid distribution information of the pipe diameter cross section. However, in order to minimize interference with the fluid flow in the pipeline, in practical applications, these sensors often have certain measurement blind spots. Interpolation methods are often used to supplement this missing information to achieve comprehensive monitoring of the pipe diameter cross section. Therefore, the choice of interpolation method is crucial to improving the accuracy and comprehensiveness of fluid flow data. Most of the spatial interpolation algorithms proposed so far mainly consider the prediction of the distribution of spatial resources and are less applicable to the interpolation of wellbore cross sections. Therefore, considering the factors affecting the visualization of slug flow in horizontal gas-liquid two-phase flow, it is urgent to study a visualization interpolation method suitable for irregular distribution of the gas phase. Summary of the Invention
[0004] The technical problem to be solved by this invention is to provide a method for visualizing horizontal gas-liquid two-phase flow slugs using optimized Kriging interpolation. This method monitors the phase distribution of gas-liquid two-phase flow slugs within a horizontal pipe, comprehensively considering the phase distribution characteristics of horizontal gas-liquid two-phase flow slugs to accurately monitor and visualize the fluid distribution.
[0005] To solve the above technical problems, the present invention adopts a technical solution: a method for visualizing slug flow of horizontal gas-liquid two-phase flow using optimized Kriging interpolation, comprising the following steps:
[0006] Step 1: Use array sensors to obtain the flow field information of the horizontal gas-liquid two-phase flow slug;
[0007] Step 2: Normalize and preprocess the acquired flow field information and integrate it into known pixels;
[0008] Step 3: Design the optimal distance factor d for slug flow kriging interpolation of horizontal gas-liquid two-phase flow ij , and use this to calculate the distance between known pixels;
[0009] Step 4: Design the spatial correlation coefficient γ for horizontal gas-liquid two-phase flow slug flow kriging interpolation, and use it to calculate the spatial correlation coefficient value γ between unknown pixels and known pixels. i0 and γ ij ;
[0010] Step 5: Use the spatial correlation coefficient γ i0 and γ ij Determine the weight coefficient λ of each known pixel point on the cross section to the unknown pixel point i ;
[0011] Step 6: Use weight coefficient λ i Interpolation calculation of unknown pixel points of horizontal gas-liquid two-phase flow slug flow;
[0012] Step 7: Visualize the cross-sectional flow field distribution using the interpolation results.
[0013] The further improvement of the technical solution of the present invention is that: Step 3 is specifically as follows: fully consider the effects of gravity and liquid viscosity factors, the gas and liquid phases have a tendency to separate in the tube, the influence of the distance in the horizontal direction is greater than that in the vertical direction, the viscosity of the fluid will cause adhesion to the tube wall, and the distance factor d is designed to be optimized. ij Specifically:
[0014]
[0015] In the formula, (x i ,y i ) and (x j ,y j ) are the coordinates of two known points or one unknown point and one known point, (x c ,y c ) is the coordinate of the center of the measuring tube section, R is the radius of the tube section, λ1 and λ2 are the distance weight coefficients;
[0016] By simulating slug flow, images of partial wellbore cross-sections were captured. The state of each sensor in the array was determined based on each image. Distance weight coefficients λ1 and λ2 of varying proportions were selected to verify the degree of optimization of the distance factor on the pipe wall and in the horizontal and vertical directions. Simple Kriging interpolation imaging was performed. The structural similarity coefficient (SSIM) was used to evaluate the image restoration effect of different weight coefficients and optimize the weight coefficients. The SSIM calculation method is as follows:
[0017]
[0018] Where μ x and μ y are the mean values of the data of the two images in the window, σ x and σy are the variances of the data of the two images in the window, σ xy It represents the covariance of the data of the two images in the window, and L is the dynamic range of the pixel value.
[0019] A further improvement of the technical solution of the present invention is that: Step 4 specifically comprises the following steps: considering that any point in the pipeline measurement section must have a certain correlation with a certain known point, and the point between two adjacent known points should have a high correlation with these two points, the spatial correlation coefficient γ is designed using the empirical correlation coefficient, wherein the applied empirical correlation coefficient is:
[0020]
[0021] Where s i is the sample point number, N is the acceptance sample point (s i ,s j ), δh is the sample point span, and after iterating the sensor data using the empirical correlation coefficient, the correlation coefficient γ is determined as follows:
[0022]
[0023] Where r is the variable range, which can be determined by fitting the actual data obtained by sensors under different working conditions.
