Non-straight pipeline leakage positioning method and system based on sparse blind separation

Abstract

The invention relates to a sparse blind separation-based non-straight pipeline leakage positioning method and system, and belongs to the technical field of pipeline leakage point positioning detection. The method comprises the steps that a plurality of sensors are arranged at the two ends of a non-straight pipeline respectively; preprocessing the observed and collected mixed signals; obtaining a time delay value based on the mixed signal obtained by preprocessing; and positioning the leakage point of the non-straight pipeline based on the time delay value and the sound velocity. The method and the system are suitable for leakage positioning detection of non-straight pipelines including T-shaped pipes, bent pipes, abrupt change pipes and the like, the degree of dependence on data integrity is low, and leakage positioning errors are small.

Description

Technical Field

[0001] The invention belongs to the technical field of pipeline leakage point positioning and detection, and in particular relates to a non-straight pipeline leakage positioning method and system based on sparse blind separation. Background Art

[0002] Accurately detecting the leakage point of the faulty pipeline is an important prerequisite for ensuring the orderly construction and safe operation of the pipeline network.

[0003] In the process of leak location detection, the noise sources can be classified into internal noise sources and external noise sources. External noise mainly comes from traffic noise, pedestrian noise, construction noise, etc. The internal noise of the pipeline mainly comes from noise sources such as T-type pipe joints and elbows. In the internal flow field of the pipeline, the fluid will be divided, divided, and decelerated when passing through the T-type pipe joint. The low-speed particles on the wall near the T-type pipe joint are subjected to the reverse pressure difference, and the speed is continuously reduced to zero. The mainstream then separates from the wall to form a vortex, and the vortex excites the pipe wall to form a vortex-induced noise. When the fluid flows through the elbow, the curvature of the elbow has an important influence on the flow characteristics of the fluid. The fluid forms a large pressure difference and velocity difference on the inner and outer walls of the elbow. At and near the elbow, due to the uneven centrifugal force of the particles of different speeds of the fluid, a pair of reverse symmetrical vortices, namely secondary vortices, will be generated. This secondary vortex excites the wall of the elbow to form an internal noise source of the pipeline that cannot be ignored.

[0004] When the faulty pipeline has a T-type pipe joint or an elbow, the detection difficulty is greatly increased. The traditional leakage detection and positioning method has limited detection capabilities for pipelines containing T-type pipe joints and elbows. The current mainstream pipeline leakage detection and positioning methods include: flow balance method, neural network learning method, time delay estimation method, pipeline leakage acoustic signal wavelet analysis method, etc. The flow balance method uses two flow sensors arranged upstream and downstream of the pipeline to locate by picking up the upstream and downstream flow data of the pipeline. In the prior art, a method for leak detection in the pipeline combining the flow balance method with the negative pressure wave method is disclosed. The technical essentials are to arrange multiple pressure sensors at equal intervals along the flow direction of the pipeline, one flow sensor is located at the beginning of the pipeline, and the other is located at the end of the pipeline. The data processor identifies and locates the flow data and pressure data. This method has high requirements on the accuracy of the flow meter, and the pressure sensor requires a large number of pressure sensors, resulting in high costs. It is almost impossible to locate pipelines with branches. The neural network learning method refers to using more than one sound or pressure sensor to pick up the leakage sound or pressure signal on the pipeline, and using BP learning method, Elman learning method and other neural network algorithms to analyze the leakage signal to complete the positioning detection work. In the prior art, a leakage point location detection method is also disclosed, which uses several pressure sensors to collect pipeline pressure data, normalizes all pressure data and uses them as input for training of a hierarchical neural network, and compares the training results with empirical samples to obtain the location of the leakage point. This method takes a lot of time to train the data, and the location of the leakage point is heavily dependent on empirical data, resulting in large positioning errors, and is not suitable for working conditions containing T-type pipe joints and elbows. The current mainstream pipeline leakage sound signal wavelet analysis method has limited positioning capabilities for pipelines containing T-type pipe joints and elbows. This is because the wavelet analysis method has a serious dependence on the selection of wavelet basis functions, and different wavelet basis functions cause different positioning errors. The delay estimation method has limited ability to suppress noise inside and outside the pipeline, and due to the presence of internal noise in the pipeline, the leakage signal correlation peak obtained by cross-correlation analysis is easily submerged by the noise correlation peak, resulting in a large positioning error. Summary of the invention

