Flow field visualization method and device based on underground earthing tank leakage

Abstract

The invention discloses a flow field visualization method and device based on underground earthing tank leakage, belongs to the technical field of underground earthing tank safety monitoring, and aims to solve the problems that safety inspection of an existing underground earthing tank is difficult, visual monitoring cannot be performed after earthing is completed, and the safety of the underground earthing tank is poor. And the technical problem that the visual state after leakage is difficult to accurately monitor is solved. The method comprises the following steps: carrying out image acquisition processing related to laser scatter points on an overground target flow field area of an underground soil covering tank to obtain a leakage disturbance image; performing de-noising processing related to pepper grain noise on the leakage disturbance image to obtain a pre-processed leakage disturbance image; performing laser scatter disturbance displacement calculation based on an overground target flow field area on the preprocessed leakage disturbance image to obtain disturbance image displacement field information at the current moment; performing triple noise correction calculation on the disturbance image displacement field information to obtain real displacement field information; and based on the real displacement field information, leakage flow field information at the current moment is determined.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of underground buried tank safety monitoring, and particularly relates to a flow field visualization method and device based on underground buried tank leakage. BACKGROUND

[0002] In daily life and industrial fields, buried tanks have been widely used in developed countries in Europe. Germany first built 30 buried tanks with a capacity of 200 m3 in 1959, and currently, more than 500 buried tank groups have been built by foreign oil companies. Based on safety, Germany has adopted buried tanks for storing liquefied petroleum gas with a capacity of more than 3 tons since 1991, and prohibited the construction of above-ground spherical tanks with a capacity of more than 1000 m3 in 1999. France prohibited the construction of above-ground spherical tanks with a capacity of more than 500 m3 in 2000. Most domestic enterprises use above-ground spherical tanks (with a maximum storage capacity of 4000 m3), and the domestic and foreign capital enterprise BASF has application cases of buried tanks. Buried tanks have been valued and gradually applied in China due to their advantages such as design conditions not being affected by atmospheric temperature and pressure changes, operation conditions being related to only medium upstream operation conditions and low temperature, saving land area, reducing safety distance from adjacent facilities, and being less affected by external fires.

[0003] Although buried tanks have many advantages compared with above-ground tanks, they bring new challenges to safety detection and equipment maintenance. In particular, it should be noted that, due to direct contact with soil on the outside, buried tanks are more prone to corrosion than above-ground tanks, and are not easy to be regularly inspected and detected. After the completion of burial, visual monitoring cannot be performed, which brings great difficulty to the maintenance of the tank. Currently, the regular inspection period of buried tanks generally refers to the regulations of Germany, which is 8-10 years. Between the longer inspection periods, the tank may be seriously corroded to cause tank rupture, and the high-pressure gas (for example, propylene, with a design operating pressure of 1.5-1.7 MPa) stored in the tank may leak.

[0004] Although no safety accidents have occurred in buried tanks since they were put into operation in the 1950s, it is still very important to study the safety of buried tanks. Currently, the safety research of buried tanks mainly focuses on the following aspects: safety distance between buried tanks and adjacent facilities and fire-fighting facilities, external corrosion of buried tanks, arrangement of gas leakage detection devices, and operation pressure of buried tanks. However, there are relatively few studies on the leakage form parameters and evolution of the leaked gas after the tank rupture. SUMMARY

[0005] The embodiment of the present application provides a flow field visualization method and device based on underground earth-covered tank leakage, and is used for solving the following technical problem: the existing underground earth-covered tank safety inspection is difficult, visual monitoring cannot be performed after earth covering is completed, and accurate monitoring of the visualized state after leakage is difficult to implement.

[0006] The embodiment of the present application adopts the following technical scheme:

[0007] In one aspect, the embodiment of the present application provides a flow field visualization method based on underground earth-covered tank leakage, comprising: through combined use of a laser device and an industrial camera, performing image acquisition and processing on a target flow field area of the underground earth-covered tank in relation to laser scattering points, to obtain a leakage disturbance image; performing noise removal processing on the leakage disturbance image in relation to pepper grain noise, to obtain a pretreated leakage disturbance image; through a cross-correlation algorithm of a multi-scale window and a fast Fourier transform, and based on a pre-acquired no-leakage reference image, performing laser scattering point disturbance displacement calculation based on the target flow field area on the pretreated leakage disturbance image, to obtain disturbance image displacement field information at a current time; performing triple noise correction calculation on the disturbance image displacement field information, to obtain real displacement field information; wherein the triple noise correction comprises: error data correction, mechanical noise correction and background noise correction; based on the real displacement field information, leakage flow field information at the current time is determined.

[0008] The flow field visualization detection of the underground earth-covered tank leakage can restore the actual underground earth-covered tank leakage situation, the leakage situation under different buried depths and leakage hole diameters of the underground earth-covered tank structure in the actual project is simulated through setting of a metal tank, different levels of soil materials and adjustable pressure and orifice area, the flow field change when the underground earth-covered tank leaks can be reproduced more truly. The disturbance in the flow field can be captured and accurately analyzed through a schlieren method and an image processing technology, high-precision flow field displacement data are obtained, so that the flow field generated by the leakage is visualized. Moreover, accurate displacement field data are obtained through image data post-processing algorithms, so that the evolution process of the flow field can be tracked and analyzed in detail. Meanwhile, the images obtained in the experimental process can be processed in real time, the displacement field data of the flow field are obtained, dynamic flow field change images are provided for the experiment, and this is helpful for in-depth analysis of the behavior of the fluid.

