A DVS-based natural gas pipeline leakage risk monitoring and early warning system and method

Abstract

The application discloses a natural gas pipeline leakage risk monitoring and early warning system and method based on DVS, the monitoring and early warning system is composed of hardware and software, the hardware includes an upper computer, a distributed optical fiber vibration sensing DVS device and a sensing optical cable, the software includes a distributed optical fiber vibration sensing system, a natural gas pipeline leakage identification system and a natural gas pipeline leakage risk monitoring and early warning system; the distributed optical fiber vibration sensing DVS device automatically sends one-dimensional vibration data collected by the sensing optical cable to the upper computer according to a sampling interval through Websocket communication; the upper computer analyzes the received one-dimensional vibration data through a software part, identifies, judges and visually displays whether the natural gas pipeline state is abnormal, and outputs a final suspected abnormal result. The application has high intelligent level by integrating hardware devices and software algorithms, and can realize real-time monitoring and accurate early warning of the natural gas pipeline leakage risk.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of natural gas pipeline leakage monitoring and early warning, and particularly relates to a natural gas pipeline leakage risk monitoring and early warning system and method based on DVS. BACKGROUND

[0002] The natural gas pipeline leakage risk monitoring and early warning system as a key safety guarantee technology has achieved certain results in actual application, but there are still some problems to be solved, mainly including imperfect system, high false alarm rate, inaccurate positioning and the like:

[0003] 1. Large system limitation: Some monitoring technologies such as leakage monitoring technology based on simulation have great limitations. For example, once the pipeline is changed or a branch line is added, the core software must be adjusted a lot, otherwise there will be false and missed reports. In addition, the system may have a risk of missing detection when facing small-scale leakage, especially in the early stage of small leakage or leakage, it may be difficult to detect the leakage event in time, resulting in delayed response.

[0004] 2. High false alarm rate: In the aspect of third-party construction damage monitoring, due to the lack of unified DVS sound wave data samples and complex background noise around the pipeline, the intelligent recognition algorithm of DVS has a high false alarm rate in the application process. The false alarm of the optical fiber early warning system mainly occurs in the area where the optical cable crosses the road. The continuous driving of vehicles causes ground vibration, which causes the system to alarm. In addition, the use of water pumps by villagers around the pipeline to irrigate the land will also cause false alarms of the optical fiber.

[0005] 3. Inaccurate positioning: The monitoring technology based on optical fiber strain (BOTDA) in monitoring natural gas pipeline leakage, as the diameter of the natural gas pipeline increases, the distance between the detection optical fiber and the leakage point is longer, and it is necessary to study how the optical fiber effectively transmits the temperature field, transmits the time, and accurately judges the change of the temperature field formed by the optical cable. It may need multiple optical cables to complete the monitoring, and the equipment cost will be further increased. In addition, the various geological conditions of the area where the pipeline passes make the heat transfer effect have great differences, and the possibility of missed and delayed alarm is larger.

[0006] Therefore, it is necessary to further optimize and improve the existing natural gas pipeline leakage risk monitoring and early warning system to improve the perfection of the system, reduce the false alarm rate, and improve the positioning accuracy, so as to better guarantee the safe operation of the natural gas pipeline. SUMMARY

[0007] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a natural gas pipeline leakage risk monitoring and early warning system and method based on DVS, which realizes real-time monitoring and early warning of natural gas pipeline leakage risk by integrating advanced hardware equipment and software algorithms.

[0008] To achieve the above objectives, the present invention adopts the following technical solution.

[0009] First, this invention provides a natural gas pipeline leakage risk monitoring and early warning system based on DVS. The monitoring and early warning system consists of two parts: hardware and software. The hardware includes a host computer, a distributed fiber optic vibration sensor (DVS) device, and a sensing optical cable. The DVS device establishes a real-time connection with the host computer via WebSocket communication. The DVS device is also connected to the sensing optical cable, which is laid in the same trench as the natural gas pipeline. The sensing optical cable is used to collect temperature, strain, and vibration information at any point along the natural gas pipeline in real time. The DVS device converts the real-time collected vibration information into one-dimensional vibration data and sends it to the host computer. The host computer parses the received one-dimensional vibration data through software, identifies, judges, and visualizes whether the natural gas pipeline is in an abnormal state.

