Pipeline leakage detection method and device and simulation verification system

Abstract

The invention provides a pipeline leakage detection method and device and a simulation verification system, and the method comprises the steps: collecting the ground dissipation current of a pipe wall when a pipeline is in a normal flowing state, and obtaining a pipe wall dissipation current signal; extracting a characteristic value of the tube wall dissipation current signal, and generating a training sample; training by using the training sample to obtain a leakage detection model, wherein the leakage detection model is a model based on an auto-encoder; and performing pipeline leakage detection by using the leakage detection model. By means of the scheme, accurate and effective pipeline leakage detection can be achieved, and the applicability to detection scenes is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline protection, and in particular to a pipeline leakage detection method and device, and also to a pipeline leakage detection simulation verification system. Background Art

[0002] During the pipeline transportation of dielectric liquids such as petroleum, charge separation occurs due to friction with the pipe wall, thereby forming an equal amount of opposite-sign double electric layers on the contact surface between the pipe wall and the oil product. A large amount of static charges are generated inside the pipe wall and the liquid. The static charges inside the liquid will flow with the liquid, generating a flow current. In actual industry, due to the fact that static electricity can cause great harm to industrial production, the method of grounding the pipeline is often used to allow the static charges on the pipe wall to dissipate to the ground through the grounding wire. This process will generate a dissipation current.

[0003] When a pipeline leaks, the leaked medium may volatilize and form an explosive mixed gas with the air. If the pipeline is not effectively grounded or the grounding fails, there is a possibility of static discharge causing a fire and explosion accident. Therefore, developing an effective pipeline leakage detection method is of great significance for ensuring the safety of pipeline transportation.

[0004] Currently, the commonly used pipeline leakage detection methods mainly include methods based on pressure, acoustics, temperature, etc. Each method has its applicable scenarios and limitations. Summary of the Invention

[0005] On the one hand, the present invention provides a pipeline leakage detection method and device to achieve accurate and effective pipeline leakage detection and improve the applicability to the detection scenario.

[0006] On the other hand, the present invention also provides a pipeline leakage detection simulation verification system, which can fully verify the accuracy of the established leakage detection model.

[0007] For this reason, the present invention provides the following technical solutions:

[0008] A pipeline leakage detection method, the method comprising:

[0009] Collect the dissipation current of the pipe wall to the ground in the normal flow state of the pipeline to obtain a pipe wall dissipation current signal;

[0010] Extract the characteristic values of the pipe wall dissipation current signal to generate training samples;

[0011] Use the training samples to train a leakage detection model, and the leakage detection model is a model based on an autoencoder;

[0012] Use the leakage detection model to perform pipeline leakage detection.

[0013] Optionally, the eigenvalue includes any one or more of the following: average value, root mean square amplitude, standard deviation, variance, kurtosis factor, margin factor, frequency domain energy, frequency domain amplitude, average power, and spectral centroid.

[0014] Optionally, the method further includes: performing noise reduction processing on the wall leakage current signal;

[0015] The extracting the eigenvalue of the wall leakage current signal and generating a training sample includes: extracting the eigenvalue of the wall leakage current signal after noise reduction processing and generating a training sample.

[0016] Optionally, the performing noise reduction processing on the wall leakage current signal includes:

[0017] Decomposing the wall leakage current signal to obtain a plurality of intrinsic mode components;

[0018] Performing signal reconstruction according to the energy proportion of the plurality of intrinsic mode components.

[0019] Optionally, the decomposing the wall leakage current signal to obtain a plurality of intrinsic mode components includes:

[0020] Using a variational mode decomposition algorithm to decompose the wall leakage current signal to obtain a plurality of intrinsic mode components.

[0021] Optionally, the method further includes:

[0022] Performing standardization processing on the eigenvalue to obtain a standardized sample eigenvalue;

[0023] Screening the standardized sample eigenvalue by a principal component analysis method to obtain a dimensionality-reduced eigenvalue.

[0024] Optionally, the method further includes:

[0025] Training a plurality of different types of autoencoder leakage detection models;

[0026] Comparing the performance of each autoencoder leakage detection model and selecting the optimal autoencoder leakage detection model.

[0027] Optionally, the plurality of different types of autoencoders include any two or more of the following: sparse autoencoder, denoising autoencoder, and convolutional autoencoder.