[0009] In an embodiment, the image acquisition and processing of the laser scattering points related to the ground target flow field area of the underground buried tank is performed by the combination of the laser device and the industrial camera, and a leakage disturbance image is obtained, which specifically includes: controlling the underground buried tank to simulate leakage; irradiating the ground target flow field area after the leakage simulation with laser scattering points by the laser device to determine a laser irradiation scattering point area; performing global image acquisition and processing on the laser irradiation scattering point area and the ground target flow field area by the adjusted focal length parameter and distance parameter of the industrial camera to obtain the leakage disturbance image.

[0010] In an embodiment, the noise removal processing of the leakage disturbance image is performed to obtain a preprocessed leakage disturbance image, which specifically includes: performing brightness enhancement processing on the laser scattering spot in the leakage disturbance image by a preset CLAHE algorithm, and obtaining a first leakage disturbance image based on the contrast of other background information in the leakage disturbance image; performing filtering processing on the pepper noise in the first leakage disturbance image by a Gaussian filtering algorithm to obtain the preprocessed leakage disturbance image.

[0011] In an embodiment, the laser scattering point disturbance displacement calculation based on the ground target flow field area is performed on the preprocessed leakage disturbance image by the cross-correlation algorithm of the multi-scale window and the fast Fourier transform and based on the pre-acquired no-leakage reference image to obtain the disturbance image displacement field information at the current time, which specifically includes: combining the no-leakage reference image and the preprocessed leakage disturbance image to obtain a disturbance image group; performing cross-correlation calculation on the current large window data corresponding to the disturbance image group by the fast Fourier transform, and obtaining the sub-pixel level displacement field data of the current large window based on the sub-pixel search algorithm; performing cross-correlation calculation on the disturbance image group at the next level of large window based on the fast Fourier transform according to the offset step of the sub-pixel level displacement data to obtain large window cross-correlation operation data; wherein the next level configuration parameter of the next level of large window is half of the current large window data configuration parameter; performing traversal processing on the large window area in the large window cross-correlation operation data by a preset small window step to obtain small window traversal result data; performing full image area traversal calculation on the preprocessed leakage disturbance image related to the laser scattering point disturbance displacement based on the small window traversal result data and by the large window step to obtain the disturbance image displacement field information at the current time.

[0012] In an implementable embodiment, the large window data corresponding to the group of perturbation images is subjected to cross-correlation calculation by the fast Fourier transform, and the sub-pixel level displacement field data of the current large window is obtained based on a sub-pixel search algorithm, specifically including: performing image gray value calculation on the non-leakage reference image and the pre-processed leakage perturbation image in the group of perturbation images respectively to obtain first large window gray value data of the non-leakage reference image and second large window gray value data of the pre-processed leakage perturbation image; performing fast Fourier transform calculation on the first large window gray value data and the second large window gray value data respectively, and based on point-by-point complex calculation of complex conjugate and inverse Fourier transform, the displacement field data of the current large window is obtained; by the sub-pixel search algorithm, the sub-pixel peak value of the displacement field data of the current large window is searched, and based on the pixel level peak position and the sub-pixel level peak position, the sub-pixel level displacement field data of the current large window is searched and generated.

[0013] In an implementable embodiment, the perturbation image displacement field information is subjected to triple noise correction calculation to obtain true displacement field information, specifically including: performing matrix conversion on the perturbation image displacement field information to obtain displacement field matrix data; by a first preset threshold, the displacement field matrix data is subjected to error data correction calculation under the influence of environmental factors to obtain first corrected displacement field matrix data; by fitting plane data, the first corrected displacement field matrix data is subjected to plane fitting to obtain a mechanical noise correction matrix; and the first corrected displacement field matrix data is subtracted by the mechanical noise correction matrix to obtain second corrected displacement field matrix data; the pre-acquired displacement field data without flow field is subjected to addition and average calculation to obtain a background noise template matrix; the second corrected displacement field matrix data is subtracted by the background noise template matrix to obtain third corrected displacement field matrix data; based on the third corrected displacement field matrix data, the true displacement field information is obtained.

[0014] In an implementable embodiment, before the first corrected displacement field matrix data is subjected to plane fitting by fitting plane data to obtain a mechanical noise correction matrix, the method further includes: selecting non-flow field region data in the first corrected displacement field matrix data; performing region fitting calculation on the non-flow field region data to obtain fitting plane data.

[0015] In an implementable embodiment, based on the real displacement field information, the leakage flow field information at the current time is determined, specifically including: mapping flow field parameters in the real displacement field information into a multi-layer color space; wherein the multi-layer color space includes: a hue layer representing a flow velocity direction, a saturation layer representing a pressure gradient amplitude, and a lightness layer representing a vorticity intensity; through a time slicing technique, a two-dimensional projection rendering of the multi-layer color space under a three-dimensional flow field is performed, to obtain the leakage flow field information at the current time.

[0016] In an implementable embodiment, the no-leakage reference image of the target flow field area on the ground before the leakage simulation is performed is collected by an industrial camera.

[0017] In another aspect, the embodiments of the present application also provide a flow field visualization device based on underground earth-covered tank leakage, including: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions capable of being executed by the at least one processor, so as to enable the at least one processor to execute the flow field visualization method based on underground earth-covered tank leakage according to any one of the above embodiments.

[0018] The embodiments of the present application provide a flow field visualization method and device based on underground earth-covered tank leakage, and compared with the prior art, the embodiments of the present application have the following beneficial technical effects:

[0019] 1. The actual underground earth-covered tank leakage situation can be restored, the leakage situation under different buried depths and leakage hole diameters of the underground earth-covered tank structure in the actual engineering is simulated by setting a metal tank, different levels of soil materials, and adjustable pressure and orifice area, and the flow field change when the underground earth-covered tank leaks can be reproduced more realistically.

[0020] 2. The disturbance in the flow field can be captured and accurately analyzed by the schlieren method and image processing technology, and high-precision flow field displacement data is obtained, so that the flow field generated by the leakage can be visualized.