[0024] The further improvement of the technical solution of the present invention is that: Step 5 is specifically: using formula (1) to calculate the optimized distance factor d between the known pixel point and the unknown pixel point i0 and the optimized distance factor d between two known pixels ij Then, we respectively substitute the fitted formula (4) to obtain γ i0 and γ ij , the spatial correlation coefficient γ between known points ij and the spatial correlation coefficient γ between unknown points and known points i0 Substitute the following matrix form equations to solve and obtain the weight coefficient λ i
[0025]
[0026] The further improvement of the technical solution of the present invention is that: Step 6 specifically comprises the following steps: using the weight coefficient λ i and the corresponding known pixel value Z i Substitute the following equation to interpolate the unknown pixels:
[0027]
[0028] Due to the adoption of the above technical solution, the technical progress achieved by the present invention is: optimizing the distance factor d by creative design ij The spatial correlation coefficient γ can effectively address the problem of poor visualization of gas-liquid two-phase flow in horizontal pipe sections with continuous and irregular gas distribution using traditional interpolation imaging methods. This makes the interpolation calculation more accurate and the final visualization better. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can derive other drawings based on these drawings without inventive effort.
[0030] Figure 1 It is a schematic flow diagram of the method of the present invention;
[0031] Figure 2 1. It is a schematic diagram of the structure of an array sensor for obtaining the flow field information of a horizontal gas-liquid two-phase flow slug;
[0032] Figure 3 It is a schematic diagram of a cross-section of a wellbore;
[0033] Figure 4 It outputs a visualization effect diagram based on the interpolation results. DETAILED DESCRIPTION
[0034] The present invention is described in further detail below in conjunction with the embodiments:
[0035] like Figure 1 The figure shows a flow chart of a slug flow visualization method for optimizing Kriging interpolation of horizontal gas-liquid two-phase flow. The specific steps are as follows:
[0036] Step 1: If Figure 2 As shown, the array sensor is used to obtain the flow field information of the horizontal gas-liquid two-phase flow slug;
[0037] Step 2: Normalize and preprocess the acquired flow field information and integrate it into known pixels;
[0038] Step 3: Design the optimal distance factor d for slug flow kriging interpolation of horizontal gas-liquid two-phase flow ij , and use this to calculate the distance between known pixels;
[0039] Taking full account of the effects of gravity and liquid viscosity, the gas and liquid phases tend to separate in the pipe. The influence of distance in the horizontal direction is greater than that in the vertical direction. The viscosity of the fluid will cause adhesion to the pipe wall. The distance factor d is designed to be optimized.ij Specifically:
[0040]
[0041] In the formula, (x i ,y i ) and (x j ,y j ) are the coordinates of two known points or one unknown point and one known point, (x c ,y c ) is the coordinate of the center of the measuring tube section, R is the radius of the tube section, λ1 and λ2 are the distance weight coefficients;
[0042] By simulating the slug flow, a partial cross-section of the wellbore is captured, such as Figure 3 As shown in the figure, the state of each sensor in the array sensor is determined according to each image. The optimization degree of the distance factor in the pipe wall and horizontal and vertical directions is verified by selecting distance weight coefficients λ1 and λ2 of different proportions, and simple Kriging interpolation imaging is performed. At the same time, the structural similarity coefficient SSIM is used to evaluate the image restoration effect of different weight coefficients. The SSIM calculation method is as follows:
[0043]
[0044] Where μ x and μ y are the mean values of the data of the two images in the window, σ x and σ y are the variances of the data of the two images in the window, σ xy represents the covariance of the data of the two images in the window, and L is the dynamic range of the pixel value. After optimization calculation in this embodiment, it is known that the optimal values are λ1=2, λ2=0.5
[0045] Step 4: Design the spatial correlation coefficient γ for horizontal gas-liquid two-phase flow slug flow kriging interpolation, and use it to calculate the spatial correlation coefficient value γ between unknown pixels and known pixels. i0 and γ ij ;
[0046] By considering that any point in the pipeline measurement section must have a certain correlation with a known point, and the point between two adjacent known points should have a high correlation with these two points, the spatial correlation coefficient γ is designed using the empirical correlation coefficient, where the empirical correlation coefficient is:
[0047]
[0048] Where s i is the sample point number, N is the acceptance sample point (s i,s j ), δh is the sample point span,
[0049] After iterating the sensor data using the empirical correlation coefficient, the correlation coefficient γ is determined as follows:
[0050]
[0051] Where r is the variable range, which can be determined by fitting the actual data obtained by sensors under different working conditions.
