Method and device for determining pipeline leakage risk and electronic equipment

Abstract

The invention discloses a method and a device for determining a pipeline leakage risk and electronic equipment, relates to the technical field of pipeline detection, and aims to solve the problem of how to accurately evaluate the leakage risk of an oil and gas pipeline. The method comprises the steps that operation maintenance data and historical leakage data of a target pipeline at the current time are acquired, and the operation data comprise operation parameters, inspection data, environment data and leakage influence information of the target pipeline; determining a feature data set based on the operation parameters and the inspection data of the target pipeline; processing the feature data set, the historical leakage data and the environmental data of the target pipeline through a risk assessment model, and determining a leakage probability value of the target pipeline at the current time; and based on the leakage probability value and the leakage influence information, determining the comprehensive leakage risk of the target pipeline at the current time. Therefore, the leakage risk of the oil and gas pipeline can be accurately evaluated so as to ensure stable operation of the oil and gas pipeline.

Description

Technical Field

[0001] This application relates to the field of pipeline inspection technology, and in particular to a method, apparatus and electronic equipment for determining pipeline leakage risk. Background Technology

[0002] As a crucial infrastructure for energy transportation, the assessment of damage risks to oil and gas pipeline networks is essential for their stable operation.

[0003] In existing technologies, risk assessments of oil and gas pipelines often rely on single data points, such as data from sensors deployed within the pipeline network, to determine the risk of leaks. However, this method is not very accurate and suffers from problems such as prediction lag and misjudgment.

[0004] Therefore, how to accurately assess the leakage risk of oil and gas pipelines in order to ensure their stable operation has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, and electronic device for determining pipeline leakage risk, aiming to solve the problem of how to accurately assess the leakage risk of a target pipeline and ensure the stable operation of the target pipeline.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, this application provides a method for determining pipeline leakage risk. The method includes: acquiring current operation and maintenance data and historical leakage data of a target pipeline, wherein the operation data includes the target pipeline's operating parameters, inspection data, environmental data, and leakage impact information, and the leakage impact information is used to indicate the impact caused by leakage in the target pipeline; determining a feature dataset based on the target pipeline's operating parameters and inspection data; processing the feature dataset, historical leakage data, and environmental data of the target pipeline using a risk assessment model to determine the leakage probability value of the target pipeline at the current time, wherein the risk assessment model is used to determine the probability of leakage in the target pipeline; and determining the comprehensive leakage risk of the target pipeline at the current time based on the leakage probability value and the leakage impact information.

[0008] In one possible approach, the operating parameters of the target pipeline include cathode potential data and pipeline internal pressure data. Based on the operating parameters and inspection data of the target pipeline, a feature dataset is determined, including: performing mode decomposition on the cathode potential data using a preset empirical mode decomposition algorithm to determine energy proportion feature data; processing the pipeline internal pressure data using a preset information entropy construction algorithm to determine energy entropy feature data, which is used to indicate the frequency distribution complexity of the internal pressure fluctuations in the target pipeline; and processing the inspection data using a preset spatiotemporal feature extraction algorithm to determine inspection spatiotemporal feature data, which is used to indicate the quality of the inspection of the target pipeline. The feature dataset includes energy proportion feature data, energy entropy feature data, and inspection spatiotemporal feature data.

[0009] In one possible approach, a risk assessment model is used to process the feature dataset, historical leakage data, and environmental data of the target pipeline to determine the leakage probability value of the target pipeline at the current time. This includes: processing the feature dataset, historical leakage data, and environmental data to determine a first fused feature parameter. The first fused feature parameter has multiple feature parameter dimensions, and each feature parameter dimension corresponds to a feature parameter type. Each data in the feature dataset, the historical leakage data, and the environmental data each correspond to a parameter type. The first fused feature parameter is then input into the risk assessment model to determine the leakage probability value.

[0010] In one possible approach, the risk assessment model includes an input layer, a hidden layer, and an output layer. The first fusion feature parameter is input into the risk assessment model to determine the leakage probability value, including: obtaining the first fusion feature parameter through the input layer;

[0011] The weights corresponding to multiple feature parameter types in the first fusion feature parameter are adjusted by the preset attention mechanism of the hidden layer to determine the second fusion feature parameter; the second fusion feature parameter is interval mapped by the hidden layer to determine the weighted feature parameter; the weighted feature parameter is normalized by the output layer to determine the leakage probability value of the target pipeline.

[0012] One possible approach involves determining the overall leakage risk of a target pipeline at the current time based on leakage probability values ​​and leakage impact information. This includes: constructing a risk assessment matrix based on the leakage probability values ​​and leakage impact information; determining a matrix consistency ratio based on the risk assessment matrix; if the matrix consistency ratio is less than a preset consistency threshold, determining the weights of the leakage probability values ​​and leakage impact information based on the risk assessment matrix; and determining the overall risk value of the target pipeline at the current time based on the leakage probability values, their weights, and the leakage impact information. This overall risk value indicates the magnitude of the risk of leakage in the target pipeline. The overall leakage risk of the target pipeline at the current time is determined based on this overall risk value.

[0013] In one possible approach, the method further includes: when the matrix consistency index is greater than or equal to a preset consistency threshold, acquiring historical risk data of the target pipeline, where the historical risk data is used to indicate the factors that caused the target pipeline to leak in a historical time; and based on the historical risk data, correcting the risk judgment matrix until the matrix consistency index of the corrected risk judgment matrix is ​​less than the preset consistency threshold.

[0014] In one possible approach, the risk assessment model is determined as follows: The incremental dataset of the target pipeline and the collection time corresponding to each data point in the incremental dataset are obtained. The incremental dataset consists of operational data of the target pipeline added in the previous preset time period before the current time. Based on the incremental dataset and the collection time, weights are assigned to the data in the incremental dataset to determine the updated dataset. The parameters of the risk assessment model are updated based on the updated dataset to determine the risk assessment model.

[0015] In one possible approach, the leakage impact information includes an impact range factor and a response time factor. Obtaining the leakage impact factor includes: obtaining the number of target objects within a preset range of the target pipeline and the leakage response time. The target objects are those affected by the leakage in the target pipeline, and the leakage response time includes the leakage alarm time and the alarm response time. Based on the number of sensitive targets and environmental data, the impact range factor is determined. Based on the leakage alarm time and the alarm response time, the response time factor is determined.

[0016] Secondly, this application provides a device for determining pipeline leakage risk. The device includes: an acquisition unit for acquiring current operation and maintenance data and historical leakage data of a target pipeline, wherein the operation data includes the target pipeline's operating parameters, inspection data, environmental data, and leakage impact information, and the leakage impact information is used to indicate the impact caused by leakage in the target pipeline; a processing unit for determining a feature dataset based on the target pipeline's operating parameters and inspection data; a processing unit for processing the feature dataset, historical leakage data, and environmental data of the target pipeline using a risk assessment model to determine the leakage probability value of the target pipeline at the current time, wherein the risk assessment model is used to determine the probability of leakage in the target pipeline; and a processing unit for determining the comprehensive leakage risk of the target pipeline at the current time based on the leakage probability value and the leakage impact information.

[0017] In one possible approach, the operating parameters of the target pipeline include cathode potential data and pipeline internal pressure data. Based on the operating parameters and inspection data of the target pipeline, a feature dataset is determined, including: performing mode decomposition on the cathode potential data using a preset empirical mode decomposition algorithm to determine energy proportion feature data; processing the pipeline internal pressure data using a preset information entropy construction algorithm to determine energy entropy feature data, which is used to indicate the frequency distribution complexity of the internal pressure fluctuations in the target pipeline; and processing the inspection data using a preset spatiotemporal feature extraction algorithm to determine inspection spatiotemporal feature data, which is used to indicate the quality of the inspection of the target pipeline. The feature dataset includes energy proportion feature data, energy entropy feature data, and inspection spatiotemporal feature data.

[0018] In one possible approach, the processing unit is specifically used to process the feature dataset, historical leakage data, and environmental data to determine a first fused feature parameter. The first fused feature parameter has multiple feature parameter dimensions, and each feature parameter dimension corresponds to a feature parameter type. Each data in the feature dataset, the historical leakage data, and the environmental data each correspond to a parameter type. The fused feature parameter is then input into a risk assessment model to determine the leakage probability value.

[0019] In one possible approach, the risk assessment model includes an input layer, a hidden layer, an output layer, and a processing unit. Specifically, the processing unit obtains first fusion feature parameters through the input layer; adjusts the weights corresponding to multiple feature parameter types in the first fusion feature parameters through a preset attention mechanism in the hidden layer to determine second fusion feature parameters; performs interval mapping on the second fusion feature parameters through the hidden layer to determine weighted feature parameters; and normalizes the weighted feature parameters through the output layer to determine the leakage probability value of the target pipeline.

[0020] In one possible approach, the processing unit is specifically configured to: construct a risk assessment matrix based on leakage probability values ​​and leakage impact information; determine a matrix consistency ratio based on the risk assessment matrix; if the matrix consistency ratio is less than a preset consistency threshold, determine the weights of the leakage probability values ​​and leakage impact information based on the risk assessment matrix; and determine the comprehensive risk value of the target pipeline at the current time based on the leakage probability values, their weights, and the leakage impact information, whereby the comprehensive risk value indicates the magnitude of the risk of leakage in the target pipeline. Based on the comprehensive risk value, the comprehensive leakage risk of the target pipeline at the current time is determined.

[0021] In one possible approach, the processing unit, specifically used to obtain historical risk data of the target pipeline when the matrix consistency index is greater than or equal to a preset consistency threshold, the historical risk data being used to indicate the factors that caused the target pipeline to leak in a historical time; based on the historical risk data, the risk judgment matrix is ​​corrected until the matrix consistency index of the corrected risk judgment matrix is ​​less than the preset consistency threshold.

[0022] In one possible approach, the processing unit is specifically used to acquire the incremental dataset of the target pipeline and the collection time corresponding to each data point in the incremental dataset. The incremental dataset consists of operational data of the target pipeline added in the previous preset time period before the current time. Based on the incremental dataset and the collection time, the data in the incremental dataset is weighted and an updated dataset is determined. Based on the updated dataset, the parameters of the risk assessment model are updated to determine the risk assessment model.

[0023] In one possible approach, the leakage impact information includes an impact range factor and a response time factor. The acquisition unit is specifically used to acquire the number of target objects within a preset range of the target pipeline and the leakage response time. The target objects are objects affected by the leakage of the target pipeline, and the leakage response time includes the leakage alarm time and the alarm response time. Based on the number of sensitive targets and environmental data, the impact range factor is determined; based on the leakage alarm time and the alarm response time, the response time factor is determined.

[0024] Thirdly, this application provides an electronic device including a memory and a processor; the memory and the processor are coupled; the memory is used to store instructions executable by the processor; when the processor executes the instructions, it performs the methods described in the first aspect and any possible implementation thereof.

[0025] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0026] Fifthly, this application provides a chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run computer programs or instructions to implement the methods described in the first aspect and any possible implementation thereof.

[0027] Sixthly, this application provides a computer program product containing instructions that, when executed by a computer, cause the computer to perform the methods described in the first aspect and any possible implementation thereof.

[0028] The technical problems that the pipeline leakage risk determination device, electronic equipment, computer storage medium, chip or computer program product can solve and the technical effects that can be achieved in the above solution can be found in the technical problems and technical effects solved in the first aspect above, and will not be repeated here. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A flowchart illustrating a method for determining pipeline leakage risk provided in an embodiment of this application;

[0031] Figure 2 A flowchart illustrating yet another method for determining pipeline leakage risk provided in this application embodiment;

[0032] Figure 3 A flowchart illustrating yet another method for determining pipeline leakage risk provided in this application embodiment;

[0033] Figure 4 This is a structural diagram of a device for determining pipeline leakage risk provided in an embodiment of this application. Detailed Implementation

[0034] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0035] In the description of this application, it should be understood that the terms "upper," "lower," "left," "right," "front," "rear," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or relative positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and for simplification, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Unless otherwise specified, the above-mentioned orientational descriptions can be flexibly set in practical applications, provided that the relative positional relationship shown in the accompanying drawings is satisfied.

[0036] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0037] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "communication" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection. They can refer to a direct connection or an indirect connection through an intermediate medium, or a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0038] In embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, article, or apparatus that includes that element.

[0039] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0040] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0041] In this embodiment, we provide a system for determining pipeline leakage risk. The system includes a data processing unit, a basic support unit, and an operation management unit.

[0042] 1-1. Data Processing Unit.

[0043] The data processing unit includes a data access module, a parsing module, a storage module, and a distribution module.

[0044] The data access module uses message queues to achieve asynchronous data reception. The parsing module handles multi-protocol conversion, including Modbus RTU, Modbus TCP, and a custom BeiDou short message protocol. The storage module is built based on a time-series database for spatiotemporal indexing and querying. The distribution module uses an in-memory database to implement data caching and publish-subscribe functionality.

[0045] 1-2. Basic support unit.

[0046] The basic support units include a GIS service module and a spatiotemporal reference module.

[0047] The GIS service module is built on a spatial database and is used for spatial indexing and topology analysis. The spatiotemporal reference module uses Coordinated Universal Time (UTC) provided by the BeiDou Navigation Satellite System as its time reference.

[0048] 1-3. Operations Management Unit.

[0049] The operation management unit includes an equipment monitoring module, a log auditing module, and an alarm module.

[0050] The equipment monitoring module is based on a monitoring system to achieve real-time status monitoring. The log auditing module is built on a log management system and is used for log storage, retrieval, and analysis. The alarm module is used to issue alarms when a leak occurs in the target pipeline.

[0051] It should be noted that the Beidou terminal in this embodiment can be the BD-810 single-mode terminal (Beidou), which supports short message communication frequency ≥ 1 time / minute, communication interface is RS485 / RS232, built-in lithium battery for extended battery life, protection level IP68, and is suitable for field pipeline monitoring scenarios; at the same time, the Beidou-3 command machine can be the A-type command machine, which supports a large number of access terminals and is equipped with a dual-machine hot standby power module to ensure the availability of the communication link.

[0052] In some embodiments, such as Figure 1 As shown, this application provides a method for determining pipeline leakage risk, specifically including: S101-S104.

[0053] S101. Obtain the current operation and maintenance data and historical leakage data of the target pipeline.

[0054] The operation and maintenance data includes the target pipeline's operating parameters, inspection data, environmental data, and leakage impact information. Leakage impact information indicates the effects of a leak in the target pipeline, while inspection data indicates the inspection route and locations. Environmental data indicates soil moisture content and soil corrosivity in the area where the target pipeline is located.

[0055] In one possible implementation, a satellite-to-ground communication link is established through the BeiDou-3 command system. A combination mechanism of packet transmission protocol, Automatic Repeat Request (ARQ), and Forward Error Correction (FEC) coding is used to obtain the initial operation and maintenance data and historical leakage data of the target pipeline at the current time.

[0056] In some embodiments, the initial operating parameters are preprocessed to determine the operating parameters of the target pipeline at the current time.

[0057] In one possible implementation, the initial operation and maintenance data are preprocessed using wavelet transform denoising algorithm, dynamic time warping outlier detection algorithm, and BeiDou spatiotemporal reference normalization algorithm to determine the operation and maintenance data of the target pipeline at the current time.

[0058] The initial operating parameters include initial cathodic protection potential data and initial pipeline internal pressure data.

[0059] For example, taking the preprocessing of initial cathodic protection potential data as an example, the initial cathodic protection potential data is decomposed into 5 levels for denoising using the db4 wavelet. Utilizing the multi-resolution analysis characteristics of wavelet transform, the data is decomposed into different frequency components, where the low-frequency part retains the main signal features, and the high-frequency part corresponds to noise components. The denoising process is implemented through a threshold shrinkage algorithm: threshold processing is applied to the high-frequency coefficients of each level, retaining coefficients greater than the threshold to preserve effective signal features, and suppressing noise components less than the threshold. The threshold is dynamically calculated based on data statistical characteristics to ensure that the abrupt changes in the potential signal (such as corrosion-related abnormal potential fluctuations) are preserved to the greatest extent possible while removing noise. This processing can effectively filter out background noise such as power frequency interference and grounding noise in the cathodic protection potential data, improving the data signal-to-noise ratio.

[0060] In another example, taking the preprocessing of initial pipeline internal pressure data as an example, a standard operating condition template is used as a benchmark. The dynamic time warping algorithm is employed to calculate the similarity distance between real-time data and the template, thereby achieving outlier detection. The standard operating condition template is constructed based on historical data from normal pipeline operation, reflecting the normal fluctuation pattern of internal pressure at different times. When the alignment distance between real-time data and the template exceeds 1.5 times the standard deviation, the corresponding data point is determined to be an anomaly. Simultaneously, anomaly repair employs a cubic spline interpolation algorithm. By constructing a piecewise smooth polynomial function, a continuous curve is fitted based on the effective data points before and after the anomaly point. This ensures that the repaired data not only conforms to the changing trend of pipeline internal pressure but also maintains the smoothness of the data sequence, avoiding the discontinuity problem caused by directly deleting anomaly points.

[0061] It should be noted that, regarding the time reference, this embodiment uses Coordinated Universal Time (UTC) provided by the BeiDou Navigation Satellite System as the unified time reference. The data from each monitoring point is synchronized using the pulse-per-second (1PPS) signal and time code output by the BeiDou terminal, eliminating clock deviations between different devices and ensuring the consistency of data timestamps. Regarding the spatial reference, the latitude and longitude coordinates obtained from BeiDou positioning are converted into UTM projected coordinates, and combined with the spatial database of the GIS service module, the spatial coordinates of all data are unified.

[0062] S102. Based on the operating parameters of the target pipeline and the inspection data, determine the feature dataset.

[0063] The operating parameters of the target pipeline include cathode location data and pipeline internal pressure data.

[0064] In one possible implementation, the cathode location data is decomposed using a pre-defined empirical mode decomposition algorithm to determine the energy proportion characteristic data.

[0065] For example, empirical mode decomposition (EMD) is performed on the cathode potential data to obtain several intrinsic mode functions (IMFs). The potential sequence in the cathode potential data is traversed, and all local maxima and minima are extracted. Cubic spline functions are used to fit the upper and lower envelopes, respectively. The standard deviation criterion is used to determine whether the intermediate signals satisfy the IMF conditions. This iteration is repeated until a valid IMF is generated. Then, the decomposition stops when the remaining trend term is less than 5% of the original signal amplitude or when an 8th-order IMF is obtained, to avoid over-decomposition.

[0066] The first three eigenmode functions are selected from the eigenmode functions and denoted as IMF1, IMF2, and IMF3.

[0067] Calculate the energy of each of the first three IMFs in the 1-10Hz frequency band, and the total energy of the first three IMFs in the 1-10Hz frequency band.

[0068] For example, the energy percentage E of each of the first three IMFs in the 1-10Hz frequency band. i It satisfies the following formula 1.

[0069]

[0070] Among them, IMF i (f) represents the frequency domain form of the i-th order eigenmode function, where f is the frequency, ∫1 10 |IMF j (f)| 2 Let i be the energy of the i-th order IMF in the 1-10Hz frequency band. This represents the total energy of the first three IMFs (j = 1, 2, 3) in the 1-10 Hz frequency band.

[0071] Furthermore, in this embodiment, the first three effective IMFs (IMF1-3) obtained from EMD decomposition are selected using cross-correlation coefficients. These three components correspond to high-frequency fluctuation characteristics in the cathodic protection potential signal (such as potential changes caused by corrosion electrochemical processes). After selection, a 1024-point Fast Fourier Transform is performed on each IMF, and a Hanning window weighting is used to reduce spectral leakage. Components in the 1-10Hz frequency band (corresponding to the frequency range of electrochemical fluctuations related to pipeline corrosion) are extracted in the frequency domain. The frequency band energy of a single-order IMF is calculated by accumulating the square of the spectral amplitude within this band and multiplying it by the frequency resolution. The total frequency band energy of the first three IMFs is then obtained. Finally, the energy proportion characteristic is obtained by the ratio of each order energy to the total energy.

[0072] In one possible implementation, the internal pressure data of the pipeline is processed by a preset information entropy construction algorithm to determine energy entropy feature data. The energy entropy feature is used to indicate the frequency distribution complexity of the internal pressure fluctuation of the target pipeline.

[0073] For example, for pipeline internal pressure data, a 3-layer db4 wavelet packet decomposition algorithm is used to divide the signal frequency domain into 8 sub-bands. During the decomposition process, threshold denoising is performed on the coefficients of each layer (the threshold is set to 0.5 times the average energy of the corresponding layer) to suppress environmental noise interference. By calculating the energy proportion of each sub-band, a wavelet packet energy entropy feature is constructed based on the principle of information entropy. This feature quantifies the frequency distribution complexity of internal pressure fluctuations. When a pipeline leaks, changes in fluid dynamics will cause abnormal energy proportions in specific frequency bands, thereby causing entropy fluctuations, which can serve as a key frequency domain feature for early warning of leaks.

[0074] In one possible implementation, the inspection data is processed by a preset spatiotemporal feature extraction algorithm to determine the spatiotemporal feature data of the inspection.

[0075] For example, the mean and standard deviation of the time intervals for the timestamp sequence of inspection trajectory points are calculated to determine the inspection time characteristic data. The inspection time characteristic data is used to identify inspection cycle anomalies (such as sudden extension or shortening of intervals) to ensure the temporal continuity of manual inspection data.

[0076] The Euclidean distance between adjacent trajectory points is calculated based on Universal Transverse Mercator (UTM) projected coordinates, and the mean distance and coefficient of variation are extracted. The Douglas-Peucker algorithm is used to compress the trajectory points, extracting the number of inflection points and curvature features as spatial feature data for inspection. This spatial feature data is used to determine whether there are missed sections or path anomalies in the inspection route. The fusion analysis of spatiotemporal features can quantitatively assess inspection quality, forming cross-validation with monitoring data and improving the reliability of risk assessment. The spatiotemporal feature data of the inspection is determined by matrix concatenation of the temporal and spatial feature data.

[0077] One possible implementation involves fusing energy proportion features, energy entropy features, and spatiotemporal inspection features into an initial feature dataset. Then, principal component analysis is used to reduce the dimensionality of the initial feature dataset, thus determining the final feature dataset.

[0078] S103. Process the feature dataset, historical leakage data, and environmental data of the target pipeline using a risk assessment model to determine the leakage probability value of the target pipeline at the current time.

[0079] The historical leakage data includes leakage events that occurred in the target pipeline over a historical period, along with the cause of each leakage event. The risk assessment model is used to determine the probability of leakage in the target pipeline at the current time.

[0080] In one possible implementation, the feature dataset, historical leakage data, and environmental data of the target pipeline are concatenated to determine the first fused feature data. This first fused feature data is then input into a risk assessment model to determine the leakage probability value of the target pipeline at the current time.

[0081] S104. Based on the leakage probability value and leakage impact information, determine the comprehensive leakage risk of the target pipeline at the current time.

[0082] In one possible implementation, a risk assessment matrix is ​​constructed based on the leakage probability value and leakage impact information. If the consistency index of the risk assessment matrix is ​​less than a preset consistency threshold, the overall leakage risk of the target pipeline at the current time is determined based on the risk assessment matrix.

[0083] Among them, the risk assessment matrix is ​​a matrix used in the analytic hierarchy process to quantify the relative importance of indicators.

[0084] As described in the above technical solution, this application obtains the current operation and maintenance data and historical leakage data of the target pipeline, and determines a feature dataset based on the pipeline's operating parameters and inspection data. This achieves the fusion of multiple aspects of the target pipeline's feature data, resulting in a more accurate assessment of leakage risk. Furthermore, a risk assessment model processes the feature dataset, historical leakage data, and environmental data to accurately determine the probability of leakage in the target pipeline. Based on the leakage probability value and leakage impact information, a comprehensive leakage risk assessment of the target pipeline is determined. This enables accurate assessment of the leakage risk of oil and gas pipelines, ensuring their stable operation.

[0085] In some embodiments, such as Figure 2 As shown, the above method S103 specifically includes: S201-S202.

[0086] S201. Process the feature dataset, historical leakage data, and environmental data to determine the fusion feature parameters.

[0087] In one possible implementation, the feature dataset, historical leakage data, and environmental data are concatenated into a matrix. The concatenated matrix is ​​then normalized to obtain the first fused feature parameters.

[0088] S202. Input the first fusion feature parameter into the risk assessment model to determine the leakage probability value.

[0089] The risk assessment model includes an input layer, a hidden layer, and an output layer.

[0090] 1-1. Input layer.

[0091] The input layer is used to obtain input data.

[0092] In one possible implementation, the first fusion feature parameters are obtained through the input layer.

[0093] 1-2, Hidden Layer.

[0094] The hidden layer is used to assign weights to the input data and determine the feature parameters.

[0095] In one possible implementation, the weights corresponding to multiple feature parameter types in the first fusion feature parameters are adjusted through a preset attention mechanism in the hidden layer to determine the second fusion feature parameters.

[0096] Specifically, the hidden layer obtains the feature map of the input layer and determines the feature mean of the c-th channel, with each channel corresponding to a feature parameter type.

[0097] For example, the feature mean satisfies the following formula two:

[0098]

[0099] Where, x c，i，j Used to represent the feature map element of the c-th channel, H×W represents the feature map size, c represents the channel index, represents the feature type, and z represents the feature map element. c Used to represent the global mean of the c-th channel, i is the index in the height direction of the feature map, and j is the index in the width direction of the feature map.

[0100] Furthermore, z is weighted by the weight matrix W1. c Dimensionality reduction: The intermediate feature vectors after dimensionality reduction satisfy the following formula (Formula 3).

[0101] y1=W1·z c Formula 3.

[0102] Where y1 represents the intermediate feature vector after dimensionality reduction, and W1 represents the weight matrix. r represents the dimensionality reduction coefficient, and C represents the total number of channels.

[0103] Next, the ReLU function is used to activate the dimensionality reduction result, and the activated nonlinear eigenvectors are determined. The nonlinear eigenvectors satisfy the following formula:

[0104] Formula 4: y2 = ReLU(y1).

[0105] Here, y2 represents the nonlinear eigenvector after ReLU activation, and ReLU represents the modified linear unit function.

[0106] The channel weights are calculated using the weight matrix W2 and the Sigmoid function, and the channel weights satisfy the following formula:

[0107] w c =σ(W2·y2) Formula 5.

[0108] Among them, w c Used to represent the attention weight of the c-th channel, σ represents the Sigmoid function, r represents the dimensionality reduction coefficient, and C represents the total number of channels.

[0109] The channel weights are multiplied by the original feature map to obtain the enhanced feature map, which is also the second fusion feature parameter.

[0110] In another possible implementation, the second fusion feature parameters are interval-mapped through a hidden layer to determine the weighted feature parameters.

[0111] 1-3, Output Layer.

[0112] The output layer is used to normalize the feature parameters output by the hidden layer, and then outputs the leakage probability value.

[0113] In one possible implementation, the leakage probability value of the target pipeline is determined by normalizing the weighted feature parameters through the output layer.

[0114] Specifically, the output layer normalizes the weighted feature parameters using the Sigmoid function to determine the leakage probability value. The leakage probability value ranges from [0,1], with a higher value indicating a higher risk of leakage.

[0115] In some embodiments, the risk assessment model is trained using a training dataset to obtain the trained risk assessment model.

[0116] The training dataset includes the first fusion feature parameters within a historical time period and the target pipeline leakage probability within a historical time period.

[0117] In one possible implementation, a predetermined number of training datasets are divided into a predetermined number of training subsets and one test set, and this process is repeated a predetermined number of times. Based on the training subsets and the test set, the F1 score and AUC value are determined. Based on the mean of the F1 score and AUC value, the model parameters corresponding to the training dataset with the largest mean among the predetermined number of training datasets are used as the model parameters of the risk assessment model, thus determining the trained risk assessment model.

[0118] As can be seen from the above technical solution, this application processes feature datasets, historical leakage data, and environmental data to determine a first fusion parameter, thereby fusing features of multiple influencing factors and making the judgment of pipeline leakage probability more accurate. Furthermore, the fused feature parameter is input into a risk assessment model to accurately determine the leakage probability value of the target pipeline.

[0119] In some embodiments, an incremental dataset of the target pipeline is obtained, and the risk assessment model is updated based on the incremental dataset using online learning algorithms and regularization methods.

[0120] In one possible implementation, the incremental dataset of the target pipeline and the acquisition time corresponding to each data point in the incremental dataset are obtained. Based on the incremental dataset and acquisition time, weights are assigned to the data in the incremental dataset to determine the updated dataset. The parameters of the risk assessment model are then updated based on the updated dataset to determine the risk assessment model.

[0121] For example, incremental data is acquired via the BeiDou satellite-to-ground communication link. The incoming incremental data is then verified; for instance, BeiDou short messages use CRC-16 verification, and the 4G link enables TCP retransmission to ensure data integrity. Furthermore, the verified new data is dynamically weighted using a time-decay method. For example, based on the data arrival time, data from the past 24 hours is given a higher weight according to an exponential decay model, while historical data has a correspondingly lower weight. Simultaneously, data is processed differently based on its data type; for example, the weight of key monitoring data such as cathodic protection potential and pipeline internal pressure is increased by 20%-50%, while the weight of auxiliary data such as inspection trajectories is reduced. The processed data is then normalized and input into the risk assessment model.

[0122] It should be noted that the online learning algorithm in this embodiment is based on the principle of weighted stochastic gradient descent, assigning higher weights to incremental data to quickly respond to changes in the target pipeline's operation, and avoiding parameter oscillations through a learning rate decay strategy. In the regularization method, L2 regularization prevents overfitting by adding penalty terms to the weights, and the Dropout mechanism randomly ignores some neurons during training to enhance the model's generalization ability.

[0123] In another possible implementation, if the amount of new data exceeds a preset threshold, a preset number of network layers in the risk assessment model are frozen, and the remaining network layers are trained with a preset learning rate.

[0124] As described in the above technical solution, this application obtains an updated dataset by acquiring an incremental dataset of the target pipeline and the acquisition time, and then assigns weights to the data in the incremental dataset. The risk assessment model is then updated using this updated dataset, enabling the risk assessment model to continuously update as the amount of data increases, and to more accurately determine the leakage probability value of the target pipeline.

[0125] In some embodiments, after updating the risk assessment model based on incremental data, the real-time performance of the risk assessment model is obtained. If the risk assessment model is continuously updated a preset number of times and the performance decline is greater than or equal to a preset performance threshold, the model is automatically rolled back to the pre-update risk assessment model. The risk assessment model is then retrained based on data from a preset historical time period.

[0126] As can be seen from the above technical solution, this application obtains the real-time performance of the risk assessment model after the risk assessment model is updated, and takes measures to retrain the risk assessment model when the performance degrades, thereby ensuring the performance of the risk assessment model.

[0127] In some embodiments, such as Figure 3 As shown, the above method S104 specifically includes: S301-S303.

[0128] S301. Construct a risk assessment matrix based on leakage probability values ​​and leakage impact information.

[0129] The risk assessment matrix is ​​used to determine the overall leakage risk of the target pipeline. Leakage influencing factors include the scope of impact and response time.

[0130] One possible implementation involves obtaining the number of target objects and the leakage response time within a preset range of the target pipeline. Based on the number of target objects and environmental data, factors affecting the scope of influence are determined. Based on the leakage response time, factors affecting the response time are determined.