[0021] 3. Real-time monitoring and data analysis can be realized, the leakage experiment process is monitored in real time by using an image acquisition device, accurate displacement field data is obtained through image data post-processing algorithms, and the evolution process of the flow field can be tracked and analyzed in detail.

[0022] 4. The images obtained during the experiment process can be processed in real time to obtain the displacement field data of the flow field, dynamic flow field change images are provided for the experiment, and the behavior of the fluid can be analyzed in depth. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below only constitute some of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings. In the drawings:

[0024] Figure 1 A flowchart of a flow field visualization method based on underground earth-covered tank leakage provided by an embodiment of the present application;

[0025] Figure 2 A structural diagram of an experimental device for simulating underground earth-covered tank leakage provided by an embodiment of the present application, which comprises: a gas source 11, a metal pipeline 12, a valve 13, a buffer tank 14, a pressure gauge 15, a connecting pipeline 16, a cuboid metal tank 21, soil 22, a porous metal plate 23, a leakage hole 24, a single-lens reflex camera 31, a laser with a grating sheet 32, and a wall 33.

[0026] Figure 3 A flowchart of a process for simulating underground earth-covered tank leakage and visualizing the leakage flow field provided by an embodiment of the present application;

[0027] Figure 4 A comparison diagram of leakage flow fields of normal earth covering and compacted earth covering provided by an embodiment of the present application;

[0028] Figure 5 A diagram of evolution of gas leakage flow field forms provided by an embodiment of the present application;

[0029] Figure 6 A comparison diagram of leakage flow fields under different sand covering thicknesses provided by an embodiment of the present application;

[0030] Figure 7 A structural schematic diagram of a flow field visualization device based on underground earth-covered tank leakage provided by an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some of the embodiments of the present application, and not all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0032] It should be noted that, Figure 2 A structural diagram of an experimental device for simulating underground earth-covered tank leakage provided by an embodiment of the present application, which comprises: a gas source 11, a metal pipeline 12, a valve 13, a buffer tank 14, a pressure gauge 15, a connecting pipeline 16, a cuboid metal tank 21, soil 22, a porous metal plate 23, a leakage hole 24, a single-lens reflex camera 31, a laser with a grating sheet 32, and a wall 33.Figure 2 As shown in the underground soil-covered tank leakage simulation and flow field visualization experimental device, including the upper end opening cubic metal tank (21), metal tank 1 m x 1 m x 0.8 m, the bottom of the metal tank is provided with a diameter of 25 mm leakage hole (24), the hole is connected with the gas supply system through the pipeline (16) with a diameter of 25 mm. The gas supply system mainly includes gas source (11), buffer tank (14) and pipeline (16) for connection, valve (13) is composed. The gas source uses carbon dioxide cylinder, the cylinder is connected with the 120L buffer tank (14) through the metal pipeline (12) with a diameter of 6mm, the buffer tank is provided with a pressure gauge (15) with a range of 0-2.5MPa, the buffer tank downstream is connected with the hole at the bottom of the metal tank through the above-mentioned 25mm pipeline with valve. The inside of the metal tank is sequentially provided with 0.2m sand and 0.6m soil (22) from bottom to top, the operating pressure of the buffer tank is set to 1.5MPa to simulate the design pressure of about 1.5MPa of the actual propylene soil-covered tank. In the metal tank, a plurality of perforated metal plates (23) are arranged every 0.2m, the 6mm diameter holes are closely distributed, and the 25mm holes are placed with closely arranged metal mesh (24). Two groups of experiments are carried out, which are the influence of loose soil / compacted soil on the leakage of soil-covered tank and the influence of sand thickness on the leakage of soil-covered tank.

[0033] Figure 3 A flow chart for simulating underground soil-covered tank leakage and visualizing leakage flow field is provided for the embodiments of the present application, as shown in Figure 3 and Figure 2The experimental and visualization overall process is shown: a single-lens reflex camera (31) and a laser with a grating sheet capable of generating a random dot image with a duty cycle of 51.75% (32) are placed on one side of the metal tank. Turn on the laser, project the laser dot image on the wall (33), and adjust the laser intensity to the point where the camera can capture the image of the laser dot. Adjust the focal length, distance of the image acquisition device so that the field of view of the image acquisition device can cover the range of the laser dot pattern and the target flow field that may exist, and adjust the lens of the image acquisition device so that the laser dot image is clearly imaged. Before the soil covering device starts to leak, use the image acquisition device to capture a background pattern photo as a reference image. Control the soil covering device to start leaking, and at the same time, the camera is installed at a certain frequency to capture images as a disturbance image set. Input the captured images into the computer, and use the CLAHE algorithm and the Gaussian filter algorithm to denoise the pre-captured images. Use the cross-correlation algorithm based on multi-scale window and fast Fourier transform, and the sub-pixel peak search algorithm to obtain the current time disturbance image displacement field information. Input the reference image and the disturbance image; set the large window side length to 32px and the step length to 16px; perform cross-correlation operation based on fast Fourier transform on the data of the corresponding large window of the reference image and the disturbance image, and use the sub-pixel search algorithm to obtain the sub-pixel level displacement data of the current large window. Set the next level window side length to 16px and the step length to 8px; based on the displacement data obtained by the large window, perform offset step, perform cross-correlation operation based on fast Fourier transform on the data of the corresponding next level window of the reference image and the disturbance image, and the calculation process is shown in Figure 3 According to the small window step length, move the small window until the entire large window area is traversed. According to the large window step length, move the large window until the entire image area is traversed, and the laser dot disturbance displacement field data caused by the leakage flow field of the soil covering tank at the current time is obtained. The obtained current time displacement information is denoised. According to the displacement information, visualize it as a cloud chart to obtain the leakage flow field information at the current time. Through the denoising algorithm, remove the error, mechanical and background noise. Finally, the leakage flow field information at the current time is obtained.

