Buried oil pipeline leakage experiment device and oil spilling detection method
By designing a buried oil pipeline leakage experimental device and combining a high-density resistivity method and a physical information neural network model, the problem of inability to simulate pipeline damage and low detection accuracy in the existing technology is solved, and high-precision and real-time oil spill detection effect is achieved.
Patent Information
- Application Number
- CN202510937225.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-08
- Publication Date
- 2025-08-15
AI Technical Summary
The existing experimental devices cannot simulate the sudden breakage of buried pipelines under stable operation conditions, and the inversion method of the traditional high-density resistivity method has low accuracy, which cannot meet the real-time and accuracy requirements of oil spill detection.
A buried oil pipeline leakage experimental device is designed, including liquid sections, gas sections, mix sections, leakage sections, circulation sections and detection systems. Combined with high-density resistivity method and physical information neural network model, the position and number of leakage holes are automatically adjusted, the flow rate is increased, and data inversion is carried out to obtain oil spill distribution images.
The accuracy and speed of simulation and oil spill detection of buried oil pipeline leakage has been improved, meeting the needs of high accuracy and real-time.
Smart Images
Figure CN120488154A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of buried oil pipeline leakage detection, and in particular to a buried oil pipeline leakage test device and an oil spill detection method. Background Art
[0002] Pipeline transportation has become the most common method of oil transportation due to its low cost, minimal losses, high transport volumes, and high safety. However, when pipelines are damaged by external forces such as natural disasters or human activity, or develop holes due to aging and corrosion, oil can spray out of these holes, posing a serious threat to the natural environment and human life and property.
[0003] Currently, many researchers conduct laboratory experiments to study the migration characteristics of spilled oil in soil. However, traditional experimental setups are unable to simulate the sudden failure of buried pipelines under stable operating conditions, nor can they account for the impact of flow patterns on leaks. Furthermore, the pipeline pressures used in laboratory experiments are low, failing to achieve the flow velocities required to generate a jet. These factors limit further research on the migration of spilled oil in soil. Furthermore, effective methods for detecting spilled oil in soil remain lacking for three-dimensional experiments. Consequently, most researchers still use two-dimensional soil chambers, which fail to accurately reflect the migration process of spilled oil in real environments. High-density resistivity is a geophysical research method widely used in fields such as topographic exploration and mining. In recent years, researchers have applied high-density resistivity to detect underground oil spills, but traditional inversion methods suffer from low accuracy. Compared to two-dimensional methods, three-dimensional high-density resistivity methods require exponentially more data, resulting in a massive inversion computational effort and low efficiency, making them unable to meet the real-time requirements of oil spill detection. Summary of the Invention
[0004] In response to the shortcomings of existing experimental devices and detection methods, the present invention provides a buried oil pipeline leakage experimental device and oil spill detection method, which overcomes the shortcomings of existing experimental devices, can simulate the sudden occurrence of leakage during the safe operation of the pipeline, consider the different locations and numbers of leakage holes, further increase the leakage flow rate through the throttling structure of the valve, and combine with the high-density resistivity method to quickly detect oil spills in the soil.
[0005] To achieve the above object, the present invention is specifically:
[0006] A buried oil pipeline leakage experimental device and an oil spill detection method are characterized in that the experimental device consists of a liquid section, a gas section, a mixing section, a leakage section, a circulation section, and a detection system.
[0007] Specifically, the liquid section includes a liquid storage tank, a pump, a first regulating valve, a first flow meter, a first pressure gauge, and a liquid pipeline; the first regulating valve, the first flow meter, and the first pressure gauge are installed on the liquid pipeline in sequence, the first regulating valve is used to regulate the flow of the liquid pipeline, and the first flow meter and the first pressure gauge are used to measure the flow and pressure of the liquid pipeline.
[0008] Specifically, the gas section includes a compressor, a gas storage tank, a second regulating valve, a second flow meter, a second pressure gauge, and a gas pipeline; the second regulating valve, the second flow meter, and the second pressure gauge are installed on the gas pipeline in sequence; the second regulating valve is used to adjust the flow of the gas pipeline, and the second flow meter and the second pressure gauge are used to measure the flow and pressure of the gas pipeline.