[0010] The software includes a distributed fiber optic vibration sensing system, a natural gas pipeline leak identification system, and a natural gas pipeline leak risk monitoring and early warning system. The distributed fiber optic vibration sensing system includes a data receiving module and a data output module. The data receiving module receives vibration information collected in real time by the sensing fiber optic cable and converts it into one-dimensional vibration data, which is then sent to the data output module. The data output module converts the one-dimensional vibration data into binary form and transmits it to the host computer's natural gas pipeline leak risk monitoring and early warning system. The natural gas pipeline leak identification system includes a data analysis module, a data conversion module, and an identification module. The data analysis module has a built-in feature value filtering algorithm to filter out outliers in the one-dimensional vibration data. The data conversion module converts the filtered one-dimensional vibration data into a two-dimensional SDP image. The identification module includes a convolutional neural network. The system includes a CNN model and a model training unit. The CNN model is used to identify, judge, and output suspected anomalies in natural gas pipelines from the converted two-dimensional SDP images. The model training unit is used to train the CNN model according to a preset training mode and automatically update the trained CNN model to a new CNN model. The natural gas pipeline leakage risk monitoring and early warning system includes a visualization module, a display module, and a data storage module. The visualization module is used to convert the output suspected anomalies in natural gas pipelines into specific leakage data, leakage early warning SDP images, latitude and longitude information, and time-domain maps. The display module is used to display the various information converted by the visualization module. The data storage module is used to classify and store leakage data and non-leakage data.

[0011] Specifically, the network communication between the host computer and the distributed fiber optic vibration sensing (DVS) device is based on the TCP / IP protocol and uses a client / server (C / S) structure connection. The data upload of the distributed fiber optic vibration sensing (DVS) device adopts an automatic transmission method. That is, after the host computer issues a data acquisition command, the distributed fiber optic vibration sensing (DVS) device automatically sends the temperature, strain, and vibration information of any point along the natural gas pipeline collected by the sensing fiber optic cable to the connected host computer according to the sampling interval.

[0012] Specifically, the natural gas pipeline leakage risk monitoring and early warning system adopts a front-end and back-end separation architecture and a three-tier web application architecture based on the B / S model. The front-end includes a client, i.e., a display module, and the back-end includes a server, i.e., a visualization module and a data storage module. The visualization module uses Vue 2.0 and axios to realize real-time data display and visualization of pipeline early warning status. The display module can display real-time leakage data, leakage early warning SDP images, latitude and longitude information, and time domain graphs.

[0013] Based on the aforementioned natural gas pipeline leakage risk monitoring and early warning system, this invention also provides an early warning method for the natural gas pipeline leakage risk monitoring system, comprising the following steps:

[0014] Step S1: The distributed fiber optic vibration sensor (DVS) automatically transmits the temperature, strain, and vibration information collected by the sensing fiber optic cable at any point along the natural gas pipeline to the host computer at sampling intervals via Websocket communication.

[0015] Step S2: The host computer analyzes and performs secondary identification on the received vibration information, and outputs the final suspected abnormal result.

[0016] Specifically, the secondary identification includes two parts: feature value screening and buried natural gas pipeline leak identification. These are implemented based on a feature value screening algorithm and a buried natural gas pipeline leak identification algorithm based on wavelet transform and the symmetric point pattern SDP algorithm, respectively. The steps are as follows:

[0017] Step S21: Narrow down the scope of suspected abnormal one-dimensional vibration data through feature value screening algorithm to reduce the amount of data identified by convolutional neural network (CNN) model;

[0018] Step S22: Using a natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm, one-dimensional vibration data is converted into two-dimensional SDP image;

[0019] The wavelet transform reduces noise by suppressing useless signals and enhancing useful signals, while preserving the local peak characteristics of useful signals. To ensure the adaptive properties of the wavelet function, the wavelet basis function is determined based on the wavelet entropy selection method. The wavelet entropy is calculated by quantizing the energy distribution of the wavelet coefficients at each decomposition level.

[0020] The symmetric point mode SDP algorithm can convert one-dimensional vibration data into a two-dimensional SDP image. The conversion formula is as follows:

[0021] ;

[0022] In the above formula, Yes transpose; The first in the time domain signal i The range of one point; and These represent the maximum and minimum values ​​of the time-domain signal, respectively; τ is the time interval factor, 1≤τ≤10; θ is the mirror symmetry angle, θ=60°; ζ represents the gain of the plotting angle, ζ≤θ; It is the first i The polar coordinate radius of each point; and These are the clockwise and counterclockwise rotation angles corresponding to the polar axis, respectively.

