A gas pipeline safety assessment and early warning system and method based on data analysis

Abstract

The present application relates to the technical field of gas pipeline safety evaluation, and more particularly to a gas pipeline safety evaluation and early warning system and method based on data analysis. The method comprises the following steps: simultaneously collecting gas pressure data of the gas pipeline at multiple points, and performing time-space synchronous processing to obtain aligned gas pressure data, thereby constructing a continuous gas pressure field of the pipeline; based on the gas pressure field, identifying differences in pipeline pressure distribution and estimating gas transient flow velocity; continuously recording the gas transient flow velocity and inferring flow velocity changes; combining the gas pressure field data to determine the deformation of the inner wall of the pipeline; positioning the deformation area and reconstructing the pipeline framework; analyzing the deformation response and identifying the deformation type; identifying the corrosion evolution trend and performing safety evaluation; and judging the degree of gas flow and performing physical safety evaluation. The present application realizes dynamic monitoring of the gas pipeline, promotes a data-driven safety management mode, improves the early warning capability for potential risks, and forms a more intelligent safety evaluation and early warning mechanism.

Description

Technical Field

[0001] This invention relates to the field of gas pipeline safety assessment technology, and in particular to a gas pipeline safety assessment and early warning system and method based on data analysis. Background Technology

[0002] Traditional gas pipeline safety assessment techniques often rely on manual inspections and periodic checks, leading to insufficient timeliness in identifying and responding to potential risks. Many accidents occur undetected. The limitations of traditional methods in data collection and analysis create blind spots in pipeline condition assessments, especially in complex environments, failing to fully reflect the actual operating conditions of pipelines and increasing safety hazards. In existing technologies, the collection and analysis of gas pressure data often lack timeliness and accuracy, making it impossible to monitor the gas flow status in real time. The identification of pressure distribution differences and the calculation of flow velocity often rely on static models, ignoring the significant impact of transient changes on pipeline safety. The identification and assessment methods for pipeline inner wall deformation are relatively simple, making it difficult to accurately classify corrosion and physical deformation types, thus affecting the accuracy and effectiveness of safety warnings. Existing technologies are ill-equipped to cope with increasingly complex pipeline operating environments, resulting in lagging safety management. Summary of the Invention

[0003] Therefore, it is necessary to provide a data analysis-based gas pipeline safety assessment and early warning system and method to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a data analysis-based method for safety assessment and early warning of gas pipelines includes the following steps:

[0005] Step S1: Collect gas pressure data from multiple points simultaneously; perform spatiotemporal synchronization enhancement on the gas pressure data to obtain aligned pressure data; construct a continuous pressure field for the pipeline based on the aligned pressure data;

[0006] Step S2: Identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; estimate the transient flow velocity of the gas through the differences in pipeline pressure distribution;

[0007] Step S3: Continuously record the transient gas flow velocity at multiple time points and infer the flow velocity changes to obtain gas flow velocity change data; determine the pipeline inner wall deformation data through the continuous gas pressure field and gas flow velocity change data.

[0008] Step S4: Locate the deformation area based on the pipeline inner wall deformation data and reconstruct the deformed pipeline frame; analyze the deformation response based on the pipeline continuous gas pressure field and gas flow velocity change data, and identify the pipeline deformation type.

[0009] Step S5: When the pipeline deformation type is pipeline corrosion deformation data, identify the corrosion evolution trend based on the pipeline corrosion deformation data, and conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, determine the gas flow level based on the pipeline physical deformation data, and conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