[0052] Step 5: Use the spatial correlation coefficient γ i0 and γ ij Determine the weight coefficient λ of each known pixel point on the cross section to the unknown pixel point i ; Use formula (1) to calculate the optimized distance factor d between known pixels and unknown pixels i0 and the optimized distance factor d between two known pixels ij Then, we respectively substitute the fitted formula (4) to obtain γ i0 and γ ij ; The spatial correlation coefficient γ between known points ij and the spatial correlation coefficient γ between unknown points and known points i0 Substitute the following matrix form equations to solve and obtain the weight coefficient λ i
[0053]
[0054] Step 6: Use weight coefficient λ i Interpolation calculation of unknown pixel points of horizontal gas-liquid two-phase flow slug flow;
[0055] Using the weight coefficient λ i and the corresponding known pixel value Z i Substitute the following equation to interpolate the unknown pixels:
[0056]
[0057] Step 7: Visualize the cross-sectional flow field distribution through interpolation results, and output the visualization effect according to the interpolation results. Figure 4 shown.
[0058] The embodiments described above are merely descriptions of preferred implementations of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present invention.
Claims
1. A method for visualizing slug flow of horizontal gas-liquid two-phase flow using optimized Kriging interpolation, characterized by: The steps include: Step 1: Use array sensors to obtain the flow field information of the horizontal gas-liquid two-phase slug flow; Step 2: normalize and preprocess the acquired flow field information and integrate it into known pixels; Step 3: Design the optimal distance factor d for slug flow kriging interpolation of horizontal gas-liquid two-phase flow ij , and use this to calculate the distance between known pixel points; Step 4: Design the spatial correlation coefficient γ for horizontal gas-liquid two-phase flow slug flow kriging interpolation, and use it to calculate the spatial correlation coefficient value γ between unknown pixels and known pixels. i0 and γ ij ; Step 5: Use the spatial correlation coefficient γ i0 and γ ij Determine the weight coefficient λ of each known pixel point on the cross section to the unknown pixel point i ; Step 6: Use weight coefficient λ i Interpolation calculation of unknown pixel points of horizontal gas-liquid two-phase slug flow; Step 7: Visualize the cross-sectional flow field distribution through interpolation results.
2. The optimized Kriging interpolation horizontal gas-liquid two-phase flow slug flow visualization method according to claim 1, characterized in that: Step 3: The specific steps are: fully consider the effects of gravity and liquid viscosity. The gas and liquid phases tend to separate in the tube. The influence of distance in the horizontal direction is greater than that in the vertical direction. The viscosity of the fluid will cause adhesion to the tube wall. Design and optimize the distance factor d. ij Specifically: In the formula, (x i ,y i ) and (x j ,y j ) are the coordinates of two known points or one unknown point and one known point, (x c ,y c ) is the coordinate of the center of the measuring tube section, R is the radius of the tube section, λ1, λ2 are the distance weight coefficients; By simulating the flow of elastic flow, we intercepted some cross-section images of the wellbore, determined the state of each sensor in the array sensor based on each image, verified the optimization degree of the distance factor in the pipe wall and horizontal and vertical directions by selecting distance weight coefficients λ1 and λ2 of different proportions, and performed simple Kriging interpolation imaging. At the same time, we used the structural similarity coefficient SSIM to evaluate the image restoration effect of different weight coefficients and optimized the weight coefficients. SSIM is calculated as follows: In the formula, μ x and μ y are the mean values of the data of the two images in the window, σ x and σ y are the variances of the data of the two images in the window, σ xy It represents the covariance of the data of the two images in the window, and L is the dynamic range of the pixel value.
3. The optimized Kriging interpolation horizontal gas-liquid two-phase flow slug flow visualization method according to claim 1, characterized in that: The specific steps of step 4 are as follows: by considering that any point in the pipeline measurement section must have a certain correlation with a known point, and the point between two adjacent known points should have a high correlation with these two points, the spatial correlation coefficient γ is designed using the empirical correlation coefficient, where the applied empirical correlation coefficient is: In the formula, s i is the sample point number, N is the acceptance sample point (s i ,s j ), δh is the sample point span, and after iterating the sensor data using the empirical correlation coefficient, the correlation coefficient γ is determined as follows: In the formula, r is the variable range, which can be determined by fitting the actual data obtained by sensors under different working conditions.
4. The optimized Kriging interpolation horizontal gas-liquid two-phase flow slug flow visualization method according to claim 1, characterized in that: Step 5 is as follows: Calculate the optimal distance factor d between the known pixel and the unknown pixel using formula (1): i0 and the optimized distance factor d between two known pixels ij Then, we substitute the fitted formula (4) to obtain γ i0 and γ ij ; The spatial correlation coefficient γ between known points ij and the spatial correlation coefficient γ between unknown points and known points i0 Substitute the following matrix form equations into the equations and solve them to obtain the weight coefficient λ i 5. The optimized Kriging interpolation horizontal gas-liquid two-phase flow slug flow visualization method according to claim 1, characterized in that: Step 6 The specific steps are: Use the weight coefficient λ i and the corresponding known pixel value Z i Substitute the following equation to interpolate unknown pixels:
