A water supply pipeline leakage diagnosis method considering uncertainty

Abstract

The present application relates to a kind of uncertainty water supply pipeline leakage diagnosis methods based on leakage before and after radar information and image correlation coefficient, belong to water supply pipeline leakage diagnosis technical field.The method includes: S1: it is demarcated survey line at certain interval perpendicular to water supply pipeline, under the condition of not leaking, the radar echo image of each survey line is detected and obtained;S2: the radar echo image of each survey line is repeatedly obtained under the condition of leaking;S3: all radar echo images are cropped according to uniform pixel size;S4: the correlation coefficient of all leakage state images and same survey line non-leakage state image is calculated;S5: the distribution of correlation coefficient under the condition of leakage is inspected, specifically: the correlation coefficient of leakage state image and non-leakage state image of each test group of same survey line is inspected, and the difference is described according to distribution form using mean or median description.The present application can improve the accuracy of buried water supply pipeline leak detection.

Description

Technical Field

[0001] This invention belongs to the field of water supply pipeline leakage diagnosis technology, and relates to the identification of pipeline leakage locations when ground-penetrating radar information before and after leakage is known. Specifically, it relates to an uncertain water supply pipeline leakage diagnosis method based on radar information before and after leakage and image correlation coefficient. Background Technology

[0002] With social development, the scale of water supply pipelines has continued to expand. However, due to being buried underground for many years, the problem of pipeline leakage has become increasingly prominent. The difficulty in locating leaks makes maintenance and repair difficult. Pipeline leakage not only wastes water resources and pollutes the environment, but also poses safety hazards.

[0003] Existing pipeline leakage detection methods are mainly divided into direct detection methods and indirect detection methods. Direct detection methods utilize various facilities and equipment to directly detect whether the pipeline itself is damaged. The most commonly used methods include endoscopic robot inspection technology and pressure gauge inspection technology. This type of inspection requires deep penetration into the pipeline, and may necessitate interrupting normal pipeline operation. Indirect detection methods utilize the results of pipeline leakage to determine the pipeline's condition. These methods do not require entry into the pipeline and have fewer environmental restrictions. They mainly include leakage noise detection technology, transient electromagnetic detection technology, and ground-penetrating radar (GPR) detection technology. Leakage noise detection technology determines the degree and location of leakage based on the sound waves emitted by a leaking pipeline that differ from those emitted by a normal pipeline. This method is easily affected by ambient noise in urban environments, resulting in poor reliability. Transient electromagnetic technology primarily identifies pipeline defects by detecting the electrical differences between buried pipelines and the surrounding soil layers. It is mainly used for oil and gas metal pipelines and is ineffective for non-metallic pipelines in the water supply sector. Ground-penetrating radar (GPR) technology is a non-destructive geophysical exploration method with advantages such as speed, non-destructive nature, and resistance to interference, and is widely used in civil engineering inspection.

[0004] Ground-penetrating radar (GPR) for detecting leaks in water supply pipelines has been developed in recent years. It allows for detection without interrupting normal pipeline operation or requiring deep excavation, enabling damage detection on road surfaces and making it highly adaptable to urban environments. However, current identification of pipeline leak radar images relies on the experience of technical personnel, lacking objective methods and making it difficult to accurately pinpoint the location of leaks. Identifying leak images is a crucial step in locating leaks and directly relates to the accurate localization of leaks in the pipeline network. Therefore, it is necessary to establish a pipeline leak diagnosis method to achieve objective and accurate location of leaks in the pipeline network. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide an uncertain water supply pipeline leakage diagnosis method based on radar information and image correlation coefficients before and after leakage. Addressing the current situation where radar image identification of leaking pipelines relies on the experience of technicians, resulting in strong subjectivity and insufficient reliability in locating pipeline leaks, this invention employs a pipeline leakage diagnosis method based on image correlation theory. Using known non-leaking pipeline images as a benchmark, the correlation coefficient between leaking and non-leaking images along the same measurement line is calculated. This correlation coefficient serves as an objective basis for diagnosing the location of pipeline leaks. Furthermore, based on statistical analysis, a distribution test is conducted using correlation coefficient indicators from multiple detection images to reduce the impact of uncertainties such as testing and calculation errors, thereby improving the reliability of water supply pipeline leak location.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An uncertain water supply pipeline leakage diagnosis method based on radar information and image correlation coefficient before and after leakage is proposed. The method uses radar echo images of water supply pipelines with and without leakage as calculation targets. Based on the basic principle of image correlation, the correlation coefficient between the leakage image and the non-leakage image is calculated, and the correlation coefficient is used as a digital standard for diagnosing pipeline leakage.