[0005] In view of the above problems, the present invention provides a non-straight pipeline leakage location method and system based on sparse blind separation.

[0006] The first object of the present invention is to provide a non-straight pipeline leakage location method based on sparse blind separation, comprising:

[0007] Place multiple sensors at both ends of a non-straight pipe;

[0008] Preprocessing the mixed signals collected by observation;

[0009] Based on the mixed signal obtained by preprocessing, a delay value is obtained;

[0010] Locate the leak point of non-straight pipes based on time delay value and sound velocity.

[0011] In a specific embodiment of the present invention, the preprocessing includes pre-whitening processing and frequency domain transformation.

[0012] In a specific embodiment of the present invention, obtaining the delay value based on the mixed signal obtained by preprocessing includes:

[0013] Normalizing the mixed signal obtained by preprocessing;

[0014] Based on the mixed matrix and mixed signal obtained through normalization, the delay value is obtained.

[0015] In a specific embodiment of the present invention, the delay value is obtained based on the mixing matrix and the mixed signal obtained by preprocessing, including:

[0016] Extracting source signals based on the mixed matrix and mixed signals obtained through preprocessing;

[0017] Based on the extracted source signal, the sample entropy is calculated;

[0018] The sample entropy value is used as the support vector machine;

[0019] Filter out leakage source signals based on support vector machine;

[0020] The delay value is obtained by estimating the cross-correlation delay of the leakage source signal.

[0021] In a specific embodiment of the present invention, the source signal is extracted according to the following formula:

[0022]

[0023] Among them, X and Y are mixed signals collected by observation after preprocessing, A and B are mixing matrices, S 1 , S 2 is the source signal corresponding to the mixing matrix A and B, and is also the sparse coefficient corresponding to the mixing matrix A and B, C 1 , C 2 are transfer functions, N 1 、N 2 are all noise disturbance terms.

[0024] In a specific embodiment of the present invention, the leakage point of the non-straight pipeline is located according to the following formula:

[0025]

[0026] Among them, τ is the time delay value, υ is the speed of sound, L is the propagation path length from the leak point to a certain sensor, and L 1is a known path.

[0027] The second object of the present invention is to provide a non-straight pipeline leakage location system based on sparse blind separation, comprising:

[0028] The installation module is used to place multiple sensors at both ends of a non-straight pipe;

[0029] The preprocessing module is used to preprocess the mixed signals collected by observation;

[0030] The delay module is used to obtain the delay value based on the mixed signal obtained by preprocessing;

[0031] The positioning module is used to locate the leakage point of non-straight pipelines based on the time delay value and sound velocity.

[0032] In a specific embodiment of the present invention, the delay module includes a normalization level 1 submodule and an acquisition level 1 submodule;

[0033] The normalization first-level submodule is used to normalize the mixed signal obtained by preprocessing;

[0034] The first-level submodule is used to obtain a delay value based on the mixed matrix and the mixed signal obtained by the normalization process.

[0035] In a specific embodiment of the present invention, the obtaining of the primary submodule includes extracting the secondary submodule, calculating the secondary submodule, supporting the secondary submodule, screening the secondary submodule and estimating the secondary submodule.