[0010] This invention achieves comprehensive monitoring of gas pipeline pressure data through multi-point synchronous gas pressure acquisition, ensuring data accuracy and consistency and laying a solid foundation for subsequent analysis. The application of spatiotemporal synchronization enhancement technology improves data alignment accuracy, making the construction of the continuous gas pressure field in the pipeline more reliable. This effectively reflects internal pressure changes within the pipeline, enhances the ability to identify differences in pipeline pressure distribution, facilitates accurate estimation of transient gas velocity, and strengthens the understanding of gas flow characteristics. Continuously recorded transient gas velocity data provides necessary information support for real-time monitoring and analysis. By combining the pipeline's continuous gas pressure field and velocity change data, the deformation of the pipeline's inner wall can be determined in a timely manner, ensuring the early identification of potential structural problems and accurate location of deformation areas. This improves the ability to analyze pipeline deformation response and helps reconstruct the deformed pipeline framework, thus providing comprehensive analysis of subsequent pressure field and velocity change data. The function of identifying pipeline deformation types further refines the basis for safety assessment and enhances the monitoring of pipeline corrosion and physical deformation. By analyzing corrosion deformation data, the system can promptly identify corrosion trends and conduct in-depth corrosion safety assessments based on pre-set safety margins, providing a scientific basis for preventative measures. The assessment of physical deformation data ensures accurate evaluation of gas flow, and the physical safety assessment combined with pressure tolerance thresholds effectively reduces safety hazards caused by pipeline damage. This enables dynamic monitoring of gas pipelines, promotes a data-driven safety management model, reduces uncertainty caused by human judgment, improves early warning capabilities for potential risks, drives the intelligent development of the gas industry, establishes a sustainable safety assessment and early warning mechanism, enhances the overall safety of the industry, provides effective guarantees for the long-term stable operation of gas pipelines, ultimately creates conditions for a safe gas environment, promotes public safety maintenance, drives the application and promotion of related technologies, provides important reference and guidance for future gas pipeline safety management, ensures the stability and security of energy supply, and promotes technological innovation and progress in the industry.

[0011] This invention also provides a data analysis-based gas pipeline safety assessment and early warning system for executing the data analysis-based gas pipeline safety assessment and early warning method described above. The data analysis-based gas pipeline safety assessment and early warning system includes:

[0012] The multi-point gas pressure synchronization module is used to simultaneously collect gas pressure data from multiple points in the gas pipeline; it performs spatiotemporal synchronization enhancement on the gas pipeline pressure data to obtain aligned pressure data; and it constructs a continuous pressure field for the pipeline based on the aligned pressure data.

[0013] The gas flow velocity estimation module is used to identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; and to estimate the transient gas flow velocity through the differences in pipeline pressure distribution.

[0014] The deformation response detection module is used to continuously record the transient flow velocity of gas at multiple time points and infer the flow velocity change to obtain gas flow velocity change data; the deformation data of the inner wall of the pipeline is judged by the continuous gas pressure field and gas flow velocity change data of the pipeline.

[0015] The deformation type identification module is used to locate the deformation area based on the deformation data of the inner wall of the pipeline and reconstruct the deformed pipeline frame; based on the deformed pipeline frame, it performs deformation response analysis on the continuous gas pressure field and gas flow velocity change data of the pipeline and identifies the pipeline deformation type.

[0016] The safety early warning decision module is used to identify the corrosion evolution trend based on the pipeline corrosion deformation data when the pipeline deformation type is pipeline corrosion deformation data, and to conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, it determines the gas flow level based on the pipeline physical deformation data, and to conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

[0017] This invention, through the application of a multi-point gas pressure synchronization module, achieves comprehensive acquisition and spatiotemporal synchronization of gas pipeline pressure data, ensuring data consistency and accuracy, and providing a foundation for subsequent analysis. The constructed continuous gas pressure field of the pipeline can effectively reflect the pressure changes inside the pipeline, helping to identify potential pressure anomalies. The gas flow velocity estimation module, by analyzing pressure distribution differences, can accurately estimate the transient flow velocity of the gas, providing a deeper understanding of the gas flow characteristics. The deformation response detection module, by continuously recording the transient flow velocity of the gas, can monitor flow velocity changes in real time, providing important data support for judging the deformation of the pipeline inner wall. The introduction of the deformation type identification module enables precise location of the deformation area and reconstruction of the deformed pipeline framework, laying the foundation for subsequent pressure field and flow velocity change data analysis. The safety early warning decision module functions to promptly identify the corrosion evolution trend when pipeline corrosion deformation is detected, and conduct in-depth corrosion risk assessment in combination with preset safety margins to ensure safety. By taking measures before corrosion problems escalate and avoiding safety hazards, the system can assess gas flow based on physical deformation and conduct physical safety assessments based on pressure tolerance thresholds. This effectively reduces the risk of accidents caused by physical damage to pipelines. The entire system integrates multiple data analysis technologies, enabling comprehensive monitoring and dynamic assessment of gas pipelines. This enhances the intelligence level of pipeline safety management, improves the real-time monitoring capability of pipeline status, strengthens the early warning capability for potential risks, reduces the impact of human factors on safety assessments, promotes a data-driven safety management model, provides a scientific basis for the long-term stable operation of gas pipelines, and ultimately achieves comprehensive protection for gas pipeline safety. This promotes the intelligent development of the gas industry, forms a sustainable safety assessment and early warning mechanism, provides important references for the future application and promotion of related technologies, improves the overall safety and economy of the industry, creates conditions for a safe gas environment in society, and promotes the protection and improvement of public safety. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the steps of a data analysis-based method for safety assessment and early warning of gas pipelines.