[0023] Step S23: Based on the convolutional neural network (CNN) model, identify the transformed two-dimensional SDP image to determine whether the risk status of the natural gas pipeline is normal or abnormal.

[0024] The Convolutional Neural Network (CNN) model comprises four convolutional layers, four pooling layers, two fully connected layers, and a softmax activation function. The specific formulas for the convolutional and pooling layers are as follows:

[0025] ;

[0026] ;

[0027] In the above formula, It is the first layer of the l-th layer. j One element; It is the first j Feature maps of the (l-1)th layer of each convolutional region; It is the weight matrix of the l-th layer; It is the bias; f(*) is the nonlinear activation function; The weight of the l-th layer is represented by the first... j Each feature map; down (*) is the downsampling function;

[0028] After multiple convolution and pooling operations, the fully connected layer transforms the multidimensional feature representation into a one-dimensional vector; the activation function of the last layer of the fully connected layer is softmax, and the mathematical expression of the softmax activation function is as follows:

[0029] ;

[0030] In the above formula, is the value of the nth category output by the fully connected layer, and K is the total number of categories in the fully connected layer; It is the predicted probability of the nth category;

[0031] Step S24: Display the identification results and related information based on the visualization module.

[0032] The calculation steps of the feature value filtering algorithm described in step S21 are as follows:

[0033] Step S211, Grouping

[0034] Let the set of all channels be . Where m is the total number of channels; C is divided into k groups, each containing q channels, i.e. Where h = 1, 2, ..., k;

[0035] Step S212: Calculate the statistics for each channel.

[0036] For each channel Calculate its average value ,variance and peak :

[0037] ;

[0038] ;

[0039] ;

[0040] In the above formula, It is the first in the channel b Data points, a It is the total number of data points in the channel; It is the maximum value among all data points in the channel; It is the minimum value among all data points in the channel;

[0041] Represent the statistics of all channels as a set. μ , And M:

[0042] ;

[0043] ;

[0044] ;

[0045] Step S213, Sorting

[0046] Sort all channels in each group from largest to smallest based on the statistical value, resulting in the sorted channel set:

[0047] ;

[0048] In the above formula, It is based on statistics μ , The index after sorting by M;

[0049] Step S214: Select the suspected abnormal channel

[0050] Let p be the percentage set by the early warning system. Select the top p% of channels as suspected abnormal channels to obtain the set of suspected abnormal channels. :

[0051] ;

[0052] In the above formula, It is a suspected abnormal channel. =1,2,..., qp , qp Round to the nearest positive integer.

[0053] Step S215: Merge Channels

[0054] The suspected abnormal channels selected from all groups based on the calculated statistics are merged into a single list to obtain the merged channel set A.

[0055] ;

[0056] Step S216: Deduplication

[0057] After removing duplicate channel numbers from the merged list, we obtain the deduplicated channel set. :

[0058] ;

[0059] Step S217, Return Result

[0060] return This serves as a list containing all suspected abnormal channels.

[0061] Furthermore, the natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm has three working modes:

[0062] a. Real-time identification mode, i.e., a two-level identification process;

[0063] b. Add / Read Data Mode

[0064] Add data mode: While collecting real-time one-dimensional vibration data, add known leakage sample data, and replace the risk status of the corresponding one-dimensional vibration data collected in real time with an abnormal status.

[0065] Data reading mode: Real-time capture of pipeline leakage signal strength, i.e., the amplitude value of leakage fluctuation, and accurate separation of leakage and non-leakage data based on the start / end time of leakage;

[0066] c. Training Mode

[0067] The known leakage sample data were repeatedly used to transform the one-dimensional vibration data into two-dimensional SDP images for specific classification learning by employing a natural gas pipeline buried leakage identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm. This optimized the convolutional neural network (CNN) model, and the trained CNN model was automatically updated to a new identification model.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. Integrated Data Concentration and Visualization: The system has a simple and reasonable structure, requiring no additional hardware equipment, thus reducing construction and maintenance costs. Through the integration and optimization of software algorithms, functional concentrating is achieved, while improving data visualization. This allows operators to quickly and intuitively grasp the safety status of the pipeline, improving the system's usability and ease of operation, enhancing the intuitiveness and understandability of monitoring results, and making monitoring work more efficient and accurate.