[0034] The underground soil covering tank leakage simulation device of the present application mainly includes two parts: a simulation device for simulating the leakage of the underground soil covering tank and a detection device for detecting the leakage parameters of the soil covering tank.

[0035] As Figure 2As shown, the simulation device includes the following structure: a metal tank with an open upper end, a vent hole is provided at the bottom of the metal tank, and the hole is connected with the gas supply system through a pipeline. The structure of the gas supply system can be: mainly including a gas source, a buffer tank and a pipeline for connection, and a valve. The gas source is connected with the buffer tank through a pipeline with a valve, a pressure instrument is installed on the buffer tank, and the downstream of the buffer tank is connected with the hole at the bottom of the metal tank through a pipeline with a valve. The structure of the gas supply system can also be: the vent hole is directly connected with the pressure container, the diameter of the connection pipe of the pressure container and the vent hole is greater than the diameter of the vent hole, and the opening and closing are controlled by an electromagnetic valve.

[0036] Further, a plurality of porous metal plates can be arranged in the metal tank, and the edges of the metal plates and the metal tank are sealed by viscous substances or sealing rings. Because the weight of the covering soil during simulation is lighter than the actual situation, it is easier to be blown away by the airflow. In order to ensure that the leakage of gas under the condition that the covering soil is not blown away can be simulated, the covering soil is prevented from being blown away by the arrangement of multiple layers of porous metal plates. The sealing between the edges of the metal plates and the metal tank can avoid the loss of airflow from the edges of the metal tank as much as possible. In this way, the difference between simulation and actual situation is avoided.

[0037] Further, a metal mesh is covered on the hole at the bottom of the metal tank. To avoid sand or soil flowing into the pipeline.

[0038] Further, according to the covering soil tank structure in actual engineering, the thickness, number of layers and order of each layer of soil, sand and other materials in the metal tank are set. The soil can be different types of soil from different regions, and the porosity of the soil can be controlled by adjusting the mesh size of the soil particles. The height of the metal tank should be consistent with the total height of each material layer. The shape and area of the hole should be set according to the actual simulation situation, and the hole area is used to simulate the leakage position. The operating pressure of the buffer tank should be set according to the design pressure in the actual covering soil tank. The soil environment after precipitation can also be simulated by watering and sedimentation.

[0039] Further, by setting a spraying device above the metal tank, the change of the leakage flow field of the covering soil tank in the rain environment can be simulated. By heating and cooling the soil layer, the change of the leakage flow field of the covering soil tank in the high temperature environment in summer and the low temperature environment in winter can be simulated.

[0040] Further, based on the experimental device, the following parameters can also be measured: by using a high-speed camera for image acquisition and post-processing algorithm visualization, the flow rate of the leakage airflow can be judged by the morphological change of the flow field. A plurality of temperature sensors and pressure sensors are arranged in the soil to obtain the temperature and pressure distribution of the leakage airflow in the soil. Through the above device and by changing the relevant parameters, the simulation of the covering soil tank leakage in different storage pressure, storage medium, soil layer thickness, soil layer porosity and other parameters, and real environment can be realized.

[0041] Moreover, most of the current researches on the buried tank or pipeline focus on the numerical simulation of how the gas spreads in the soil after leakage, and there are relatively few researches on how the gas leaks into the atmosphere through the soil after leakage in real conditions, especially the researches on the flow field pattern and flow rate of the gas after passing through the soil. Based on the simulation device and the current technical deficiency, the following detection device and method are provided:

[0042] The detection device comprises the following devices: a laser emitting device provided with a grating sheet capable of dispersing laser into a scattered dot image; and an image acquisition equipment (an industrial camera) installed on the fixed device and the tripod. Compared with the traditional background schlieren method, the laser speckle is used as a carrier for depicting the flow field pattern, and the cost is lower, and a suitable background plate does not need to be made for a specific scene, and the laser device can be directly irradiated on any vertical surface or directly on the soil, and different angle and scene leakage and visualization experiments of the buried tank can be completed by using a set of laser devices; meanwhile, additional light supplement is not needed. In addition, when the leakage experiment of the buried tank with a large area is monitored, the leakage approximate position can be traced back by adjusting different angles.

[0043] That is, the underground buried tank leakage simulation and flow field visualization experiment device provided in the application comprises a metal tank with an open upper end, and the bottom of the metal tank is provided with a vent hole connected with a gas supply system through a pipeline. The gas supply system mainly comprises a gas source, a buffer tank, and a pipeline and a valve for connection. The gas source is connected with the buffer tank through a pipeline provided with a valve, and a pressure instrument is installed on the buffer tank. The downstream of the buffer tank is connected with the hole at the bottom of the metal tank through a pipeline provided with a valve. The simulation experiment and the flow field visualization process are as follows: the focal length of the image acquisition equipment is adjusted so that the field of view of the image acquisition equipment can cover the background plate and the flow field to be observed, the image acquisition equipment is adjusted so that the image acquisition equipment can be focused on the background plate, and the on-site background plate should be ensured to have sufficient illumination; before the valve is opened for ventilation, a background plate photo is collected by using the image acquisition equipment, or before the valve is opened for ventilation, the image acquisition equipment starts to collect the background plate photo at a certain frequency; the gas source is opened, the pressure of the buffer tank is adjusted to reach the experimental operating pressure P, the downstream pipeline valve is opened, the gas is sprayed from the hole at the bottom of the metal tank, and at the same time, the camera starts to collect the image of the background plate after the simulation leakage; all the valves are closed; the image data collected by the image acquisition equipment is transmitted to the computer for image data post-processing, and finally the visual underground buried tank leakage flow field image is obtained.