[0131] The target objects are those affected by the leak in the target pipeline, such as residential areas, water sources, and critical facilities. The leak response time includes the leak alarm time and the alarm response time. The leak alarm time is the time when monitoring pipeline data triggers an early warning, and the alarm response time is the time when manual confirmation of a leak in the target pipeline and the initiation of emergency measures.

[0132] Specifically, the number of target objects, the current pressure level of the pipeline, and the soil permeability are normalized and mapped according to preset thresholds to determine the influencing factors. The sum of the first warning time and the second confirmation time is compared with the preset emergency event threshold, and the response time factor is determined after normalization.

[0133] It should be noted that the range of the influence factor is [0,1], and the larger the value, the wider the range of the target pipeline leakage. The range of the response time factor is [0,1].

[0134] One possible implementation involves constructing a risk assessment matrix based on the leakage probability value and leakage impact information.

[0135] Among them, the risk judgment matrix is ​​a risk indicator judgment matrix with a scale of 1-9, and the element a in the matrix ij Used to indicate the importance of index i relative to index j.

[0136] S302. Based on the risk assessment matrix, determine the weights of the leakage probability value and the weights of the leakage impact information.

[0137] In one possible implementation, a matrix consistency index is determined based on the risk assessment matrix. If the matrix consistency index is less than a preset consistency threshold, the weights of the leakage probability value and the leakage impact information are determined based on the risk assessment matrix.

[0138] For example, the matrix consistency metric satisfies the following formula six:

[0139]

[0140] Where CI is used to represent the matrix consistency index, λ max The largest eigenvalue of the matrix used to represent the risk assessment matrix.

[0141] For example, the matrix consistency ratio satisfies the following formula seven.

[0142]

[0143] Wherein, CR is used to represent the matrix consistency ratio, CI is used to represent the matrix consistency index, and RI is used to represent the matrix random consistency index.

[0144] For example, with a consistency threshold of 0.1, when the consistency index of the risk judgment matrix is ​​less than 0.1, the weights of the leakage probability value and the leakage impact information are determined based on the risk judgment matrix.

[0145] In some embodiments, historical risk data of the target pipeline is acquired when the matrix consistency ratio is greater than or equal to a preset threshold. Then, based on the historical risk data, the risk assessment matrix is ​​corrected until the matrix consistency ratio of the corrected risk assessment matrix is ​​less than the preset consistency threshold.

[0146] Historical risk data is used to indicate the factors that caused the target pipeline to leak during historical events.

[0147] For example, historical risk data may include the frequency of correlations between factors such as geological fault zones and fluctuations in the operating pressure of the target pipeline and leakage events in the target pipeline within a historical time period.

[0148] Specifically, when the matrix consistency ratio is greater than or equal to a preset threshold, the element with the largest inconsistency is determined by calculating the consistency ratio of each row of the risk assessment matrix. Then, based on historical risk data of the target pipeline, the relative importance scale of the indicator corresponding to the element with the largest inconsistency is adjusted. This process continues until the matrix consistency ratio of the corrected risk assessment matrix is ​​less than the preset consistency threshold.

[0149] S303. Based on the weight of the leakage probability value, the leakage probability value, the leakage impact information, and the weight of the leakage impact information, determine the comprehensive leakage risk of the target pipeline at the current time.

[0150] In one possible implementation, the leakage probability value and leakage impact information are weighted and summed using the weights of the leakage probability value and the leakage impact information respectively, to determine the comprehensive leakage risk value of the target pipeline at the current time. Based on the comprehensive leakage risk value and a first risk threshold, a second risk threshold, and a third risk threshold, the comprehensive leakage risk of the target pipeline at the current time is determined.

[0151] Specifically, if the overall leakage risk value is less than the first risk threshold, the overall leakage risk of the target pipeline is low; if the overall leakage risk value is greater than or equal to the first risk threshold and less than the second risk threshold, the overall leakage risk of the target pipeline is medium; if the overall leakage risk value is greater than or equal to the second risk threshold and less than the third risk threshold, the overall leakage risk of the target pipeline is high; and if the overall leakage risk value is greater than or equal to the third risk threshold, the overall leakage risk of the target pipeline is extremely high.

[0152] For example, taking a first risk threshold of 0.3, a second risk threshold of 0.6, and a third risk threshold of 0.8 as an example: If the overall leakage risk value is less than 0.3, the overall leakage risk of the target pipeline is low; if the overall leakage risk value is greater than or equal to 0.3 and less than 0.6, the overall leakage risk of the target pipeline is medium; if the overall leakage risk value is greater than or equal to 0.6 and less than 0.8, the overall leakage risk of the target pipeline is high; and if the overall leakage risk value is greater than or equal to 0.8, the overall leakage risk of the target pipeline is extremely high.

[0153] In another possible implementation, emergency measures are determined based on the overall leakage risk of the target pipeline at the current time.

[0154] For example, if the overall leakage risk of the target pipeline is low, the pipeline is in safe operation and routine monitoring is maintained. If the overall leakage risk of the target pipeline is high, the data monitoring density needs to be increased to eliminate potential risks. If the overall leakage risk of the target pipeline is extremely high, an emergency shutdown plan should be implemented immediately, and on-site handling should be carried out simultaneously.

[0155] As described in the above technical solution, this application constructs a risk assessment matrix based on leakage probability values ​​and leakage impact information. When the consistency ratio of the risk assessment matrix is ​​less than a preset consistency threshold, the weights of the leakage probability values ​​and leakage impact information are determined. Furthermore, based on the leakage probability values ​​and leakage impact information, the overall risk value of the target pipeline is comprehensively assessed. This accurately determines the overall leakage risk of the target pipeline at the current time.

[0156] In some embodiments, a risk heatmap is determined based on the overall leakage risk of the target pipeline at the current time.

[0157] The risk heatmap is used to indicate the risk level of each area in the target pipeline network. The target pipeline network includes multiple target pipelines.

[0158] In one possible implementation, the comprehensive risk value corresponding to multiple target pipelines in the pipeline network is obtained through the methods S101-S104 described above. Then, the target pipeline network is divided into multiple preset regions, and a risk heatmap is determined based on the comprehensive risk value corresponding to the multiple target pipelines, the alarm threshold, and the multiple preset regions.