[0070] 2. Improved Real-Time Performance and Accuracy: The system integrates advanced algorithms such as WebSocket communication technology, feature value filtering algorithm, natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern (SDP), and convolutional neural network (CNN) model. Through feature value filtering and real-time interactive conversion of SDP images, the CNN model is used for identification, and the results are displayed on the interface through visualization methods. This enables real-time monitoring and early warning of natural gas pipeline leaks, improves the real-time performance and accuracy of leak monitoring, and allows for faster response to potential leak risks.

[0071] 3. High level of intelligence: The training mode of the convolutional neural network (CNN) model enables the system to continuously learn and optimize, improve the accuracy of leakage identification, and automatically update the model to adapt to new data features, thereby enhancing the intelligence level of the system. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the natural gas pipeline leakage risk monitoring and early warning system based on DVS according to the present invention.

[0073] Figure 2 This is a flowchart of the natural gas pipeline leakage risk monitoring and early warning method based on DVS according to the present invention.

[0074] Figure 3 This is a flowchart illustrating the feature value screening and buried natural gas pipeline leak identification process of the present invention.

[0075] Figure 4 This is a schematic diagram of the structure of the Convolutional Neural Network (CNN) model of the present invention;

[0076] Figure 5 The SDP image generated in the embodiments of the present invention;

[0077] Figure 6 These are the time-domain and frequency-domain diagrams in the embodiments of the present invention;

[0078] Figure 7 This is a schematic diagram of latitude and longitude early warning positioning in an embodiment of the present invention.

[0079] In the diagram: 1. Natural gas pipeline; 2. Leakage hole; 3. Communication fiber optic bundle tube; 4. Communication fiber optic cable; 5. Host computer; 6. DVS equipment; 7. Gas gathering station. Detailed Implementation

[0080] To facilitate understanding and implementation of the present invention by those skilled in the art, the various steps of the method proposed in this invention are described in detail below. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various modifications or alterations to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0081] Example

[0082] like Figure 1 As shown, this invention discloses a natural gas pipeline leakage risk monitoring and early warning system based on DVS. The monitoring and early warning system consists of two parts: hardware and software. The hardware includes a host computer, a distributed fiber optic vibration sensor (DVS) device, and a sensing optical cable. The DVS device establishes a real-time connection with the host computer via Websocket communication. The DVS device is also connected to the sensing optical cable. The sensing optical cable is laid in the same trench as the natural gas pipeline. The sensing optical cable is used to collect temperature, strain, and vibration information at any point along the natural gas pipeline in real time. The DVS device converts the real-time collected vibration information into one-dimensional vibration data and sends it to the host computer. The host computer parses the received one-dimensional vibration data through software, identifies, judges, and visualizes whether the natural gas pipeline status is abnormal.

[0083] The software includes a distributed fiber optic vibration sensing system, a natural gas pipeline leak identification system, and a natural gas pipeline leak risk monitoring and early warning system. The distributed fiber optic vibration sensing system includes a data receiving module and a data output module. The data receiving module receives vibration information collected in real time by the sensing fiber optic cable and converts it into one-dimensional vibration data, which is then sent to the data output module. The data output module converts the one-dimensional vibration data into binary form and transmits it to the host computer's natural gas pipeline leak risk monitoring and early warning system. The natural gas pipeline leak identification system includes a data analysis module, a data conversion module, and an identification module. The data analysis module has a built-in feature value filtering algorithm to filter out outliers in the one-dimensional vibration data. The data conversion module converts the filtered one-dimensional vibration data into a two-dimensional SDP image. The identification module includes a convolutional neural network. The system includes a CNN model and a model training unit. The CNN model is used to identify, judge, and output suspected anomalies in natural gas pipelines from the converted two-dimensional SDP images. The model training unit is used to train the CNN model according to a preset training mode and automatically update the trained CNN model to a new CNN model. The natural gas pipeline leakage risk monitoring and early warning system includes a visualization module, a display module, and a data storage module. The visualization module is used to convert the output suspected anomalies in natural gas pipelines into specific leakage data, leakage early warning SDP images, latitude and longitude information, and time-domain maps. The display module is used to display the various information converted by the visualization module. The data storage module is used to classify and store leakage data and non-leakage data.