[0044] Based on the above experiment device and detection device, the process of the leakage simulation and measurement of the buried tank is as follows:

[0045] The embodiment of the application provides a flow field visualization method based on underground buried tank leakage, as shown in Figure 1As shown, the underground buried tank leakage flow field visualization method specifically includes steps S101-S105:

[0046] S101, by using the combination of laser equipment and industrial camera, the ground target flow field area of the underground buried tank is subjected to image acquisition processing related to laser scattering points, and a leakage disturbance image is obtained.

[0047] Specifically, it is necessary to first acquire a no-leakage reference image of the ground target flow field area before the leakage simulation by the industrial camera. Then the underground buried tank is controlled to perform leakage simulation.

[0048] Further, by the laser equipment, the ground target flow field area after the leakage simulation is subjected to laser scattering point irradiation processing, and a laser irradiation scattering point area is determined.

[0049] Further, by the adjusted focal length parameter and distance parameter of the industrial camera, the laser irradiation scattering point area and the ground target flow field area are subjected to global image acquisition processing, and a leakage disturbance image is obtained.

[0050] In one embodiment, before the buried device starts to leak, an image acquisition device is used to capture a background pattern photo as a reference image. Then the laser is turned on, and the laser scattering point image is projected on the facade or soil surface. The laser intensity is adjusted to the point that the camera can capture the laser scattering point image. Then the focal length and distance of the image acquisition device are adjusted so that the field of view of the image acquisition device can cover the laser scattering point pattern and the possible range of the target flow field. The lens of the image acquisition device is adjusted so that the laser scattering point image is clearly imaged. Then the buried device is controlled to start leaking, and the camera is installed to capture images at a certain frequency as a disturbance image set. The higher the capture frequency, the better the flow field pattern evolution and leakage speed of the leaked gas through the soil can be exhibited. Finally, the leakage disturbance image is captured and obtained.

[0051] S102, the leakage disturbance image is subjected to noise removal processing related to pepper noise, and a pretreated leakage disturbance image is obtained.

[0052] Specifically, the laser scattering point light spot in the leakage disturbance image also needs to be subjected to brightness enhancement processing by a preset CLAHE algorithm, and based on the contrast of other background information in the leakage disturbance image, a first leakage disturbance image is obtained. Then the pepper noise in the first leakage disturbance image is subjected to filtering processing by a Gaussian filtering algorithm, and a pretreated leakage disturbance image is obtained.

[0053] In one embodiment, the collected images are input into a computing system, and a CLAHE algorithm and a Gaussian filtering algorithm are used to denoise the pre-collected images. Since the soil tank leakage experiment is usually carried out in an open environment, the laser speckle pattern is not clear enough under the influence of natural light, and therefore the CLAHE algorithm is used to enhance the contrast of the image and improve the distinction between the laser speckle and other background information, thereby providing more details for the subsequent cross-correlation calculation. However, since this operation will enhance the pepper noise generated by the camera, especially when the natural light is insufficient, the Gaussian filtering algorithm is added to remove the pepper noise.

[0054] S103, by the cross-correlation algorithm of the multi-scale window and the fast Fourier transform, and based on the pre-collected no-leakage reference image, the pre-processed leakage disturbance image is calculated based on the laser speckle disturbance displacement caused by the ground target flow field region, to obtain the disturbance image displacement field information at the current time.

[0055] Specifically, the no-leakage reference image and the pre-processed leakage disturbance image are first combined to obtain a disturbance image group.

[0056] Further, the current large window data corresponding to the disturbance image group is calculated by the fast Fourier transform, and based on the sub-pixel search algorithm, the sub-pixel level displacement field data of the current large window is obtained.

[0057] As a feasible implementation, the no-leakage reference image and the pre-processed leakage disturbance image in the disturbance image group are respectively calculated to obtain the first large window gray value data of the no-leakage reference image and the second large window gray value data of the pre-processed leakage disturbance image. The first large window gray value data and the second large window gray value data are respectively calculated by the fast Fourier transform, and based on the point-by-point complex calculation of the complex conjugate and the inverse Fourier transform, the displacement field data of the current large window is obtained. The sub-pixel peak value search is performed on the displacement field data of the current large window by the sub-pixel search algorithm, and based on the pixel level peak position and the sub-pixel level peak position, the sub-pixel level displacement field data of the current large window is searched and generated.

[0058] Further, the next level large window cross-correlation calculation is also performed on the disturbance image group according to the offset step of the sub-pixel level displacement data and based on the fast Fourier transform, to obtain the large window cross-correlation operation data. The next level configuration parameter of the next level large window is half of the configuration parameter of the current large window data.

[0059] Further, the large window region in the large window cross-correlation operation data is processed by a preset small window step to obtain small window traversal result data.

[0060] Further, based on the small window traversal result data, and through the large window step, the pre-processed leakage disturbance image is calculated for the full image region traversal under the laser scattering point disturbance displacement, and finally the disturbance image displacement field information at the current time is obtained.

[0061] In one embodiment, the cross-correlation algorithm based on multi-scale window and fast Fourier transform, and the sub-pixel peak search algorithm are used to obtain the disturbance image displacement field information at the current time. The soil covering leakage visualization experiment may need to be reconstructed in the later stage Air flow density field distribution, which requires sufficient visualization image accuracy and resolution. Multi-scale window cross-correlation calculation can improve the accuracy and spatial resolution of cross-correlation analysis. Fast Fourier cross-correlation calculation can improve the calculation speed.

[0062] As a feasible implementation, the cross-correlation algorithm based on multi-scale window and fast Fourier transform includes the following steps:

[0063] (1) First, input the reference image and the disturbance image.