[0008] The method specifically includes the following steps:

[0009] S1: Draw test lines perpendicular to the water supply pipeline at certain intervals (based on the original records of pipeline construction, for large-diameter pipelines with large water flow, the leakage coverage area may be wider, so test lines can be laid out more sparsely; for small-diameter pipelines with small water flow, the leakage coverage area is relatively smaller, so more dense test lines need to be drawn to avoid missing detection), and detect and obtain the radar echo image of each test line in the non-leakage state, referred to as the non-leakage state image;

[0010] S2: Repeatedly acquire radar echo images of each survey line under leakage conditions, referred to as leakage condition images;

[0011] S3: Crop all radar echo images to a uniform pixel size;

[0012] S4: Calculate the correlation coefficients between all leakage status images and non-leakage status images on the same test line;

[0013] S5: Examine the distribution of correlation coefficients under leakage conditions. Specifically, perform a distribution test on the correlation coefficients between the leakage condition images and the non-leakage condition images of each test group on the same test line, and determine whether to use the mean or the median to describe the difference based on the distribution pattern.

[0014] Furthermore, in step S1 or S2, the radar echo image is a grayscale image. The same radar parameters and sampling rate are used during the image acquisition process. During the detection process, calibration and distance uniformization can be used in combination to make the image size as consistent as possible. The image is in JPG format.

[0015] Further, step S4 specifically includes: using the non-leaking image as a reference, calculating the correlation coefficient between the leaking image and the non-leaking image along the same measurement line; arbitrarily selecting a pixel point P(x0,y0) in the reference image, and selecting a rectangular region of (2m+1)×(2m+1) centered on this point as a reference subset, where M is the radius of this reference subset; then similarly selecting a calculation point P'(x0',y0') in the target image, and selecting a target subset centered on this point, with the same size as the reference subset; the correlation coefficient C between the two is:

[0016]

[0017] in, f(x0,y0) is the gray value of the selected pixel P(x0,y0) in the reference image; m The reference subset represents the average grayscale value; g(x0',y0') represents the grayscale value of pixel P'(x0',y0') in the target image; g m The average gray value of the target subset.

[0018] Due to the localized nature of the leakage, the following two conclusions can be drawn based on the correlation coefficient:

[0019] 1) Areas far from the leak point are not affected by the leakage. Even after the pipeline leaks, the radar echo image is still an image of the non-leaking state. That is, the image features before and after the pipeline leak are consistent and have good correlation.

[0020] 2) The radar echo image characteristics of the test line near the leak point change due to the influence of pipeline leakage. The correlation between the leak image and the non-leak image of the same test line decreases. Therefore, the range of the test line where the leak point is located can be determined by the correlation coefficient of the radar echo images before and after the leakage.

[0021] Furthermore, in step S5, based on the fundamental theories of statistical analysis and combined with the distribution test of correlation indicators from multiple image detections, a probabilistic or non-probabilistic uncertainty-based method for diagnosing water supply network leakage is proposed to optimize the above diagnostic results. By detecting leakage images multiple times along the same test line, the correlation coefficients with images in a non-leakage state are calculated, and the distribution of these correlation coefficients is tested. Based on the distribution pattern, the mean or median is used to describe the difference. This reduces the impact of uncertainties such as testing and calculation errors, improving the accuracy of buried water supply pipeline detection.

[0022] Based on the distribution pattern, determine whether to use the mean or median to describe the differences, specifically including:

[0023] 1) If it is a normal distribution, when the correlation coefficient of a certain measurement line is significantly lower than the population mean, the area of ​​that measurement line is determined to be the area where the pipeline leak is located;

[0024] 2) If the distribution is non-normal, when the correlation coefficient of a certain test line is significantly lower than the median of the population, the area of ​​the test line is determined to be the area where the pipeline leak is located.

[0025] The beneficial effects of this invention are as follows: This invention can effectively identify leaking pipeline images and locate pipeline leaks. By calculating the correlation with images of non-leaking pipelines along the same survey line, it uses the correlation coefficient as an objective standard for determining the location of pipeline leaks, thus overcoming the current reliance on the experience of technical personnel for radar image identification. Simultaneously, by combining the distribution verification of correlation indicators from multiple detection images, it effectively reduces the adverse effects of uncertainties such as testing and calculation errors on the diagnostic results, improving the accuracy of leak detection in buried water supply pipelines.

[0026] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0027] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0028] Figure 1 This is a flowchart of the water supply pipeline leakage diagnosis method based on the uncertainty of radar information and image correlation coefficient before and after leakage according to the present invention.