[0084] Specifically, the network communication between the host computer and the distributed fiber optic vibration sensing (DVS) device is based on the TCP / IP protocol and uses a client / server (C / S) structure connection. The data upload of the distributed fiber optic vibration sensing (DVS) device adopts an automatic transmission method. That is, after the host computer issues a data acquisition command, the distributed fiber optic vibration sensing (DVS) device automatically sends the temperature, strain, and vibration information of any point along the natural gas pipeline collected by the sensing fiber optic cable to the connected host computer according to the sampling interval.

[0085] Specifically, the natural gas pipeline leakage risk monitoring and early warning system adopts a front-end and back-end separation architecture and a three-tier web application architecture based on the B / S model. The front-end includes a client, i.e., a display module, and the back-end includes a server, i.e., a visualization module and a data storage module. The visualization module uses Vue 2.0 and axios to realize real-time data display and visualization of pipeline early warning status. The display module can display real-time leakage data, leakage early warning SDP images, latitude and longitude information, and time domain graphs.

[0086] Based on the aforementioned natural gas pipeline leakage risk monitoring and early warning system, such as Figure 2As shown in the figure, this embodiment also provides an early warning method for the natural gas pipeline leakage risk monitoring system, including the following steps:

[0087] Step S1: The distributed fiber optic vibration sensor (DVS) automatically transmits the temperature, strain, and vibration information collected by the sensing fiber optic cable at any point along the natural gas pipeline to the host computer at sampling intervals via Websocket communication.

[0088] Step S2: The host computer analyzes and performs secondary identification on the received vibration information, and outputs the final suspected abnormal result.

[0089] Specifically, such as Figure 3 As shown, the secondary identification includes two parts: feature value screening and buried natural gas pipeline leak identification. These are implemented based on a feature value screening algorithm and a buried natural gas pipeline leak identification algorithm based on wavelet transform and the symmetric point pattern SDP algorithm, respectively. The steps are as follows:

[0090] Step S21: Narrow down the scope of suspected abnormal one-dimensional vibration data through feature value screening algorithm to reduce the amount of data identified by convolutional neural network (CNN) model;

[0091] Step S22: Using a natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm, one-dimensional vibration data is converted into two-dimensional SDP image;

[0092] The wavelet transform reduces noise by suppressing useless signals and enhancing useful signals, while preserving the local peak characteristics of useful signals. To ensure the adaptive properties of the wavelet function, the wavelet basis function is determined based on the wavelet entropy selection method. The wavelet entropy is calculated by quantizing the energy distribution of the wavelet coefficients at each decomposition level.

[0093] The Symmetric Point Mode SDP algorithm can convert the one-dimensional time-domain waveform of a vibration signal into a polar coordinate snowflake image. The conversion formula is as follows:

[0094] ;

[0095] In the above formula, Yes transpose; The first in the time domain signal i The range of one point; and These represent the maximum and minimum values ​​of the time-domain signal, respectively; τ is the time interval factor, 1≤τ≤10; θ is the mirror symmetry angle, θ=60°; ζ represents the gain of the plotting angle, ζ≤θ; It is the first i The polar coordinate radius of each point; and These are the clockwise and counterclockwise rotation angles corresponding to the polar axis, respectively.

[0096] Step S23: Based on the convolutional neural network (CNN) model, identify the transformed two-dimensional SDP image to determine whether the risk status of the natural gas pipeline is normal or abnormal.

[0097] like Figure 4 As shown, the Convolutional Neural Network (CNN) model comprises four convolutional layers, four pooling layers, two fully connected layers, and a softmax activation function. The specific formulas for the convolutional and pooling layers are as follows:

[0098] ;

[0099] ;

[0100] In the above formula, It is the j-th element of the l-th layer; It is the feature map of the (l-1)th layer of the j-th convolutional region; It is the weight matrix of the l-th layer; It is a bias; f (*) is a non-linear activation function; This represents the j-th feature map of the l-th layer with weights; down (*) is the downsampling function;

[0101] After multiple convolution and pooling operations, the fully connected layer transforms the multidimensional feature representation into a one-dimensional vector; the activation function of the last layer of the fully connected layer is softmax, and the mathematical expression of the softmax activation function is as follows:

[0102] ;

[0103] In the above formula, is the value of the nth category output by the fully connected layer, and K is the total number of categories in the fully connected layer; It is the predicted probability of the nth category;

[0104] Step S24: Display the identification results and related information based on the visualization module.