[0064] (2) Set the large window side length as M, and the step as M / 2. Perform cross-correlation operation based on fast Fourier transform on the data corresponding to the large window of the non-leakage reference image and the pre-processed leakage disturbance image, and use the sub-pixel search algorithm to obtain the sub-pixel level displacement data of the current large window.

[0065] (3) Set the next window side length as M / 2, and the step as M / 4. Based on the displacement data obtained by the large window, perform offset step, and perform cross-correlation operation based on fast Fourier transform on the data corresponding to the next window of the reference image and the disturbance image.

[0066] (4) Move the small window according to the small window step until the entire large window region is traversed.

[0067] (5) Move the large window according to the large window step until the entire image region is traversed, and obtain the laser scattering point disturbance displacement field data caused by the soil covering tank leakage flow field at the current time, that is, the disturbance image displacement field information at the current time.

[0068] Wherein, the cross-correlation algorithm based on fast Fourier transform has the following calculation process:

[0069] f(x,y) and g(x,y) represent the image gray value data corresponding to the window of the non-leakage reference image and the pre-processed leakage disturbance image respectively, and N is the side length of the window. Through fast Fourier transformation, F(m,n) and G(m,n) are obtained, and the calculation formula is as follows:

[0070]

[0071]

[0072] Then the complex conjugate of F(m, n) is multiplied with G(m, n) Point-by-point complex multiplication is performed to obtain R(m, n):

[0073]

[0074] Then the inverse Fourier transform of R(m, n) is performed to obtain the displacement field data r(x, y) of the current window:

[0075]

[0076] The sub-pixel peak search algorithm uses a three-point two-dimensional Gaussian fitting method to search for a sub-pixel peak, (Δu, Δv) is the pixel-level peak position, and (Δx, Δy) is the sub-pixel-level peak position. The calculation formula is as follows:

[0077]

[0078]

[0079] S104, triple noise correction calculation is performed on the perturbed image displacement field information to obtain the true displacement field information. The triple noise correction includes error data correction, mechanical noise correction, and background noise correction.

[0080] Specifically, first, the perturbed image displacement field information is converted into a matrix to obtain displacement field matrix data.

[0081] Further, the error data correction calculation under the influence of environmental factors is performed on the displacement field matrix data by using a first preset threshold to obtain the first corrected displacement field matrix data.

[0082] In an embodiment, when detecting the topsoil leakage, since the soil or sand particles will move with the change of the environment, thus causing error data in the displacement field data, error data correction should be performed. Since the displacement data caused by the movement of these particles is larger than the displacement data caused by the disturbance of the laser scattering points by the flow field, the error data correction can be removed by the following steps. The error data correction step is: for the m-row n-column displacement field matrix data A=[a ij ], a threshold α is set, and the error-corrected displacement field matrix data (the first corrected displacement field matrix data) A1=[a′ ij ] can be represented as: Generally, the threshold is set according to the actual situation according to the maximum displacement number caused by the flow field.

[0083] Further, the data of the non-flow field region in the first corrected displacement field matrix data is selected; the data of the non-flow field region is calculated by regional fitting to obtain fitting plane data.

[0084] Further, the first corrected displacement field matrix data is fitted by the fitting plane data to obtain a mechanical noise correction matrix. The first corrected displacement field matrix data is subtracted by the mechanical noise correction matrix to obtain second corrected displacement field matrix data.

[0085] In an embodiment, since the implementation of the soil covering tank leakage generally needs to be carried out in a well-ventilated environment, especially for toxic and harmful, flammable and explosive gases, the entire experimental device should be carried out in an open environment. Therefore, the entire experimental operation will inevitably be affected by the natural environment, such as natural wind. The existence of natural wind will cause a certain jitter of the camera or laser used for shooting, thereby causing mechanical noise in the final displacement data field, which is generally regularly distributed in the data. The mechanical noise correction step is: selecting the data of the region without flow field in the displacement data A1, performing least square calculation on the region to obtain the fitting plane calculation formula f(a″ ij )=ai+bj+c, wherein a, b, and c are obtained by the least square method. The fitting plane of the entire displacement data is calculated by the calculation formula to obtain a mechanical noise correction matrix B1. The corrected displacement data (second corrected displacement field matrix data) A2 is obtained: A2=A1-B1

[0086] Further, the pre-acquired displacement field data without flow field is calculated by summation and average to obtain a background noise template matrix. The second corrected displacement field matrix data is subtracted by the background noise template matrix to obtain third corrected displacement field matrix data.

[0087] In an embodiment, at the same time, there is a kind of noise in the displacement field data which has not been mentioned in the previous background schlieren related research. When analyzing the displacement field data, the background noise is generated due to the joint action of the background pattern and the displacement evaluation algorithm. The background noise has the characteristics of spatial random distribution and temporal stable distribution. Since the background noise distribution is consistent when the background pattern is consistent, the stable background noise template for the current laser speckle pattern can be obtained after the summation and average of the displacement data without flow field at multiple times. The background noise correction step is: N displacement field data A′ i are calculated by summation and average operation to obtain a background noise template matrix B2: Finally, the corrected displacement data (third corrected displacement field matrix data) A3 is obtained: A3=A2-B2.

[0088] Further, based on the displacement field matrix data after the third correction, real displacement field information is obtained. That is, through the above noise removal step, most of the noise in the displacement information that can depict the flow field is removed, thereby obtaining the real displacement field information.

[0089] S105, based on the real displacement field information, the leakage flow field information at the current time is determined.

[0090] Specifically, the flow field parameters in the real displacement field information are mapped to a multi-layer color space. The multi-layer color space includes a hue layer representing the flow direction, a saturation layer representing the pressure gradient amplitude, and a lightness layer representing the vorticity intensity.