[0036] The extraction secondary submodule is used to extract the source signal based on the mixed matrix and mixed signal obtained by preprocessing;

[0037] The calculation secondary submodule is used to calculate the sample entropy based on the extracted source signal;

[0038] The secondary submodule is used to support the sample entropy value as a support vector machine;

[0039] The secondary screening submodule is used to screen out leakage source signals based on support vector machines;

[0040] The estimation secondary submodule is used to estimate the delay based on the cross-correlation of the leakage source signal to obtain the delay value.

[0041] A third object of the present invention is to provide an electronic device, comprising: a processor, wherein the processor is coupled to a memory;

[0042] The memory is used to store computer programs;

[0043] The processor is used to execute the computer program stored in the memory so that the electronic device executes the method as described above.

[0044] A fourth object of the present invention is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores a program or instruction, and when the program or instruction is executed on a computer, the computer executes the method as described above.

[0045] Beneficial effects of the present invention:

[0046] The present invention provides a non-straight pipeline leakage location method and system based on sparse blind separation, which is suitable for leakage location detection of non-straight pipelines containing T-shaped pipes, curved pipes, and mutant pipes;

[0047] When separating the observed signal, blind separation technology is used: sparse component analysis technology. In the pipeline leakage detection containing T-tubes and elbows, the internal noise source and the leakage signal are superimposed together. The noise caused by the T-tube joints and elbows can be considered as an additive noise, which is theoretically independent and irrelevant to the leakage signal. However, in practice, even if the signal is whitened, this internal noise source cannot remain strictly independent and irrelevant to the leakage source, and in most cases the observed signal to be separated is in an underdetermined condition; compared with empirical mode decomposition techniques such as EMD and EEMD, this method has more rigorous mathematical principles and solves the problems of under-decomposition and over-decomposition in empirical mode decomposition;

[0048] When screening leakage source signals and internal noise source signals, sample entropy and support vector machine are combined to complete the screening task. Experience shows that the sample entropy values ​​of leakage signals are basically greater than 1, while the sample entropy values ​​of internal noise signals are generally less than 1. Sample entropy reflects the degree of randomness of the signal. The larger the sample entropy value, the greater the randomness of the signal. Compared with approximate entropy, data length has little effect on the calculation result of sample entropy; sample entropy has better consistency; sample entropy has lower requirements for data integrity, indicating that the method provided by the present invention has a low degree of dependence on data integrity and a small leakage location error.

[0049] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A flowchart of a non-straight pipeline leakage location method based on sparse blind separation according to an embodiment of the present invention is shown;

[0052] Figure 2 A schematic diagram showing a plurality of sensors arranged at both ends of a T-tube according to an embodiment of the present invention is shown;

[0053] Figure 3 A schematic diagram showing a plurality of sensors arranged at both ends of a curved pipe according to an embodiment of the present invention is shown;

[0054] Figure 4 A framework diagram of a non-straight pipeline leakage location system based on sparse blind separation according to an embodiment of the present invention is shown;

[0055] Figure 5 A framework diagram of an electronic device according to an embodiment of the present invention is shown;

[0056] In the figure:

[0057] Installation module 1; pre-processing module 2; delay module 3; positioning module 4; electronic device 300; processor 301; memory 302. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, a non-straight pipeline leakage location method based on sparse blind separation according to an embodiment of the present invention includes:

[0060] Step S1, placing a plurality of sensors at both ends of the non-straight pipeline respectively;

[0061] Step S2, preprocessing the observed and collected mixed signal;

[0062] Step S3, obtaining a delay value based on the mixed signal obtained by preprocessing;

[0063] Step S4: Locate the leakage point of the non-straight pipeline based on the time delay value and the sound velocity.