[0159] For example, the comprehensive risk value of the target pipeline in each preset region is obtained. A Gaussian kernel function is used to smooth the comprehensive risk value of the target pipeline in each preset region to determine the corresponding regional risk value. If the regional risk value is less than a first alarm threshold, the region is considered risk-free. If the regional risk value is greater than or equal to the first alarm threshold but less than a second alarm threshold, the region is displayed as a yellow risk area. If the regional risk value is greater than the second alarm threshold, the region is displayed as a red risk area.

[0160] As can be seen from the above technical solution, this application constructs a risk heat map of the entire target pipeline network based on the comprehensive risk value of the target pipeline in each preset area. This makes the risks of the pipeline network visible, which is beneficial for maintaining the stability of the pipeline network operation.

[0161] The foregoing mainly describes the solutions provided in the embodiments of this application from a methodological perspective. It is understood that the pipeline leakage risk determination device, in order to achieve the above functions, includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that the pipeline leakage risk determination method steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0162] This application also provides a device for determining pipeline leakage risk. This device can be a server, a CPU within the server, a module within the server used to determine pipeline leakage risk, or a client within the server used to determine pipeline leakage risk.

[0163] This application embodiment can divide the pipeline leakage risk determination device into functional modules or functional units according to the above method example. For example, each function can be divided into a separate functional module or functional unit, or two or more functions can be integrated into one processing unit. The integrated module can be implemented in hardware or in software functional modules or functional units. The module or unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0164] When dividing each function into modules according to its corresponding function. Figure 4 A structural diagram of a device for determining pipeline leakage risk provided in this application is shown below. Figure 4 As shown, the device for determining pipeline leakage risk can be used to perform... Figure 1 , Figure 2 , Figure 3 The method for determining pipeline leakage risk is shown. The pipeline leakage risk determination device 40 includes: an acquisition unit 401 and a processing unit 402.

[0165] The acquisition unit 401 is used to acquire the current operation and maintenance data and historical leakage data of the target pipeline. The operation data includes the target pipeline's operating parameters, inspection data, environmental data, and leakage impact information. The leakage impact information is used to indicate the impact caused by the leakage of the target pipeline.

[0166] Processing unit 402 is used to determine a feature dataset based on the operating parameters of the target pipeline and inspection data;

[0167] The processing unit 402 is used to process the feature dataset, historical leakage data and environmental data of the target pipeline through a risk assessment model to determine the leakage probability value of the target pipeline at the current time. The risk assessment model is used to determine the probability of leakage of the target pipeline.

[0168] Processing unit 402 is used to determine the overall leakage risk of the target pipeline at the current time based on the leakage probability value and leakage impact information.

[0169] In one possible approach, the operating parameters of the target pipeline include cathode potential data and pipeline internal pressure data; the processing unit 402 is specifically used to perform mode decomposition on the cathode potential data using a preset empirical mode decomposition algorithm to determine energy proportion feature data; to process the pipeline internal pressure data using a preset information entropy construction algorithm to determine energy entropy feature data, the energy entropy feature being used to indicate the frequency distribution complexity of the internal pressure fluctuations of the target pipeline; and to process the inspection data using a preset spatiotemporal feature extraction algorithm to determine inspection spatiotemporal feature data, the inspection spatiotemporal feature being used to indicate the quality of the inspection of the target pipeline; the feature dataset includes energy proportion feature data, energy entropy feature data, and inspection spatiotemporal feature data.

[0170] In one possible approach, the processing unit 402 is specifically used to process the feature dataset, historical leakage data, and environmental data to determine a first fused feature parameter. The first fused feature parameter has multiple feature parameter dimensions, and each feature parameter dimension corresponds to a feature parameter type. Each data in the feature dataset, the historical leakage data, and the environmental data each correspond to a parameter type. The fused feature parameter is then input into a risk assessment model to determine the leakage probability value.

[0171] In one possible approach, the risk assessment model includes an input layer, a hidden layer, and an output layer. The processing unit 402 is specifically used to obtain first fusion feature parameters through the input layer; adjust the weights corresponding to multiple feature parameter types in the first fusion feature parameters through a preset attention mechanism in the hidden layer to determine second fusion feature parameters; perform interval mapping on the second fusion feature parameters through the hidden layer to determine weighted feature parameters; and normalize the weighted feature parameters through the output layer to determine the leakage probability value of the target pipeline.

[0172] In one possible approach, processing unit 402 is specifically configured to: construct a risk assessment matrix based on leakage probability values ​​and leakage impact information; determine a matrix consistency ratio based on the risk assessment matrix; if the matrix consistency ratio is less than a preset consistency threshold, determine the weights of the leakage probability values ​​and leakage impact information based on the risk assessment matrix; and determine the comprehensive risk value of the target pipeline at the current time based on the leakage probability values, their weights, the leakage impact information, and their weights, whereby the comprehensive risk value indicates the magnitude of the risk of leakage in the target pipeline. Based on the comprehensive risk value, the comprehensive leakage risk of the target pipeline at the current time is determined.

[0173] In one possible approach, the processing unit 402 is specifically used to obtain historical risk data of the target pipeline when the matrix consistency index is greater than or equal to a preset consistency threshold. The historical risk data is used to indicate the factors that caused the target pipeline to leak in a historical time. Based on the historical risk data, the risk judgment matrix is ​​corrected until the matrix consistency index of the corrected risk judgment matrix is ​​less than the preset consistency threshold.

[0174] In one possible approach, the processing unit 402 is specifically used to acquire the incremental dataset of the target pipeline and the collection time corresponding to each data point in the incremental dataset. The incremental dataset consists of the operational data of the target pipeline that was added in the previous preset time period before the current time. Based on the incremental dataset and the collection time, the data in the incremental dataset are weighted and an updated dataset is determined. Based on the updated dataset, the parameters of the risk assessment model are updated to determine the risk assessment model.

[0175] In one possible approach, the leakage impact information includes an impact range factor and a response time factor. The acquisition unit 401 is specifically used to acquire the number of target objects within a preset range of the target pipeline and the leakage response time. The target objects are objects affected by the leakage of the target pipeline, and the leakage response time includes the leakage alarm time and the alarm response time. Based on the number of sensitive targets and environmental data, the impact range factor is determined; based on the leakage alarm time and the alarm response time, the response time factor is determined.

[0176] This disclosure also provides a computer-readable storage medium storing instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the pipeline leakage risk determination method provided in the embodiments of this disclosure described above.

[0177] This disclosure also provides a computer program product containing instructions that, when run on an electronic device, cause the electronic device to execute the pipeline leakage risk determination method provided in the above-described embodiments of this disclosure.