[0105] Specifically, the calculation steps of the feature value filtering algorithm described in step S21 are as follows:

[0106] Step S211, Grouping

[0107] Let the set of all channels be . Where m is the total number of channels; C is divided into k groups, each containing q channels, i.e. Where h = 1, 2, ..., k;

[0108] Step S212: Calculate the statistics for each channel.

[0109] For each channel Calculate its average value ,variance and peak :

[0110] ;

[0111] ;

[0112] ;

[0113] In the above formula, It is the first in the channel b One data point; a It is the total number of data points in the channel; It is the maximum value among all data points in the channel; It is the minimum value among all data points in the channel;

[0114] Represent the statistics of all channels as a set. μ , And M:

[0115] ;

[0116] ;

[0117] ;

[0118] Step S213, Sorting

[0119] Sort all channels in each group from largest to smallest based on the statistical value, resulting in the sorted channel set:

[0120] ;

[0121] In the above formula, It is based on statistics μ , And the index after sorting by M;

[0122] Step S214: Select the suspected abnormal channel

[0123] Let p be the percentage set by the early warning system. Select the top p% of channels as suspected abnormal channels to obtain the set of suspected abnormal channels. :

[0124] ;

[0125] In the above formula, It is a suspected abnormal channel. =1,2,..., qp , qp Round to the nearest positive integer.

[0126] Step S215: Merge Channels

[0127] The suspected abnormal channels selected from all groups based on the calculated statistics are merged into a single list to obtain the merged channel set A.

[0128] ;

[0129] Step S216: Deduplication

[0130] After removing duplicate channel numbers from the merged list, we obtain the deduplicated channel set. :

[0131] ;

[0132] Step S217, Return Result

[0133] return This serves as a list containing all suspected abnormal channels.

[0134] Furthermore, the natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm has three working modes:

[0135] a. Real-time identification mode, i.e., a two-level identification process;

[0136] b. Add / Read Data Mode

[0137] Add data mode: While collecting real-time one-dimensional vibration data, add known leakage sample data, and replace the risk status of the corresponding one-dimensional vibration data collected in real time with an abnormal status.

[0138] Data reading mode: Real-time capture of pipeline leakage signal strength, i.e., the amplitude value of leakage fluctuation, and accurate separation of leakage and non-leakage data based on the start / end time of leakage;

[0139] c. Training Mode

[0140] The known leakage sample data were repeatedly used to transform the one-dimensional vibration data into two-dimensional SDP images for specific classification learning by employing a natural gas pipeline buried leakage identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm. This optimized the convolutional neural network (CNN) model, and the trained CNN model was automatically updated to a new identification model.

[0141] Working Process: The distributed fiber optic vibration sensing (DVS) device and host computer are activated. The DVS device automatically transmits real-time temperature, strain, and vibration information collected at any point along the natural gas pipeline to the host computer. The host computer's natural gas pipeline leakage risk monitoring and early warning system analyzes the DVS data and then performs a two-stage identification algorithm: first, a feature value filtering algorithm is used to reduce the amount of data recognized by the convolutional neural network (CNN) model; then, the symmetric point pattern (SDP) algorithm is used to convert the one-dimensional vibration data into a two-dimensional SDP image. The converted two-dimensional SDP image is shown below. Figure 5 As shown, a convolutional neural network (CNN) model is used to identify two-dimensional SDP images to determine whether the pipeline status is abnormal. Finally, the identification results and information are visualized on the interface of the natural gas pipeline leakage risk monitoring and early warning system. Figure 6 and Figure 7 The images show the time-domain and frequency-domain graphs, as well as the latitude and longitude coordinates of the leak location, displayed on the interface of the natural gas pipeline leak risk monitoring and early warning system. This invention uses a visualization method to intuitively display the monitoring results and useful information from the identification process on the interface, enabling operators to quickly grasp the safety status of the pipeline. By optimizing and updating the convolutional neural network (CNN) model through training modes, the system can continuously learn and adapt to new data features, effectively prevent and respond to leak events in a timely manner, and ensure the safe operation of the pipeline.