[0064] In step S1, a plurality of sensors are placed at both ends of a non-straight pipe, wherein the non-straight pipe is one of a T-shaped pipe, an elbow pipe, and a sudden change pipe. In the embodiment of the present invention, a schematic diagram of arranging a plurality of sensors at both ends of a T-shaped pipe and an elbow pipe is exemplarily shown, as shown in FIG. Figure 2 and Figure 3 As shown;

[0065] The multiple sensors are respectively placed at both ends of the non-straight pipeline, that is, multiple sensors are symmetrically arranged at each end of the pipeline, and the sensors at the same end are close to each other; the specific arrangement method is: the accelerometer with a magnetic base is adsorbed on the pipeline, and the magnetic base cannot be tightened too much to avoid the accelerometer and the magnetic base being too tightly screwed together, resulting in failure to respond to the leakage signal in time.

[0066] In step S2, the mixed signal collected by observation is a mixed signal obtained by amplification and filtering by a data acquisition unit, and the signal contains a noise leakage signal;

[0067] Preprocessing the mixed signal of the noisy leakage signal, the preprocessing includes pre-whitening processing and frequency domain transformation, that is, pre-whitening processing is performed on the collected noisy leakage signal to maximize the mutual independence of the source mixed signals in the observed signal, thereby improving the effectiveness of the blind separation algorithm, wherein the pre-whitening processing is a processing technology commonly used in the technical field, and the present invention will not be repeated here;

[0068] The series of signals obtained by pre-whitening are transformed in the frequency domain, that is, the correlation of the second-order statistics in different frequency bands (correlation refers to the calculation of the mutual correlation coefficient of different frequency bands in the frequency domain) is further processed to obtain a mixed signal. In order to more clearly express the signal processing steps in the embodiment of the present invention, the mixed signal obtained by pre-processing is expressed as X=[X 1 , X 2 , …, X N ],Y=[Y 1 , Y 2 , …, Y N ];

[0069] In step S3, the delay value is obtained based on the mixed signal obtained by preprocessing, including:

[0070] Step A1, normalizing the mixed signal obtained by preprocessing;

[0071] Step A2: Obtain a delay value based on the mixed matrix and mixed signal obtained through normalization processing.

[0072] In step A1, the mixed signal obtained by preprocessing is normalized (normalized)

[0073] ), the output after processing is a mixing matrix, which is expressed as A = [A1 , A 2 , …, A N ]B=[B 1 , B 2 , …, B N ].

[0074] In step A2, the delay value is obtained based on the mixing matrix and the mixed signal obtained by preprocessing, including:

[0075] Step B1, extracting source signals based on the mixed matrix and mixed signals obtained through preprocessing;

[0076] Step B2, calculating sample entropy based on the extracted source signal;

[0077] Step B3, using the sample entropy value as a support vector machine;

[0078] Step B4: filtering out leakage source signals based on a support vector machine;

[0079] Step B5: Estimating the time delay based on the cross-correlation of the leakage source signal to obtain a time delay value.

[0080] In step B1, the source signal is extracted according to formula (1):

[0081]

[0082] Among them, S 1 , S 2 is the source signal corresponding to the mixing matrix A and B, and is also the sparse coefficient corresponding to the mixing matrix A and B, C 1 , C 2 are transfer functions, N 1 、N 2 are noise disturbance terms, among which, C 1 , C 2 is the transfer function of the leakage signal in the pipeline, which can be deduced from the pipeline propagation theory. 1 、N 2 In order to detect the on-site noise, it is actually collected by the sensor.

[0083] Formula (1) is the blind separation formula model. Formula (1) constructs the mixing matrices A and B and their sparse coefficients S 1 , S 2 The corresponding relationship between them. At this time, the purpose of the algorithm is to find a mixing matrix A and B, and a transfer function C based on the mixed signals X and Y obtained after preprocessing 1 , C 2 , so that the sparse coefficient S 1 , S 2As sparse as possible. In order to highlight the effectiveness of sparsity for blind source separation, blind source separation will be divided into two different stages:

[0084] ① The mixed source signals are uncorrelated, that is, independent of each other. The uncorrelated source signals are denoted as S I , S J Therefore, the source separation model is constructed as shown in formula (2):

[0085]

[0086] Among them, W 1 , W 2 is the separation matrix;