[0019] Figure 2 This is a detailed flowchart illustrating the implementation steps of step S2;

[0020] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0022] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0023] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0024] To achieve the above objectives, please refer to Figures 1 to 2 A data-driven method for safety assessment and early warning of gas pipelines includes the following steps:

[0025] Step S1: Collect gas pressure data from multiple points simultaneously; perform spatiotemporal synchronization enhancement on the gas pressure data to obtain aligned pressure data; construct a continuous pressure field for the pipeline based on the aligned pressure data;

[0026] Step S2: Identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; estimate the transient flow velocity of the gas through the differences in pipeline pressure distribution;

[0027] Step S3: Continuously record the transient gas flow velocity at multiple time points and infer the flow velocity changes to obtain gas flow velocity change data; determine the pipeline inner wall deformation data through the continuous gas pressure field and gas flow velocity change data.

[0028] Step S4: Locate the deformation area based on the pipeline inner wall deformation data and reconstruct the deformed pipeline frame; analyze the deformation response based on the pipeline continuous gas pressure field and gas flow velocity change data, and identify the pipeline deformation type.

[0029] Step S5: When the pipeline deformation type is pipeline corrosion deformation data, identify the corrosion evolution trend based on the pipeline corrosion deformation data, and conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, determine the gas flow level based on the pipeline physical deformation data, and conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

[0030] This invention achieves comprehensive monitoring of gas pipeline pressure data through multi-point synchronous gas pressure acquisition, ensuring data accuracy and consistency and laying a solid foundation for subsequent analysis. The application of spatiotemporal synchronization enhancement technology improves data alignment accuracy, making the construction of the continuous gas pressure field in the pipeline more reliable. This effectively reflects internal pressure changes within the pipeline, enhances the ability to identify differences in pipeline pressure distribution, facilitates accurate estimation of transient gas velocity, and strengthens the understanding of gas flow characteristics. Continuously recorded transient gas velocity data provides necessary information support for real-time monitoring and analysis. By combining the pipeline's continuous gas pressure field and velocity change data, the deformation of the pipeline's inner wall can be determined in a timely manner, ensuring the early identification of potential structural problems and accurate location of deformation areas. This improves the ability to analyze pipeline deformation response and helps reconstruct the deformed pipeline framework, thus providing comprehensive analysis of subsequent pressure field and velocity change data. The function of identifying pipeline deformation types further refines the basis for safety assessment and enhances the monitoring of pipeline corrosion and physical deformation. By analyzing corrosion deformation data, the system can promptly identify corrosion trends and conduct in-depth corrosion safety assessments based on pre-set safety margins, providing a scientific basis for preventative measures. The assessment of physical deformation data ensures accurate evaluation of gas flow, and the physical safety assessment combined with pressure tolerance thresholds effectively reduces safety hazards caused by pipeline damage. This enables dynamic monitoring of gas pipelines, promotes a data-driven safety management model, reduces uncertainty caused by human judgment, improves early warning capabilities for potential risks, drives the intelligent development of the gas industry, establishes a sustainable safety assessment and early warning mechanism, enhances the overall safety of the industry, provides effective guarantees for the long-term stable operation of gas pipelines, ultimately creates conditions for a safe gas environment, promotes public safety maintenance, drives the application and promotion of related technologies, provides important reference and guidance for future gas pipeline safety management, ensures the stability and security of energy supply, and promotes technological innovation and progress in the industry.