[0028] A pipeline leakage detection device, the device includes:

[0029] A current acquisition module, configured to acquire the wall-to-ground leakage current of the pipeline in a normal flow state to obtain a wall leakage current signal;

[0030] A sample generation module, configured to extract the eigenvalue of the wall stray current signal and generate training samples;

[0031] A model training module, configured to train a leakage detection model by using the training samples, where the leakage detection model is a model based on an autoencoder;

[0032] A detection module, configured to perform pipeline leakage detection by using the leakage detection model.

[0033] Optionally, the device further includes:

[0034] A noise reduction processing module, configured to perform noise reduction processing on the wall stray current signal;

[0035] The sample generation module extracts the eigenvalue of the wall stray current signal after noise reduction processing and generates training samples.

[0036] Optionally, the device further includes:

[0037] A normalization module, configured to perform normalization processing on the eigenvalue to obtain a normalized sample eigenvalue;

[0038] A dimensionality reduction processing module, configured to screen the normalized sample eigenvalue by using a principal component analysis method to obtain a dimensionality reduction eigenvalue.

[0039] A computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the pipeline leakage detection method are executed.

[0040] A computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the pipeline leakage detection method are implemented.

[0041] A pipeline leakage detection simulation verification system, the system includes: a pipeline transportation system and a signal acquisition and processing system:

[0042] The pipeline transportation system includes: a storage tank and a dissipation tank connected by a connecting pipeline; an insulating pipe section and a test pipe section are arranged on the connecting pipeline, and two ends of the test pipe section are connected to the connecting pipeline through the insulating pipe section; a controllable leakage port is arranged on the test pipe section;

[0043] The storage tank stores a flowable medium;

[0044] The dissipation tank is configured to allow the medium in the pipeline to stand in the dissipation tank for a certain time before flowing into the test pipe section, so that the electric charge carried by the medium when flowing into the test pipe section is zero;

[0045] The signal acquisition and processing system comprises: a micro-ammeter and a computer; the input end of the micro-ammeter is connected to the pipe wall of the test pipe section, and the output end of the micro-ammeter is connected to the computer;

[0046] The micro-ammeter is grounded and used to measure the current dissipated from the pipe wall to the ground;

[0047] The computer is used to establish a leakage detection model according to the leakage current from the pipe wall to the ground, and to verify the accuracy of the leakage detection model.

[0048] Optionally, the controllable leakage port is a controllable ball valve.

[0049] The pipeline leakage detection method and device provided by the present invention obtains the pipeline leakage current signal by collecting the pipeline wall leakage current to the ground when the pipeline is in a normal flow state, extracts characteristic values ​​from the pipeline wall leakage current signal to generate training samples, and obtains a leakage detection model based on an autoencoder through training, and uses the leakage detection model to perform pipeline leakage detection. The scheme of the present invention performs pipeline leakage detection through pipeline electrostatic current signals, without the need to install pressure, acoustic and other sensors in advance. Moreover, the scheme of the present invention is not limited by the pipeline environment, and can be applied to a variety of detection scenarios to achieve real-time detection of leaking pipelines.

[0050] The pipeline leakage detection simulation verification system provided by the present invention can simulate the normal flow state and the pipeline leakage state of the pipeline, and by collecting the pipe wall to ground leakage current in the two states, using the training sample data to establish a leakage detection model based on the autoencoder, and using the sample data to be tested to test the leakage detection model, it can further prove the feasibility and reliability of the pipeline leakage detection method and device provided by the present invention in realizing pipeline leakage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 It is a flow chart of the pipeline leakage detection method provided by the present invention;

[0053] Figure 2 It is a structural schematic diagram of the pipeline leakage detection device provided by the present invention;

[0054] Figure 3 It is a structural schematic diagram of a pipeline leakage detection simulation verification system provided by the present invention;

[0055] Figure 4 It is the flow chart of the verification process of the leakage detection model in the method of the present invention by using the pipeline leakage detection simulation verification system shown in Figure 3 .

[0056] Reference numerals:

[0057] 1. Storage tank; 2. Pump; 3. Valve; 4. Connecting pipeline; 5. Escape tank; 6. Insulated pipe section; 7. Test pipe section; 8. Simulated leakage orifice; 9. Microammeter; 10. Computer. Specific implementation manners

[0058] In order to enable those skilled in the art of the present technology to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0059] In the description of the present invention, it should be noted that, unless otherwise clearly defined and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0060] As Figure 1 shown, it is a flow chart of a pipeline leakage detection method provided by the present invention, including the following steps:

[0061] Step 101, collect the stray current from the pipe wall to the ground in the normal flow state of the pipeline to obtain the pipe wall stray current signal.

[0062] For example, a microammeter can be used to collect the stray current from the pipe wall to the ground in the normal flow state. First, the microammeter is reliably connected to the pipe section to be measured, and it is ensured that the microammeter is reliably grounded. Then, the grounding wire of the pipeline is disconnected so that the static charge generated on the pipe wall can flow into the ground through the microammeter. When the flow state in the pipe is stable, the acquisition function of the microammeter is turned on to collect the stray current value in the normal flow state as the training sample data.

[0063] It is understandable that when media such as oil products are transported in a pipeline, friction with the pipe wall will generate an equal amount of opposite-sign double electric layers at the solid-liquid interface. Among them, the charges on the pipe wall can be dissipated to the ground through a grounding wire, forming a dissipation current, and a microammeter can be used to measure the value of the dissipation current of the pipe wall. When the pipeline is flowing normally, the value of the dissipation current of the pipe wall to the ground is stable; when the pipeline leaks, the leaked fluid quickly rubs against the pipe wall, increasing the static charges on the pipe wall and causing the value of the dissipation current of the pipe wall to the ground to increase. Therefore, by analyzing the change in the value of the dissipation current, it can be determined whether the pipeline leaks.

[0064] Step 102: Extract the characteristic values of the dissipation current signal of the pipe wall to generate training samples.

[0065] The characteristic values include, but are not limited to, any one or more of the following: average value, root mean square amplitude, standard deviation, variance, kurtosis factor, margin factor, frequency domain energy, frequency domain amplitude, average power, spectral centroid, etc.

[0066] In some embodiments, noise reduction processing can also be performed on the dissipation current signal of the pipe wall.

[0067] Specifically, decompose the dissipation current signal of the pipe wall to obtain a number of intrinsic mode components. For example, algorithms such as variational mode decomposition (VMD) can be used, but are not limited to, to decompose the dissipation current signal of the pipe wall. VMD is a signal processing technology that estimates each signal component by solving a frequency domain variational optimization problem. Specifically, the frequency center and bandwidth of each component in the signal are determined by iteratively searching for the optimal solution of the variational model. This method can adaptively achieve the frequency domain dissection of the signal and the effective separation of each component. Then, signal reconstruction is performed according to the energy ratio of the number of intrinsic mode components obtained by decomposition.

[0068] Through noise reduction processing, the extracted characteristic values can better reflect the characteristics of the leaking pipeline.

[0069] In other embodiments, the principal component analysis method can also be used to screen the characteristic values to obtain reduced-dimensional characteristic values; the reduced-dimensional characteristic values are standardized to obtain standardized sample characteristic values.

[0070] Through dimensionality reduction processing, the computational amount of model training can be greatly simplified, and the model training efficiency can be improved.

[0071] Step 103: Use the training samples to train a leakage detection model, and the leakage detection model is a model based on an autoencoder.

[0072] The autoencoder model mainly consists of an encoder and a decoder. Its main purpose is to convert the input x into an intermediate variable y, and then convert y into an output Then compare the input x and the output

[0073] The autoencoder can have various types, such as but not limited to any of the following: sparse autoencoder, denoising autoencoder, convolutional autoencoder, etc.

[0074] The leakage detection model is a three-layer neural network structure, namely the input layer, the encoding layer, and the output layer.

[0075] Select the mean squared error (MSE) and the mean absolute error (MAE) as thresholds. Assume the training samples are:

[0076] X = (X 1 , X 2 , …, Xn)

[0077] The result reconstructed by the autoencoder is:

[0078]

[0079] MSE is calculated by the following formula:

[0080]

[0081] MAE is calculated by the following formula:

[0082]

[0083] Through training, appropriate MSE and MAE thresholds can be selected.

[0084] Step 104, use the leakage detection model to detect pipeline leakage.

[0085] Specifically, collect the stray current from the pipe wall to the ground of the pipeline to be detected to obtain the stray current signal of the pipe wall; extract the eigenvalues of the stray current signal of the pipe wall, input the eigenvalues into the trained leakage detection model, and determine whether the pipeline leaks according to the output of the model.

[0086] It should be noted that for the stray current from the pipe wall to the ground of the pipeline to be detected, it is also possible to first perform noise reduction processing on it, and then extract the eigenvalues of the stray current from the pipe wall to the ground after noise reduction.

[0087] Furthermore, the extracted eigenvalues can also be standardized and dimension-reduced.

[0088] The pipeline leakage detection method provided by the embodiment of the present invention performs pipeline leakage detection through the pipeline static current signal. It is not necessary to install sensors such as pressure and acoustics in advance, and can perform leakage detection by analyzing the pipeline static current signal, which is simple and convenient, and has high accuracy.

[0089] Further, in order to obtain better model performance, in another embodiment, multiple different types of autoencoders can be selected respectively, such as sparse autoencoders, denoising autoencoders, convolutional autoencoders, etc. Using the training samples generated in the above step 102, train multiple different types of autoencoder leakage detection models for pipeline leakage detection respectively, compare the time cost, detection accuracy rate, etc. of different types of model autoencoder leakage detection models, and determine the optimal autoencoder leakage detection model as the final leakage detection model, so that the leakage detection model has better performance, thereby improving the detection effect.

[0090] Correspondingly, the embodiment of the present invention also provides a pipeline leakage detection device, as Figure 2 shown, which is a schematic structural diagram of the device.

[0091] The pipeline leakage detection device 200 includes the following modules:

[0092] The current acquisition module 201 is used to acquire the wall-to-ground dissipation current of the pipeline in the normal flow state to obtain the wall dissipation current signal;

[0093] The sample generation module 202 is used to extract the eigenvalue of the wall dissipation current signal to generate a training sample;

[0094] The model training module 203 is used to train a leakage detection model 20 using the training sample, and the leakage detection model 20 is a model based on an autoencoder;

[0095] The detection module 204 is used to perform pipeline leakage detection using the leakage detection model 20.

[0096] The current acquisition module 201 can be, for example, a microammeter or other devices.

[0097] In a non-limiting embodiment, the pipeline leakage detection device 200 may further include: a noise reduction processing module (not shown in the figure), which is used to perform noise reduction processing on the wall dissipation current signal. Correspondingly, the sample generation module 202 extracts the eigenvalue of the wall dissipation current signal after noise reduction processing to generate a training sample.

[0098] In another non-limiting embodiment, the pipeline leakage detection device 200 may further include: a standardization module and a dimensionality reduction processing module (not shown in the figure). Among them:

[0099] The standardization module is used to standardize the eigenvalue to obtain a standardized sample eigenvalue; the dimensionality reduction processing module is used to screen the standardized sample eigenvalue by the principal component analysis method to obtain a dimensionality reduction eigenvalue.

[0100] Correspondingly, the model training module 203 uses the dimensionality-reduced eigenvalue as a training sample to train and obtain a leakage detection model 20.

[0101] In some embodiments, the model training module 203 can respectively select various different types of autoencoders, such as sparse autoencoders, denoising autoencoders, convolutional autoencoders, etc., and train various different types of autoencoder leakage detection models for pipeline leakage detection, compare the time cost, detection accuracy, etc. of different types of model autoencoder leakage detection models, and determine the optimal autoencoder leakage detection model as the final leakage detection model, so that the leakage detection model has better performance, thereby improving the detection effect.

[0102] For more descriptions of the above modules, reference can be made to the descriptions in the embodiments of the present invention above, and details are not described herein again.

[0103] The pipeline leakage detection device provided by the embodiments of the present invention performs pipeline leakage detection through the pipeline static current signal, without the need to install sensors such as pressure and acoustics in advance, and can perform leakage detection by analyzing the pipeline static current signal, which is simple and convenient, and has high accuracy.

[0104] Using the pipeline leakage detection method and device provided by the embodiments of the present invention, the real-time detection of the pipeline state can be simply and conveniently realized, without being restricted by the pipeline environment, and can be applied to a variety of detection scenarios.

[0105] In order to fully verify the accuracy of the leakage detection model established by the embodiments of the present invention, the embodiments of the present invention also provide a pipeline leakage detection simulation verification system, as Figure 3 shown, which is a schematic structural diagram of the system.

[0106] Referring to Figure 3 , the pipeline leakage detection simulation verification system includes a pipeline transportation system and a signal acquisition and processing system. Among them:

[0107] The pipeline transportation system includes: a storage tank 1 and a dispersion tank 5 connected by a connecting pipeline 4; an insulating pipe section 6 and a test pipe section 7 are arranged on the connecting pipeline 4, and both ends of the test pipe section 7 are connected to the connecting pipeline 4 through the insulating pipe section 6. Through the insulating pipe section 6, the static electricity generated on the test pipe section 7 can be made not affected by other components.

[0108] A simulated leakage port 8 is arranged on the test pipe section 7.

[0109] The storage tank 1 stores a flowable medium;

[0110] The dissipation tank 5 is used to allow the medium in the pipeline to stand in the dissipation tank 5 for a certain period of time before flowing into the test pipe section 7, so that the charge carried by the medium when flowing into the test pipe section 7 is zero, and will not affect the electrostatic measurement on the test pipe section 7;

[0111] The test pipe section 7 is then connected to the storage tank 1 through the connecting pipe 4 to achieve the circulation flow of the medium.

[0112] The signal acquisition and processing system includes: a micro-ammeter 9 and a computer 10; the input end of the micro-ammeter 9 is connected to the wall of the test pipe section 7, and the output end of the micro-ammeter 9 is connected to the computer 10;

[0113] In this embodiment, the microammeter 9 is grounded to measure the leakage current from the pipe wall to the ground; the computer 10 is used to establish a leakage detection model according to the leakage current from the pipe wall to the ground, and to verify the accuracy of the leakage detection model.

[0114] The microammeter 9 is grounded, and the test pipe section 7 is not grounded. The static charge generated by the flow of the medium on the test pipe section 7 flows through the microammeter 9 and dissipates to the ground through the grounding wire of the microammeter 9 to measure the dissipated current of the pipe wall to the ground.

[0115] like Figure 3 As shown, a pump 2 and a valve 3 are sequentially arranged on the connecting pipe 4 on one side between the storage tank 1 and the escape tank 5 to control the flow rate of the medium out of the storage tank 1 and in the pipe.

[0116] The simulated leakage port 8 may be simulated by a ball valve, and the ball valve may be a manual ball valve or an electric ball valve, which is not limited in the embodiment of the present invention.

[0117] The process of verifying the accuracy of the leakage detection model established by the scheme of the present invention using the pipeline leakage detection simulation verification system is as follows: Figure 4 shown.

[0118] In step 401, the pipe wall-to-ground leakage current signals are collected in a normal flow state and a leakage state to obtain training sample data and test sample data.

[0119] Reference Figure 3, first, reliably connect the microammeter 9 to the test pipe section 7 and ensure that the microammeter 9 is reliably grounded. Then disconnect the grounding wire of the pipeline so that the static charges generated on the pipe wall flow into the ground through the microammeter 9. When the flow state in the pipe is stable, turn on the acquisition function of the microammeter 9 to acquire the dissipated current value under the normal flow state as the training sample data. When the pipeline flow state is stable, turn on the acquisition function of the microammeter 9, and then open the simulated leakage port 8 on the test pipe section 7 to cause leakage in the pipeline, and measure the dissipated current value on the pipe wall during leakage as the sample data to be measured.

[0120] In step 402, establish a leakage detection model based on an autoencoder using the training sample data.

[0121] The method for establishing the leakage detection model can be referred to the description in the embodiment shown above, and will not be elaborated here. Figure 1 shown in the above embodiments and will not be repeated here.

[0122] In step 403, input the sample data to be measured into the autoencoder of the leakage detection model to obtain the reconstructed data.

[0123] In step 404, determine the performance of the leakage detection model according to the difference between the sample data to be measured and the reconstructed data.

[0124] The smaller the difference between the sample to be measured and the reconstructed data, the higher the accuracy of the leakage detection model.

[0125] Correspondingly, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, it executes Figure 1 all or part of the steps of the pipeline leakage detection method.

[0126] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not elaborated in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] In several embodiments provided by the present invention, it should be understood that the disclosed device can also be implemented in other ways.

[0128] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless manner.

[0129] The above embodiments of the present invention have been described in detail. Specific implementation manners have been used in this article to elaborate on the present invention. The descriptions of the above embodiments are only used to help understand the method and system of the present invention. They are only some of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. The content of this specification should not be construed as a limitation to the present invention. Therefore, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A pipeline leakage detection method, characterized in that: The method comprises: Collect the pipe wall-to-ground leakage current when the pipeline is in normal flow state to obtain the pipe wall leakage current signal; Extracting characteristic values ​​of the pipe wall escaped current signal to generate training samples; Using the training samples to train a leakage detection model, the leakage detection model is a model based on an autoencoder; The leakage detection model is used to perform pipeline leakage detection.