[0178] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections having one or more wires; portable computer disks; hard disks; random access memory (RAM); read-only memory (ROM); erasable programmable read-only memory (EPROM); registers; hard disks; optical fibers; portable compact disc read-only memory (CD-ROM); optical storage devices; magnetic storage devices; or any suitable combination thereof; or any other form of computer-readable storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In the embodiments of this application, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0179] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0180] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0181] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the classified units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0182] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0183] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, essentially, or the part that contributes to the prior art, or a complete or partial classification of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0184] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining pipeline leakage risk, characterized in that, The method includes: The system acquires the current operation and maintenance data and historical leakage data of the target pipeline. The operation data includes the operating parameters, inspection data, environmental data, and leakage impact information of the target pipeline. The leakage impact information is used to indicate the impact caused by the leakage of the target pipeline. Based on the operating parameters of the target pipeline and the inspection data, a feature dataset is determined; The risk assessment model is used to process the feature dataset, the historical leakage data, and the environmental data of the target pipeline to determine the leakage probability value of the target pipeline at the current time. The risk assessment model is used to determine the probability of the target pipeline leaking. Based on the leakage probability value and the leakage impact information, the overall leakage risk of the target pipeline at the current time is determined.

2. The method according to claim 1, characterized in that, The operating parameters of the target pipeline include cathode potential data and pipeline internal pressure data; The determination of the feature dataset based on the operating parameters of the target pipeline and the inspection data includes: The cathode potential data is decomposed into modes using a preset empirical mode decomposition algorithm to determine the energy proportion characteristic data. The internal pressure data of the pipeline is processed by a preset information entropy construction algorithm to determine the energy entropy feature data. The energy entropy feature is used to indicate the frequency distribution complexity of the internal pressure fluctuation of the target pipeline. The inspection data is processed by a preset spatiotemporal feature extraction algorithm to determine the inspection spatiotemporal feature data, which is used to indicate the quality of the inspection of the target pipeline. The feature dataset includes energy percentage feature data, energy entropy feature data, and inspection spatiotemporal feature data.

3. The method according to claim 1, characterized in that, By processing the feature dataset, historical leakage data, and environmental data of the target pipeline using a risk assessment model, the leakage probability value of the target pipeline at the current time is determined, including: The feature dataset, the historical leakage data, and the environmental data are processed to determine a first fusion feature parameter. The first fusion feature parameter has multiple feature parameter dimensions, and each feature parameter dimension corresponds to a feature parameter type. Each data in the feature dataset, the historical leakage data, and the environmental data each correspond to a parameter type. The first fusion feature parameter is input into the risk assessment model to determine the leakage probability value.

4. The method according to claim 3, characterized in that, The risk assessment model includes an input layer, a hidden layer, and an output layer. The step of inputting the first fused feature parameters into the risk assessment model to determine the leakage probability value includes: The first fused feature parameters are obtained through the input layer; The second fusion feature parameter is determined by adjusting the weights corresponding to multiple feature parameter types in the first fusion feature parameter through the hidden layer based on a preset attention mechanism; The weighted feature parameters are determined by performing interval mapping on the second fused feature parameters through the hidden layer; The output layer normalizes the weighted feature parameters to determine the leakage probability value of the target pipeline.

5. The method according to claim 1, characterized in that, The determination of the comprehensive leakage risk of the target pipeline at the current time based on the leakage probability value and the leakage impact information includes: Based on the leakage probability value and the leakage impact information, a risk assessment matrix is ​​constructed; Based on the risk assessment matrix, determine the matrix consistency ratio; If the matrix consistency ratio is less than a preset consistency threshold, the weights of the leakage probability value and the leakage impact information are determined based on the risk judgment matrix. Based on the leakage probability value, the weight of the leakage probability value, the leakage impact information, and the weight of the leakage impact information, the comprehensive risk value of the target pipeline at the current time is determined, and the comprehensive risk value is used to indicate the magnitude of the risk of leakage in the target pipeline; Based on the comprehensive risk value, the comprehensive leakage risk of the target pipeline at the current time is determined.

6. The method according to claim 5, characterized in that, The method further includes: If the matrix consistency index is greater than or equal to a preset consistency threshold, historical risk data of the target pipeline is obtained. The historical risk data is used to indicate the factors that caused the target pipeline to leak at the historical time. Based on the historical risk data, the risk judgment matrix is ​​corrected until the matrix consistency index of the corrected risk judgment matrix is ​​less than the preset consistency threshold.

7. The method according to any one of claims 1-6, characterized in that, The risk assessment model was determined in the following way: Obtain the incremental dataset of the target pipeline and the collection time corresponding to each data point in the incremental dataset. The incremental dataset consists of the operational data of the target pipeline that was added in the previous preset time period before the current time. Based on the incremental dataset and the collection time, weights are assigned to the data in the incremental dataset to determine the updated dataset; The risk assessment model is updated with parameters based on the updated dataset to determine the risk assessment model.

8. The method according to claim 1 or 5, characterized in that, The leakage impact information includes an impact range factor and a response time factor. Obtaining the leakage impact factors includes: The number of target objects and the leakage response time within a preset range of the target pipeline are obtained. The target objects are objects affected by the leakage of the target pipeline. The leakage response time includes the leakage alarm time and the alarm response time. The influence range factor is determined based on the number of sensitive targets and the environmental data. The response time factor is determined based on the leakage alarm time and the alarm response time.

9. A device for determining pipeline leakage risk, characterized in that, The device includes: The acquisition unit is used to acquire the current operation and maintenance data and historical leakage data of the target pipeline. The operation data includes the operation parameters, inspection data, environmental data and leakage impact information of the target pipeline. The leakage impact information is used to indicate the impact caused by the leakage of the target pipeline. The processing unit is used to determine a feature dataset based on the operating parameters of the target pipeline and the inspection data; The processing unit is used to process the feature dataset, the historical leakage data, and the environmental data of the target pipeline through a risk assessment model to determine the leakage probability value of the target pipeline at the current time. The risk assessment model is used to determine the probability of the target pipeline leaking. The processing unit is used to determine the overall leakage risk of the target pipeline at the current time based on the leakage probability value and the leakage impact information.

10. An electronic device, characterized in that, It includes a memory and a processor; the memory and the processor are coupled; the memory is used to store instructions executable by the processor; when the processor executes the instructions, it performs the method as described in any one of claims 1-8.

Images