[0009] Specifically, the mixing section includes a mixing pipe, in which the gas and liquid are mixed and fully developed; by adjusting the flow rate of liquid and gas, pipeline leakage experiments with different flow types and flow rates can be achieved, and by changing the medium of liquid and gas, pipeline leakage experiments with different media can be achieved.
[0010] Specifically, the leakage section includes a leakage pipe, a leakage hole, an electric valve, a single-chip microcomputer, and a differential pressure sensor; the leakage hole is located on the leakage pipe, and unused leakage holes can be sealed with a wire plug; the flow channel of the electric valve adopts a throttling structure to increase the leakage flow rate; the electric valve is installed on the leakage hole through a thread, and the single-chip microcomputer is connected to the electric valve; several leakage holes can be drilled on the leakage pipe according to experimental requirements, and the electric valve is installed on each leakage hole. The inlet wall of the electric valve is flush with the inner wall of the pipe. The opening and closing of each electric valve is controlled by the single-chip microcomputer, and the number and position of the leakage holes can be automatically adjusted; the differential pressure sensor is installed on the electric valve, and the leakage flow can be calculated by the formula.
[0011] Specifically, the circulation section includes a circulation pipe, a cyclone separator, a gas discharge pipe, and a liquid discharge pipe; the cyclone separator is connected to the circulation pipe, the gas discharge pipe and the liquid discharge pipe are connected to the cyclone separator, and the liquid discharge pipe is connected to the liquid storage tank. After the fluid enters the cyclone separator, it is separated, the gas phase enters the gas discharge pipe and is discharged, and the liquid phase enters the liquid discharge pipe and is discharged into the liquid storage tank.
[0012] Specifically, the detection system includes a soil box, an electrode array, a high-density resistivity measuring device, and a data processing device; the electrode array is composed of a number of electrodes and is inserted into the soil on the surface of the soil box; the high-density resistivity measuring device is connected to the electrode array and is used to measure the soil resistivity; the data processing device is connected to the high-density resistivity measuring device and is used to process the measured resistivity data. i And inversion imaging.
[0013] A buried oil pipeline leakage test device and oil spill detection method, the oil spill detection method steps are as follows:
[0014] S1. At the beginning of the experiment, the pump and compressor are turned on, and the first and second regulating valves are adjusted to make the mixing pipeline reach the predetermined leakage condition. After the condition stabilizes, the electric valve is controlled by the single chip microcomputer to open the leakage hole, and the leakage flow rate is calculated by the differential pressure sensor;
[0015] S2. Set the electrode spacing, arrange the electrode array, record the electrode coordinates, and collect resistivity data ρ through a high-density resistivity measurement device. i ;
[0016] S3, data preprocessing, removing bad pixels, and organizing data format;
[0017] S4. Define the computational domain and boundaries, and determine the scope and boundary conditions of the underground space;
[0018] S5. Perform forward calculation to obtain a forward geoelectric model;
[0019] S6. Establish a physical information neural network model, input the measured resistivity and its spatial coordinates into the deep neural network for training, output the flow field parameters, use automatic differentiation technology to calculate the spatial gradient of these variables, construct a loss function L, and minimize the loss function L. Perform error backpropagation based on the chain rule, train the neural network parameters (w, b) until convergence, and finally output the inversion results;
[0020] S7. Compare the inversion result with the forward modeling result. If the error is less than or equal to 5%, output the resistivity distribution image and finally obtain the oil spill distribution image. Otherwise, continue to train the physical information neural network until the error meets the requirements.
[0021] The deep neural network consists of an input layer, a hidden layer, and an output layer. The output flow field parameters include flow velocity u, water saturation S w , oil saturation S o The loss function L is given by the control equation loss function L PDE And the measurement point matching loss function L Data The structure is used to ensure that the field variables predicted by the neural network satisfy the physical equations and are consistent with the currently known electric field information.