[0091] Further, through the time slicing technology, the multi-layer color space is projected and rendered in two dimensions under the three-dimensional flow field, and the leakage flow field information at the current time is obtained.

[0092] Meanwhile, the present application can: 1) accurately simulate the leakage of underground buried tank. The underground buried tank leakage simulation device can accurately simulate the leakage process of the underground buried tank, especially the leakage flow field under complex soil structure and pressure conditions, helping researchers to fully understand and predict the impact of leakage on the surrounding environment. 2) Adjustable experimental parameters. The device can adjust the thickness of the soil layer in the metal tank, the shape and area of the orifice, the operating pressure of the buffer tank and other parameters according to actual needs, so as to realize accurate simulation of different underground buried tank leakage scenarios, improve the applicability and diversity of the experiment. 3) Real-time data acquisition and processing. The scheme can process the images obtained during the experiment in real time to obtain the displacement field data of the flow field, provide dynamic flow field change images for the experiment, and help in-depth analysis of the behavior of the fluid.

[0093] Moreover, the lack of high-precision underground buried tank leakage flow field simulation technology. Current domestic and foreign research on underground buried tank leakage mostly focuses on leakage volume, leakage rate and its impact on the environment, and lacks high-precision and real-time simulation of underground buried tank leakage flow field. The present application realizes the dynamic visualization of the flow field in the underground buried tank leakage process by introducing the class streak technology and combining high-resolution image acquisition equipment, filling the gap in this research field.

[0094] As a feasible embodiment, Figure 4 A normal soil covering and compaction soil covering leakage flow field comparison chart provided by an embodiment of the present application, Figure 5 A gas leakage flow field shape evolution chart provided by an embodiment of the present application, Figure 6 A leakage flow field comparison chart under different sand thicknesses provided by an embodiment of the present application, such as Figure 4 、 5, 6, the above simulation results are obtained through the flow field visualization method based on underground buried tank leakage, and the problems of lack of underground buried tank experimental device and high cost and complex setting of the previous buried tank leakage flow field visualization method are solved.

[0095] In addition, the embodiment of the present application also provides a flow field visualization device based on underground buried tank leakage, as shown in Figure 7 The flow field visualization device 700 based on underground buried tank leakage specifically comprises:

[0096] at least one processor 701, and a memory 702 connected with the at least one processor 701. The memory 702 stores instructions executable by the at least one processor 701, so that the at least one processor 701 can execute the following steps:

[0097] Through the combination of the laser device and the industrial camera, the image acquisition and processing of the laser scattering points on the target flow field area of the underground buried tank are performed, and a leakage disturbance image is obtained;

[0098] The leakage disturbance image is subjected to the noise removal processing related to the pepper noise, and a pretreated leakage disturbance image is obtained;

[0099] Through the cross-correlation algorithm of the multi-scale window and the fast Fourier transform, and based on the pre-acquired non-leakage reference image, the pretreated leakage disturbance image is subjected to the calculation of the laser scattering point disturbance displacement caused by the target flow field area on the ground, and a disturbance image displacement field information at the current time is obtained;

[0100] The disturbance image displacement field information is subjected to the triple noise correction calculation, and a real displacement field information is obtained. The triple noise correction includes error data correction, mechanical noise correction and background noise correction;

[0101] Based on the real displacement field information, the leakage flow field information at the current time is determined.

[0102] The application can restore the actual underground buried tank leakage situation through the underground buried tank leakage flow field visualization detection, simulate the leakage situation of the underground buried tank structure with different burial depths and leakage hole diameters in the actual engineering by setting the metal tank, different levels of soil materials and adjustable pressure and orifice area, and can more truly reproduce the flow field change when the underground buried tank leaks. The flow field disturbance can be captured and accurately analyzed through the schlieren method and image processing technology, and high-precision flow field displacement data can be obtained, so that the flow field generated by the leakage can be visualized. Moreover, accurate displacement field data can be obtained through image data post-processing algorithm, so that the evolution process of the flow field can be tracked and analyzed in detail. At the same time, the images obtained during the experiment can be processed in real time to obtain the displacement field data of the flow field, which provides dynamic flow field change images for the experiment and helps to analyze the behavior of the fluid.

[0103] Each of the embodiments in the application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments. Especially, the device embodiments are described simply because they are basically similar to the method embodiments, and the relevant parts can be referred to the part of the method embodiments.

[0104] The above describes specific embodiments of the application. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims can be performed in a different order than those described in the embodiments and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order in order to achieve the desired results. In some implementations, multitasking and parallel processing can be advantageous or necessary.

[0105] The above only describes the embodiments of the application and does not limit the application. The embodiments of the application can be variously changed and modified by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the application shall be included in the scope of the claims of the application.

Claims

1. A flow field visualization method based on leakage from an underground soil-covered tank, characterized in that, The method includes: By combining laser equipment with industrial cameras, images of laser scatter points are acquired and processed in the above-ground target flow field area of ​​underground soil-covered tanks to obtain leakage disturbance images. The leakage disturbance image is denoised to remove pepper-like noise, resulting in a preprocessed leakage disturbance image. By using a cross-correlation algorithm of multi-scale window and fast Fourier transform, and based on the pre-acquired leak-free reference image, the displacement of the laser scatter point disturbance caused by the flow field region of the ground target is calculated on the pre-processed leak disturbance image to obtain the displacement field information of the disturbance image at the current moment. The displacement field information of the disturbed image is subjected to triple noise correction calculation to obtain the true displacement field information; wherein, the triple noise correction includes: error data correction, mechanical noise correction and background noise correction; Based on the actual displacement field information, the leakage flow field information at the current moment is determined.