[0029] Figure 2 This is a schematic diagram of the survey line division;

[0030] Figure 3 A schematic diagram illustrating the standardization of pixel size in grayscale images. Detailed Implementation

[0031] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0032] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0033] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0034] Please see Figures 1-3 This invention provides a method for diagnosing uncertain water supply pipeline leaks based on the correlation coefficient of radar information and images before and after leakage. The method uses ground-penetrating radar grayscale images before and after the leak to calculate the correlation coefficient to determine the location of the leak in the water supply network, and combines multiple detection images to accurately identify the leak location through the distribution test of correlation indicators. The method includes the following steps:

[0035] Step 1: Draw test lines perpendicular to the water supply pipeline at certain intervals (for pipes with a diameter of DN300 or less, the test line spacing should be less than 10m; for pipes with a diameter of DN500 or more, the test line spacing can be 10-20m; the specific test line spacing must be considered comprehensively based on the pipe diameter, pipeline pressure, and site environment)). Obtain radar echo images for each test line under leak-free conditions. The radar echo images are grayscale images. The same radar parameters and sampling rate are used during image acquisition. Calibration and distance uniformization can be used in conjunction with the detection process to ensure consistent image size. Images are in JPG format.

[0036] Step 2: Repeatedly acquire radar images of each survey line under the leakage condition, and all images are in JPG format.

[0037] Step 3: Crop all radar grayscale images to a uniform pixel size.

[0038] Step 4: Calculate the correlation coefficient between all leakage images and non-leakage images along the same survey line. Using the non-leakage images as a baseline, calculate the correlation coefficient between leakage status images and non-leakage images along the same survey line. In the baseline image, arbitrarily select a pixel point P(x0,y0), and select a rectangular region of (2m+1)×(2m+1) centered on this point as a reference subset, where M is the radius of this reference subset. Then, similarly select a calculation point P'(x0',y0') on the target image, and select a target subset centered on this point, with the same size as the reference subset. The correlation coefficient C between the two is:

[0039]

[0040] in, f(x0,y0) is the gray value of the selected pixel P(x0,y0) in the reference image; m The reference subset represents the average grayscale value; g(x0',y0') represents the grayscale value of pixel P'(x0',y0') in the target image; g m The average gray value of the target subset.

[0041] Step 5: Examine the distribution of correlation coefficients under leakage conditions. Perform a distribution test on the correlation coefficients between the leakage condition images and the non-leakage images of each test group on the same test line, and determine whether to use the mean or the median to describe the differences based on the distribution pattern.

[0042] 1) If it is a normal distribution, when the correlation coefficient of a certain test line is significantly lower than the population mean, the test line area is determined to be the area where the pipeline leak is located.

[0043] 2) If the distribution is non-normal, when the correlation coefficient of a certain test line is significantly lower than the median of the population, the area of ​​the test line is determined to be the area where the pipeline leak is located.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for diagnosing water supply pipeline leakage considering uncertainties, characterized in that, The method specifically includes the following steps: S1: Draw measuring lines perpendicular to the water supply pipeline at intervals, and detect and acquire radar echo images of each measuring line under non-leakage conditions, referred to as non-leakage condition images; S2: Repeatedly acquire radar echo images of each survey line under leakage conditions, referred to as leakage condition images; S3: Crop all radar echo images to a uniform pixel size; S4: Calculate the correlation coefficients between all leakage status images and non-leakage status images along the same survey line. Specifically, this includes: using the non-leakage status image as a reference, calculating the correlation coefficients between the leakage status images and the non-leakage status images along the same survey line; arbitrarily selecting a pixel point at coordinates in the reference image. And select with that point as the center The rectangular region is the reference subset. M Therefore, referencing the radius of the subset, calculation points are then selected similarly on the target image. A target subset, centered at this point and with the same size as the reference subset, is selected; the correlation coefficient between the two is... C for: in, , ; For the pixels selected in the reference image grayscale value; The average grayscale value of the reference subset; For target image pixels grayscale value; The average gray value of the target subset; S5: Examine the distribution of correlation coefficients under leakage conditions. Specifically, perform a distribution test on the correlation coefficients of leakage state images and non-leakage state images of each test group on the same test line. Determine whether to use the mean or median to describe the difference based on the distribution pattern: If it is a normal distribution, when the correlation coefficient of a certain test line is significantly lower than the population mean, then the area of ​​that test line is determined to be the area where the pipeline leaks; if it is a non-normal distribution, when the correlation coefficient of a certain test line is significantly lower than the population median, then the area of ​​that test line is determined to be the area where the pipeline leaks.

2. The method for diagnosing water supply pipeline leakage according to claim 1, characterized in that, In step S1 or S2, the radar echo image is a grayscale image. The same radar parameters and sampling rate are used during the image acquisition process. During the detection process, calibration and distance uniformization are used in combination to make the image size consistent. The image is in JPG format.

3. The method for diagnosing water supply pipeline leakage according to claim 1, characterized in that, In step S4, due to the locality of the leakage, the following conclusions can be drawn based on the correlation coefficient: 1) In areas far from the leak point, the image features before and after the pipe leakage are consistent and have good correlation; 2) Near the leak point, the characteristics of radar echo images before and after leakage change, and the correlation between leakage images and non-leakage images on the same survey line decreases. The range of the survey line where the leak point is located is determined by the correlation coefficient of radar echo images before and after leakage.

Images