[0022] The present invention provides a buried oil pipeline leakage test device and an oil spill detection method, which have the following beneficial effects: the present invention overcomes the defects of indoor experiments on buried oil pipeline leakage, and can independently adjust the leakage flow rate, flow type and fluid medium; by changing the opening and closing of the electric valve, the situation of sudden pipeline leakage under stable working conditions can be simulated, and the position and number of leakage holes can be automatically controlled by a single-chip microcomputer; the throttling structure of the electric valve can increase the outlet flow rate, which is more consistent with the actual working conditions on site, and the leakage flow rate can be calculated; the soil is detected by a high-density resistivity method, and the measurement data is inverted in combination with a physical information neural network model to obtain an oil spill distribution image, which greatly improves the imaging accuracy and accelerates the inversion speed, meeting the high-precision and real-time requirements in oil spill detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for describing the present invention.
[0024] Figure 1 This is the flow chart of the buried pipeline leakage experimental device.
[0025] Figure 2 Schematic diagram of the leakage section structure.
[0026] Figure 3 Schematic diagram of the flow path structure of the electric valve.
[0027] Figure 4 Flow chart of the detection method.
[0028] Figure 5 This is a diagram of the physical information neural network model.
[0029] Reference numerals include:
[0030] 1—Liquid storage tank; 2—Pump; 3—First regulating valve; 4—First flow meter; 5—Pressure gauge 1; 6—Compressor; 7—Air storage tank;
[0031] 8—Second regulating valve; 9—Second flow meter; 10—Second pressure gauge; 11—Liquid pipeline; 12—Gas pipeline;
[0032] 13—mixing pipe; 14—leakage pipe; 141—leakage hole; 142—electric valve; 143—single-chip microcomputer; 144—differential pressure sensor;
[0033] 15—circulation pipe; 16—cyclone separator; 17—gas discharge pipe; 18—liquid discharge pipe; 19—soil box; 20—electrode array;
[0034] 21—High-density resistivity measurement device; 22—Data processing device; 23—Input layer; 24—Hidden layer; 25—Output layer. DETAILED DESCRIPTION
[0035] In order to make the technical solutions, objectives and advantages of the present invention more clear, the technical solutions are further described in detail below in conjunction with specific implementation methods.
[0036] like Figure 1 As shown, a buried oil pipeline leakage experimental device and an oil spill detection method are provided. The experimental device consists of a liquid section, a gas section, a mixing section, a leakage section, a circulation section, and a detection system.
[0037] The liquid section includes a liquid storage tank 1, a pump 2, a first regulating valve 3, a first flow meter 4, a first pressure gauge 5, and a liquid pipeline 11; the first regulating valve 3, the first flow meter 4, and the first pressure gauge 5 are installed on the liquid pipeline 11 in sequence, the first regulating valve 3 is used to adjust the flow of the liquid pipeline, and the first flow meter 4 and the first pressure gauge 5 are used to measure the flow and pressure of the liquid pipeline 11.
[0038] The gas section includes a compressor 6, a gas storage tank 7, a second regulating valve 8, a second flow meter 9, a second pressure gauge 10, and a gas pipeline 12; the second regulating valve 8, the second flow meter 9, and the second pressure gauge 10 are installed on the gas pipeline in sequence; the second regulating valve 8 is used to adjust the flow of the gas pipeline, and the second flow meter 9 and the second pressure gauge 10 are used to measure the flow and pressure of the gas pipeline 12.
[0039] The mixing section includes a mixing pipe 13, in which the gas and liquid are mixed and fully developed; by adjusting the flow rate of the liquid and gas, pipeline leakage experiments with different flow types and flow rates can be realized, and by changing the medium of the liquid and gas, pipeline leakage experiments with different media can be realized.