[0142] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A natural gas pipeline leakage risk monitoring and early warning system based on DVS, characterized in that, The monitoring and early warning system consists of two parts: hardware and software. The hardware includes a host computer, a distributed fiber optic vibration sensor (DVS) device, and a sensing optical cable. The DVS device establishes a real-time connection with the host computer via WebSocket communication. The DVS device is also connected to the sensing optical cable, which is laid in the same trench as the natural gas pipeline. The sensing optical cable is used to collect temperature, strain, and vibration information at any point along the natural gas pipeline in real time. The DVS device converts the real-time collected vibration information into one-dimensional vibration data and sends it to the host computer. The host computer parses the received one-dimensional vibration data using software, identifies, judges, and visualizes whether the natural gas pipeline is in an abnormal state. The software includes a distributed fiber optic vibration sensing system, a natural gas pipeline leak identification system, and a natural gas pipeline leak risk monitoring and early warning system. The distributed fiber optic vibration sensing system includes a data receiving module and a data output module. The data receiving module receives vibration information collected in real time by the sensing fiber optic cable and converts it into one-dimensional vibration data, which is then sent to the data output module. The data output module converts the one-dimensional vibration data into binary form and transmits it to the host computer's natural gas pipeline leak risk monitoring and early warning system. The natural gas pipeline leak identification system includes a data analysis module, a data conversion module, and an identification module. The data analysis module has a built-in feature value filtering algorithm to filter out outliers in the one-dimensional vibration data. The data conversion module converts the filtered one-dimensional vibration data into a two-dimensional SDP image. The identification module includes a convolutional neural network. The system includes a CNN model and a model training unit. The CNN model is used to identify, judge, and output suspected anomalies in natural gas pipelines from the converted two-dimensional SDP images. The model training unit is used to train the CNN model according to a preset training mode and automatically update the trained CNN model to a new CNN model. The natural gas pipeline leakage risk monitoring and early warning system includes a visualization module, a display module, and a data storage module. The visualization module is used to convert the output suspected anomalies in natural gas pipelines into specific leakage data, leakage early warning SDP images, latitude and longitude information, and time-domain maps. The display module is used to display the various information converted by the visualization module. The data storage module is used to classify and store leakage data and non-leakage data.

2. The natural gas pipeline leakage risk monitoring and early warning system based on DVS according to claim 1, characterized in that, The network communication between the host computer and the distributed fiber optic vibration sensing (DVS) device is based on the TCP / IP protocol and uses a client / server (C / S) structure. The data upload of the distributed fiber optic vibration sensing (DVS) device adopts an automatic transmission method. That is, after the host computer issues a data acquisition command, the distributed fiber optic vibration sensing (DVS) device automatically sends the temperature, strain, and vibration information of any point along the natural gas pipeline collected by the sensing fiber optic cable to the connected host computer according to the sampling interval.

3. The natural gas pipeline leakage risk monitoring and early warning system based on DVS according to claim 1, characterized in that, The natural gas pipeline leakage risk monitoring and early warning system adopts a front-end and back-end separation architecture and a three-tier web application architecture based on the B / S model. The front-end includes a client, i.e., a display module, and the back-end includes a server, i.e., a visualization module and a data storage module. The visualization module uses Vue 2.0 and axios to realize real-time data display and visualization of pipeline early warning status. The display module can display real-time leakage data, leakage early warning SDP images, latitude and longitude information, and time domain graphs.

4. A method for monitoring and early warning of natural gas pipeline leakage risks based on DVS, wherein the method is based on the natural gas pipeline leakage risk monitoring and early warning system as described in any one of claims 1-3, characterized in that, Includes the following steps: Step S1: The distributed fiber optic vibration sensor (DVS) automatically transmits the temperature, strain, and vibration information collected by the sensing fiber optic cable at any point along the natural gas pipeline to the host computer at sampling intervals via Websocket communication. Step S2: The host computer analyzes and performs secondary identification on the received vibration information, and outputs the final suspected abnormal result.