[0087] ②Choose a suitable transfer basis function C 1 , C 2 , and the independent source signal S I , S J Directly as the sparse coefficient S 1 , S 2 , by selecting appropriate transfer basis functions to make the sparse coefficients as sparse as possible, so as to improve the separation effect, and then realize "extraction of source signals based on the mixing matrix and mixed signals obtained by preprocessing";

[0088] That is, in step B1, the source signal S is extracted according to the mixed signal X and the mixing matrix A. I , S I =[s 1 ,s 2 ,…,s N ], extract the source signal S from the mixed signal Y and the mixing matrix B J , S J =[S 1 , S 2 , …, S N ].

[0089] In step B2, the sample entropy is calculated based on the extracted source signal, that is, the S obtained in step B1 is I and S J Calculate and obtain the sample entropy.

[0090] In step B3, the sample entropy value is used as a support vector machine, that is, the sample entropy value is used as an input value of the support vector machine SVM for classification;

[0091] In step B4, the leakage source signal is screened out based on the support vector machine, that is, in step B3, the source signal with a higher SVM screening entropy value is classified as the leakage source signal S i , S j .

[0092] In step B5, the cross-correlation delay estimation based on the leakage source signal obtains the delay value, that is, the cross-correlation function is processed using a bandpass filter based on the coherence function, which can suppress the interference of the detection environment noise and make the cross-correlation peak more obvious. The time corresponding to the peak of the cross-correlation function is the delay value.

[0093] In step S4, the leakage point of the non-straight pipeline is located according to formula (3):

[0094]

[0095] In formula (3), τ is the time delay value, υ is the speed of sound, the time delay value, L 1 Substitute the sound speed (experience value can be taken) into formula (3), that is, locate the propagation path length of the non-L leakage point to a certain sensor, L 1 is a known path, that is, the leakage point of the straight pipeline determined in the previous steps.

[0096] like Figure 4 As shown, a non-straight pipeline leakage location system based on sparse blind separation according to an embodiment of the present invention includes:

[0097] The installation module 1 is used to place multiple sensors at both ends of the non-straight pipe;

[0098] The preprocessing module 2 is used to preprocess the mixed signal collected by observation;

[0099] The delay module 3 is used to obtain a delay value based on the mixed signal obtained by preprocessing;

[0100] The positioning module 4 is used to locate the leakage point of the non-straight pipeline based on the time delay value and the sound speed.

[0101] In the embodiment of the present invention, the delay module 3 includes a normalization level 1 submodule and an acquisition level 1 submodule;

[0102] The normalization first-level submodule is used to normalize the mixed signal obtained by preprocessing;

[0103] The first-level submodule is used to obtain a delay value based on the mixed matrix and the mixed signal obtained by the normalization process.

[0104] In the embodiment of the present invention, the obtaining of the primary submodule includes extracting the secondary submodule, calculating the secondary submodule, supporting the secondary submodule, screening the secondary submodule and estimating the secondary submodule.

[0105] The extraction secondary submodule is used to extract the source signal based on the mixed matrix and mixed signal obtained by preprocessing;

[0106] The calculation secondary submodule is used to calculate the sample entropy based on the extracted source signal;

[0107] The secondary submodule is used to support the sample entropy value as a support vector machine;

[0108] The secondary screening submodule is used to screen out leakage source signals based on support vector machines;

[0109] The estimation secondary submodule is used to estimate the delay based on the cross-correlation of the leakage source signal to obtain the delay value.

[0110] like Figure 5 As shown, some embodiments of the present invention provide an electronic device, the electronic device 300, comprising: a processor 301, the processor 301 is coupled to a memory 302;

[0111] The memory 302 is used to store computer programs;

[0112] The processor 301 is configured to execute the computer program stored in the memory 302 so that the electronic device executes the method described in the above embodiment.

[0113] In certain embodiments of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a program or instruction. When the program or instruction is executed on a computer, the computer executes the method described in the above embodiments.