[0040] The leakage section includes a leakage pipe 14, a leakage hole 141, an electric valve 142, a single chip computer 143, and a differential pressure sensor 144. Figure 2 As shown, the leakage hole 141 is located on the leakage pipe 14, and the unused leakage holes 141 can be sealed with a wire plug; the flow channel of the electric valve 142 adopts a throttling structure, which can increase the fluid flow rate; the electric valve 142 is installed on the leakage hole 141 through a thread, and the inlet wall of the electric valve 142 is flush with the inner wall of the leakage pipe 14, and the single-chip microcomputer 143 is connected to the electric valve 142; a number of leakage holes 141 can be punched on the leakage pipe 14 according to experimental requirements, and the electric valve 142 is installed on each leakage hole 141. The opening and closing of each electric valve 142 is controlled by the single-chip microcomputer 143, and the number and position of the leakage holes 141 can be adjusted independently; the differential pressure sensor 144 is installed on the electric valve 142, and the flow channel structure of the electric valve 142 is as shown Figure 3 As shown, the leakage flow can be calculated by the following formula:
[0041]
[0042] Where Q is the outlet flow rate, S1 is the inlet section area, S2 is the throat area, ρ is the fluid density, ΔP is the differential pressure, and C is the discharge coefficient.
[0043] The circulation section includes a circulation pipe 15, a cyclone separator 16, a gas discharge pipe 17, and a liquid discharge pipe 18; the cyclone separator 16 is connected to the circulation pipe 15, the gas discharge pipe 17 and the liquid discharge pipe 18 are connected to the cyclone separator 16, and the liquid discharge pipe 18 is connected to the liquid storage tank 1. After the fluid enters the cyclone separator 16, it is separated, the gas phase enters the gas discharge pipe 17 and is discharged, and the liquid phase enters the liquid discharge pipe 18 and is discharged into the liquid storage tank 1.
[0044] The detection system includes a soil box 19, an electrode array 20, a high-density resistivity measuring device 21, and a data processing device 22; the electrode array 20 is composed of a plurality of electrodes and is inserted into the soil on the surface of the soil box 19, the high-density resistivity measuring device 21 is connected to the electrode array 20, and is used to measure the soil resistivity, and the data processing device 22 is connected to the high-density resistivity measuring device 21, and is used to process the measured resistivity data ρ i And inversion imaging.
[0045] According to the experimental requirements, set the fluid medium, fluid working conditions, fluid flow pattern, leakage hole position, and leakage hole number, and build the experimental device. Figure 4 The specific steps are as follows:
[0046] S1. The experiment begins by turning on the pump 2 and the compressor 6, and adjusting the first regulating valve 3 and the second regulating valve 8 to make the mixing pipe 13 reach a predetermined leakage condition. After the condition stabilizes, the single chip microcomputer 143 controls the electric valve 142 to open the leakage hole 141, and the differential pressure sensor 144 calculates the leakage flow rate.
[0047] S2. Set the electrode spacing, arrange the electrode array 20, record the electrode coordinates, and collect resistivity data ρ through the high-density resistivity measurement device 21. i ;
[0048] S3, data preprocessing, removing bad pixels, and organizing data format;
[0049] S4. Define the computational domain and boundaries, and determine the scope and boundary conditions of the underground space;
[0050] S5. Perform forward calculation to obtain a forward geoelectric model;
[0051] S6. Establish a physical information neural network model, input the measured resistivity and its spatial coordinates into the deep neural network for training, output the flow field parameters, use automatic differentiation technology to calculate the spatial gradient of these variables, construct a loss function L, and minimize the loss function L. Perform error backpropagation based on the chain rule, train the neural network parameters (w, b) until convergence, and finally output the inversion results;
[0052] S7. Compare the inversion result with the forward modeling result. If the error is less than or equal to 5%, output the resistivity distribution image and finally obtain the oil spill distribution image. Otherwise, continue to train the physical information neural network until the error meets the requirements.
[0053] like Figure 5 As shown, the deep neural network includes an input layer 23, a hidden layer 24 and an output layer 25. The output flow field parameters include flow velocity u, water saturation S w , oil saturation S o , the loss function L is determined by the control equation loss function L PDE And the measurement point matching loss function L Data The structure is used to ensure that the field variables predicted by the neural network satisfy the physical equations and are consistent with the currently known electric field information.