5. The method for monitoring and early warning of natural gas pipeline leakage risks based on DVS according to claim 4, characterized in that, The secondary identification comprises two parts: feature value screening and buried natural gas pipeline leak identification. These are implemented based on a feature value screening algorithm and a buried natural gas pipeline leak identification algorithm based on wavelet transform and the symmetric point pattern SDP algorithm, respectively. The steps are as follows: Step S21: Narrow down the scope of suspected abnormal one-dimensional vibration data through feature value screening algorithm to reduce the amount of data identified by convolutional neural network (CNN) model; Step S22: Using a natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm, one-dimensional vibration data is converted into two-dimensional SDP image; The wavelet transform reduces noise by suppressing useless signals and enhancing useful signals, while preserving the local peak characteristics of useful signals. To ensure the adaptive properties of the wavelet function, the wavelet basis function is determined based on the wavelet entropy selection method. The wavelet entropy is calculated by quantizing the energy distribution of the wavelet coefficients at each decomposition level. The symmetric point mode SDP algorithm can convert one-dimensional vibration data into a two-dimensional SDP image. The conversion formula is as follows: ； In the above formula, Yes transpose; The first in the time domain signal i The range of one point; and These represent the maximum and minimum values ​​of the time-domain signal, respectively; τ is the time interval factor, 1≤τ≤10; θ is the mirror symmetry angle, θ=60°; ζ represents the gain of the plotting angle, ζ≤θ; It is the first i The polar coordinate radius of each point; and These are the clockwise and counterclockwise rotation angles corresponding to the polar axis, respectively. Step S23: Based on the convolutional neural network (CNN) model, identify the transformed two-dimensional SDP image to determine whether the risk status of the natural gas pipeline is normal or abnormal. The Convolutional Neural Network (CNN) model comprises four convolutional layers, four pooling layers, two fully connected layers, and a softmax activation function. The specific formulas for the convolutional and pooling layers are as follows: ； ； In the above formula, It is the first layer of the l-th layer. j One element; It is the first j Feature maps of the (l-1)th layer of each convolutional region; It is the weight matrix of the l-th layer; It is a bias; f (*) is a non-linear activation function; The weight of the l-th layer is represented by the first... j Each feature map; down (*) is the downsampling function; After multiple convolution and pooling operations, the fully connected layer transforms the multidimensional feature representation into a one-dimensional vector; the activation function of the last layer of the fully connected layer is softmax, and the mathematical expression of the softmax activation function is as follows: ； In the above formula, is the value of the nth category output by the fully connected layer, and K is the total number of categories in the fully connected layer; It is the predicted probability of the nth category; Step S24: Display the identification results and related information based on the visualization module.

6. The method for monitoring and early warning of natural gas pipeline leakage risks based on DVS according to claim 5, characterized in that, The calculation steps of the feature value filtering algorithm described in step S21 are as follows: Step S211, Grouping Let the set of all channels be . Where m is the total number of channels; C is divided into k groups, each containing q channels, i.e. Where h = 1, 2, ..., k; Step S212: Calculate the statistics for each channel. For each channel Calculate its average value ,variance and peak : ； ； ； In the above formula, It is the first in the channel b One data point; a It is the total number of data points in the channel; It is the maximum value among all data points in the channel; It is the minimum value among all data points in the channel; Represent the statistics of all channels as a set. μ , And M: ； ； ； Step S213, Sorting Sort all channels in each group from largest to smallest based on the statistical value, resulting in the sorted channel set: ； In the above formula, It is based on statistics μ , The index after sorting by M; Step S214: Select the suspected abnormal channel Let p be the percentage set by the early warning system. Select the top p% of channels as suspected abnormal channels to obtain the set of suspected abnormal channels. : ； In the above formula, It is a suspected abnormal channel. =1,2,..., qp , qp Round to the nearest positive integer. Step S215: Merge Channels The suspected abnormal channels selected from all groups based on the calculated statistics are merged into a single list to obtain the merged channel set A. ； Step S216: Deduplication After removing duplicate channel numbers from the merged list, we obtain the deduplicated channel set. : ； Step S217, Return Result return This serves as a list containing all suspected abnormal channels.

7. The method for monitoring and early warning of natural gas pipeline leakage risks based on DVS according to claim 5, characterized in that, The natural gas pipeline buried leak identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm has three working modes: a. Real-time identification mode, i.e., a two-level identification process; b. Add / Read Data Mode Add data mode: While collecting real-time one-dimensional vibration data, add known leakage sample data, and replace the risk status of the corresponding one-dimensional vibration data collected in real time with an abnormal status. Data reading mode: Real-time capture of pipeline leakage signal strength, i.e., the amplitude value of leakage fluctuation, and accurate separation of leakage and non-leakage data based on the start / end time of leakage; c. Training Mode The known leakage sample data were repeatedly used to transform the one-dimensional vibration data into two-dimensional SDP images for specific classification learning by employing a natural gas pipeline buried leakage identification algorithm based on wavelet transform and symmetric point pattern SDP algorithm. This optimized the convolutional neural network (CNN) model, and the trained CNN model was automatically updated to a new identification model.

Images