[0114] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an electronic device or a device.

[0115] 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 embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A non-straight pipeline leakage location method based on sparse blind separation, It is characterized in that include: Place multiple sensors at both ends of a non-straight pipe; Preprocessing the mixed signals collected by observation; Based on the mixed signal obtained by preprocessing, a delay value is obtained; Locate the leak point of non-straight pipes based on time delay value and sound velocity.

2. The non-straight pipeline leakage location method based on sparse blind separation according to claim 1, It is characterized in that The preprocessing includes pre-whitening processing and frequency domain transformation.

3. The non-straight pipeline leakage location method based on sparse blind separation according to claim 1, It is characterized in that The delay value is obtained based on the mixed signal obtained by preprocessing, including: Normalizing the mixed signal obtained by preprocessing; Based on the mixed matrix and mixed signal obtained through normalization, the delay value is obtained.

4. The non-straight pipeline leakage location method based on sparse blind separation according to claim 3, It is characterized in that The delay value is obtained based on the mixing matrix and the mixed signal obtained by preprocessing, including: Extracting source signals based on the mixed matrix and mixed signals obtained through preprocessing; Based on the extracted source signal, the sample entropy is calculated; The sample entropy value is used as the support vector machine; Filter out leakage source signals based on support vector machine; The delay value is obtained by estimating the cross-correlation delay of the leakage source signal.

5. The non-straight pipeline leakage location method based on sparse blind separation according to claim 4, It is characterized in that The source signal is extracted according to the following formula: Among them, X and Y are mixed signals collected by observation after preprocessing, A and B are mixing matrices, S 1 , S 2 is the source signal corresponding to the mixing matrix A and B, and is also the sparse coefficient corresponding to the mixing matrix A and B, C 1 , C 2 are transfer functions, N 1 、N 2 are all noise disturbance terms.

6. The non-straight pipeline leakage location method based on sparse blind separation according to any one of claims 1 to 5, It is characterized in that The leakage point of the non-straight pipeline is located according to the following formula: Among them, τ is the time delay value, υ is the speed of sound, L is the propagation path length from the leak point to a certain sensor, and L 1 is a known path.

7. A non-straight pipeline leakage location system based on sparse blind separation, It is characterized in that include: The installation module is used to place multiple sensors at both ends of a non-straight pipe; The preprocessing module is used to preprocess the mixed signals collected by observation; The delay module is used to obtain the delay value based on the mixed signal obtained by preprocessing; The positioning module is used to locate the leakage point of non-straight pipelines based on the time delay value and sound velocity.

8. The non-straight pipeline leakage location system based on sparse blind separation according to claim 7, It is characterized in that The delay module includes a normalization level 1 submodule and an acquisition level 1 submodule; The normalization first-level submodule is used to normalize the mixed signal obtained by preprocessing; The first-level submodule is used to obtain a delay value based on the mixed matrix and the mixed signal obtained by the normalization process.

9. The non-straight pipeline leakage location system based on sparse blind separation according to claim 8, It is characterized in that The obtaining of the primary submodule includes extracting the secondary submodule, calculating the secondary submodule, supporting the secondary submodule, screening the secondary submodule and estimating the secondary submodule. The extraction secondary submodule is used to extract the source signal based on the mixed matrix and mixed signal obtained by preprocessing; The calculation secondary submodule is used to calculate the sample entropy based on the extracted source signal; The secondary submodule is used to support the sample entropy value as a support vector machine; The secondary screening submodule is used to screen out leakage source signals based on support vector machines; The estimation secondary submodule is used to estimate the delay based on the cross-correlation of the leakage source signal to obtain the delay value.

10. An electronic device, It is characterized in that include: a processor coupled to the memory; The memory is used to store computer programs; The processor is configured to execute the computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 6.

11. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a program or an instruction, and when the program or the instruction is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 6.

Images