2. The pipeline leakage detection method according to claim 1, characterized in that: The characteristic values ​​include any one or more of the following: mean value, root amplitude, standard deviation, variance, kurtosis factor, margin factor, frequency domain energy, frequency domain amplitude, average power, and spectrum centroid.

3. The pipeline leakage detection method according to claim 1, characterized in that: The method further includes: performing noise reduction processing on the pipe wall stray current signal; The extracting the characteristic value of the pipe wall leakage current signal and generating a training sample comprises: The characteristic value of the pipe wall escaped current signal after noise reduction is extracted to generate training samples.

4. The pipeline leakage detection method according to claim 3, characterized in that: The noise reduction process of the pipe wall leakage current signal comprises: Decomposing the pipe wall leakage current signal to obtain a plurality of inherent modal components; Signal reconstruction is performed according to the energy proportions of the plurality of natural mode components.

5. The pipeline leakage detection method according to claim 4, characterized in that: The decomposition of the pipe wall leakage current signal to obtain several natural modal components includes: The pipe wall escape current signal is decomposed by using a variational mode decomposition algorithm to obtain a number of inherent mode components.

6. The pipeline leakage detection method according to claim 1, characterized in that: The method further comprises: Performing standardization on the characteristic values ​​to obtain standardized sample characteristic values; The standardized sample eigenvalues ​​are screened by a principal component analysis method to obtain reduced dimension eigenvalues.

7. The pipeline leakage detection method according to any one of claims 1 to 6, characterized in that: The method further comprises: Train various types of autoencoder leakage detection models; Compare the performance of each encoder leakage detection model and select the best autoencoder leakage detection model.

8. The pipeline leakage detection method according to claim 7, characterized in that: The multiple different types of autoencoders include any two or more of the following: sparse autoencoders, denoising autoencoders, and convolutional autoencoders.

9. A pipeline leakage detection device, characterized in that: The device comprises: The current acquisition module is used to collect the pipe wall-to-ground leakage current when the pipeline is in a normal flow state, and obtain the pipe wall leakage current signal; A sample generation module, used to extract the characteristic value of the pipe wall escape current signal and generate a training sample; A model training module, used to train a leakage detection model using the training samples, wherein the leakage detection model is a model based on an autoencoder; The detection module is used to perform pipeline leakage detection using the leakage detection model.

10. The pipeline leakage detection device according to claim 9, characterized in that: The device also includes: A noise reduction processing module, used for performing noise reduction processing on the pipe wall escape current signal; The sample generation module extracts the characteristic value of the pipe wall escaped current signal after noise reduction processing to generate a training sample.

11. The pipeline leakage detection device according to claim 10, characterized in that: The device also includes: A standardization module, used for performing standardization processing on the characteristic value to obtain a standardized sample characteristic value; The dimension reduction processing module is used to screen the standardized sample characteristic values ​​through the principal component analysis method to obtain the dimension reduction characteristic values.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the pipeline leakage detection method according to any one of claims 1 to 8 are executed.

13. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the pipeline leakage detection method according to any one of claims 1 to 8 are implemented.

14. A pipeline leakage detection simulation verification system, characterized in that: The system includes: pipeline transportation system and signal acquisition and processing system: The pipeline transportation system comprises: a storage tank and a fugitive tank connected by a connecting pipeline; an insulating pipe section and a test pipe section are arranged on the connecting pipeline, and both ends of the test pipe section are connected to the connecting pipeline through the insulating pipe section; and a controllable leakage port is arranged on the test pipe section; The storage tank stores a flowable medium; The dissipation tank is used to allow the medium in the pipeline to stand in the dissipation tank for a certain period of time before flowing into the test pipe section, so that the charge carried by the medium when flowing into the test pipe section is zero; The signal acquisition and processing system comprises: a micro-ammeter and a computer; the input end of the micro-ammeter is connected to the pipe wall of the test pipe section, and the output end of the micro-ammeter is connected to the computer; The micro-ammeter is grounded and used to measure the current dissipated from the pipe wall to the ground; The computer is used to establish a leakage detection model according to the leakage current from the pipe wall to the ground, and to verify the accuracy of the leakage detection model.

15. The pipeline leakage detection simulation verification system according to claim 14, characterized in that: The controllable leakage port is a controllable ball valve.

Images