[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned implementation methods, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the present invention.
Claims
1. A buried oil pipeline leakage test device and oil spill detection method, characterized in that: The experimental device consists of a liquid section, a gas section, a mixing section, a leakage section, a circulation section, and a detection system; The liquid section comprises a liquid storage tank (1), a pump (2), a first regulating valve (3), a first flow meter (4), a first pressure gauge (5), and a liquid pipeline (11); the first regulating valve (3), the first flow meter (4), and the first pressure gauge (5) are sequentially installed on the liquid pipeline (11); The gas section comprises a compressor (6), a gas storage tank (7), a second regulating valve (8), a second flow meter (9), a second pressure gauge (10), and a gas pipeline (12); the second regulating valve (8), the second flow meter (9), and the second pressure gauge (10) are sequentially installed on the gas pipeline (12); The mixing section includes a mixing pipe (13); The leakage section comprises a leakage pipe (14), a leakage hole (141), an electric valve (142), a single chip microcomputer (143), and a differential pressure gauge (144); the leakage hole (141) is located on the leakage pipe (14); the electric valve (142) is installed on the leakage hole (141) by means of a thread; the inlet wall surface of the electric valve (142) is flush with the inner wall surface of the leakage pipe (14); the single chip microcomputer (143) is connected to the electric valve (142); and the differential pressure gauge (143) is installed on the electric valve (142); The circulation section comprises a circulation pipeline (15), a cyclone separator (16), a gas discharge pipe (17), and a liquid discharge pipe (18); the cyclone separator (16) is connected to the circulation pipeline (15), the gas discharge pipe (17) and the liquid discharge pipe (18) are connected to the cyclone separator (16), and the liquid discharge pipe (18) is connected to the liquid storage tank (1); The detection system comprises a soil box (19), an electrode array (20), a high-density resistivity measuring device (21), and a data processing device (22); the electrode array (20) is composed of a plurality of electrodes and is inserted into the soil on the surface of the soil box; the high-density resistivity measuring device (21) is connected to the electrode array (20); and the data processing device (22) is connected to the high-density resistivity measuring device (21).
2. The buried oil pipeline leakage test device and oil spill detection method according to claim 1, characterized in that: The oil spill detection method steps are as follows: S1. The experiment begins by turning on the pump (2) and the compressor (6), adjusting the first regulating valve (3) and the second regulating valve (8) to make the mixing pipe (13) reach a predetermined leakage condition. After the condition stabilizes, the electric valve (142) is controlled by the single chip microcomputer (143) to open the leakage hole (141); S2, setting the electrode spacing, arranging the electrode array (20), recording the electrode coordinates, and collecting resistivity data ρ through a high-density resistivity measuring device (21) i ; S3, data preprocessing, removing bad pixels, and organizing data format; S4. Define the computational domain and boundaries, and determine the scope and boundary conditions of the underground space; S5. Perform forward calculation to obtain a forward geoelectric model; S6. Establish a physical information neural network model, input the measured resistivity and its spatial coordinates into the deep neural network for training, output the flow field parameters, use automatic differentiation technology to calculate the spatial gradient of these variables, construct a loss function L, and minimize the loss function L. Perform error backpropagation based on the chain rule, train the neural network parameters (w, b) until convergence, and finally output the inversion results; S7. Compare the inversion result with the forward modeling result. If the error is less than or equal to 5%, output the resistivity distribution image and finally obtain the oil spill distribution image. Otherwise, continue to train the physical information neural network until the error meets the requirements.
3. The physical information neural network according to claim 2, characterized in that: The deep neural network includes an input layer (23), a hidden layer (24) and an output layer (25), and the output flow field parameters include oil spill velocity u, water saturation S w , oil saturation S o , the loss function L is governed by the loss function L PDE And the measurement point matching loss function L Data The structure is used to ensure that the field variables predicted by the neural network satisfy the physical equations and are consistent with the currently known electric field information.