2. The flow field visualization method based on leakage from an underground soil-covered tank according to claim 1, characterized in that, By combining laser equipment with an industrial camera, images of laser scatter points are acquired and processed in the above-ground target flow field area of ​​the underground soil-covered tank to obtain leakage disturbance images, specifically including: The underground soil-covered tank was controlled to simulate a leak. The laser device is used to irradiate the ground target flow field area after the leakage simulation with laser scattering to determine the laser irradiation scattering area. By using the adjusted focal length and distance parameters of the industrial camera, the laser-irradiated scattered area and the ground target flow field area are subjected to full-domain image acquisition and processing to obtain the leakage disturbance image.

3. The flow field visualization method based on leakage from an underground covered tank according to claim 1, characterized in that, The leakage disturbance image is denoised to remove pepper-like noise, resulting in a preprocessed leakage disturbance image, specifically including: The laser scattering spot in the leakage disturbance image is brightness equalized using the preset CLAHE algorithm, and a first leakage disturbance image is obtained based on the contrast of other background information in the leakage disturbance image. The pepper-grain noise in the first leakage disturbance image is filtered using a Gaussian filtering algorithm to obtain the preprocessed leakage disturbance image.

4. The flow field visualization method based on leakage from an underground covered tank according to claim 1, characterized in that, Using a cross-correlation algorithm combining multi-scale windows and fast Fourier transform, and based on a pre-acquired leak-free reference image, the displacement of the preprocessed leak disturbance image is calculated based on the laser scattering disturbance induced by the ground target flow field region. This yields the displacement field information of the disturbance image at the current moment, specifically including: The leak-free reference image and the preprocessed leak perturbation image are combined to obtain a perturbation image group; The fast Fourier transform is used to perform cross-correlation calculation on the current large window data corresponding to the perturbed image group, and the sub-pixel displacement field data of the current large window is obtained based on the sub-pixel search algorithm. Based on the offset step size of the subpixel displacement data and the fast Fourier transform, the cross-correlation calculation of the next-level large window is performed on the perturbed image group to obtain the large window cross-correlation operation data; wherein, the next-level configuration parameter of the next-level large window is half of the current large window data configuration parameter; By using a preset small window step size, the large window region in the large window cross-correlation calculation data is traversed to obtain the small window traversal result data; Based on the small window traversal result data, and using a large window step size, the preprocessed leaked disturbance image is subjected to full image region traversal calculation under the laser scatter point disturbance displacement to obtain the displacement field information of the disturbance image at the current moment.

5. The flow field visualization method based on leakage from an underground covered tank according to claim 4, characterized in that, The fast Fourier transform is used to perform cross-correlation calculations on the large window data corresponding to the perturbed image group, and based on the sub-pixel search algorithm, the sub-pixel level displacement field data of the current large window is obtained, specifically including: The grayscale values ​​of the leak-free reference image and the preprocessed leak perturbation image in the perturbation image group are calculated respectively to obtain the grayscale value data of the first large window of the leak-free reference image and the grayscale value data of the second large window of the preprocessed leak perturbation image. Fast Fourier transform calculations are performed on the grayscale data of the first large window and the grayscale data of the second large window respectively, and the displacement field data of the current large window is obtained based on the pointwise complex number calculation and inverse Fourier transform of the complex conjugate quantity. The subpixel search algorithm is used to perform subpixel peak search on the displacement field data of the current large window, and the subpixel displacement field data of the current large window is searched and generated based on the pixel-level peak position and the subpixel-level peak position.

6. The flow field visualization method based on leakage from an underground covered tank according to claim 1, characterized in that, The displacement field information of the disturbed image is subjected to triple noise correction calculation to obtain the true displacement field information, specifically including: The displacement field information of the disturbed image is matrix transformed to obtain displacement field matrix data; By setting a threshold, the erroneous data of the displacement field matrix data is corrected and calculated to obtain the displacement field matrix data after the first correction. The displacement field matrix data after the first correction is fitted to a plane using the least squares method to obtain the mechanical noise correction matrix; and the mechanical noise correction matrix is ​​subtracted from the displacement field matrix data after the first correction to obtain the displacement field matrix data after the second correction. The background noise template matrix is ​​obtained by summing and averaging the pre-acquired displacement field data that do not have a flow field. Subtract the background noise template matrix from the displacement field matrix data after the second correction to obtain the displacement field matrix data after the third correction. Based on the displacement field matrix data after the third correction, the true displacement field information is obtained.

7. The flow field visualization method based on leakage from an underground covered tank according to claim 6, characterized in that, Before performing plane fitting on the first corrected displacement field matrix data by fitting plane data to obtain the mechanical noise correction matrix, the method further includes: Select the data from the displacement field matrix data that does not show the flow field region; For data in regions where no flow field is observed, regional fitting calculations are performed to obtain fitted plane data.

8. The flow field visualization method based on leakage from an underground covered tank according to claim 1, characterized in that, Based on the actual displacement field information, the leakage flow field information at the current moment is determined, specifically including: The flow field parameters in the real displacement field information are mapped to a multi-layer color space; wherein, the multi-layer color space includes: a hue layer representing the flow velocity direction, a saturation layer representing the pressure gradient magnitude, and a brightness layer representing the vorticity intensity. By using time slicing technology, two-dimensional projection rendering under a three-dimensional flow field is performed on the multi-layer color space to obtain the leakage flow field information at the current moment.

9. The flow field visualization method based on leakage from an underground covered tank according to claim 1, characterized in that, The leak-free reference image of the ground target flow field region was acquired using an industrial camera before leakage simulation was performed.

10. A flow field visualization device based on leakage from an underground covered tank, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform a flow field visualization method based on leakage from an underground soil-covered tank, as described in any one of claims 1-9.

Images