[0031] In this embodiment of the invention, the gas pipeline safety assessment and early warning method based on data analysis includes the following steps:

[0032] Step S1: Collect gas pressure data from multiple points simultaneously; perform spatiotemporal synchronization enhancement on the gas pressure data to obtain aligned pressure data; construct a continuous pressure field for the pipeline based on the aligned pressure data;

[0033] In this embodiment, when simultaneously collecting gas pressure data from multiple points in the gas pipeline, high-precision pressure sensors of model HPT500 are deployed. The range is 0 to 5 MPa, and the accuracy class is 0.1. A group of sensor nodes is set up every 10 meters. Each group of nodes contains three independent channels, recording the gas pressure value, ambient temperature, and measurement point timestamp, respectively. The sampling frequency is set to 10Hz. The data acquisition terminal uses the Modbus RTU protocol to synchronously transmit data to the central data server, based on the distributed time synchronization protocol PTP (Precision Time Protocol). A precise time protocol is used to uniformly correct the timestamps of all measuring points, with the error limited to within 0.01 seconds. A bidirectional interpolation correction method is used to linearly interpolate data with delays and packet loss, and outliers are removed. When data is missing for more than 1 second consecutively, a data missing marker is recorded, and no interpolation compensation is performed. Finally, aligned air pressure data under the same time reference for the entire pipeline is obtained. Based on these data, bilinear interpolation is used to interpolate and fill the spatial dimension, generating a spatial continuous air pressure field with 0.5-meter intervals. The output is a three-dimensional continuous air pressure data matrix with a spatial resolution of 0.5 meters and a time resolution of 0.1 seconds.

[0034] Step S2: Identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; estimate the transient flow velocity of the gas through the differences in pipeline pressure distribution;

[0035] In this embodiment, based on the continuous gas pressure field of the pipeline, the gas pressure values ​​at each location at the same time point are first extracted, the pressure gradient difference between two adjacent points is calculated, and the pressure difference threshold is set to 0.02 MPa. Adjacent points exceeding this threshold are identified as pressure anomaly areas, and the locations of the anomaly areas are recorded as pressure abrupt change points. Based on the location of the abrupt change point and the pressure change direction of the measuring points before and after it, the pressure change rate per unit time is calculated using the first-order difference method. Then, the transient flow velocity derivation formula is used, and the pressure change rate, pipeline cross-sectional area, and gas medium density (taken as 0.72 kg / m³) are substituted to obtain the gas transient flow velocity value. The flow velocity resolution is 0.1 m / s. The flow velocity values ​​at each location at the corresponding time are recorded to generate a gas transient flow velocity matrix.

[0036] Step S3: Continuously record the transient gas flow velocity at multiple time points and infer the flow velocity changes to obtain gas flow velocity change data; determine the pipeline inner wall deformation data through the continuous gas pressure field and gas flow velocity change data.

[0037] In this embodiment, transient gas flow velocity data is continuously recorded at 5 samples per second for 30 minutes to form a transient gas flow velocity sequence containing 18,000 time points. Based on this flow velocity sequence, the maximum, minimum, mean, and standard deviation of the time series are calculated for each sampling point. The flow velocity change rate data is obtained by using the first-order difference to calculate the flow velocity sequence. Combined with the continuous gas pressure field data of the pipeline, the local high-frequency flow velocity change area is compared and analyzed with the corresponding pressure field area based on the pressure-flow velocity coupling model. If the flow velocity fluctuation amplitude at a certain location is greater than 20% of the mean and is accompanied by pressure anomalies, the location is determined to be a deformation-affected area. Furthermore, based on the duration and intensity of the flow velocity change, the deformation value of the pipeline inner wall in this area is derived. The deformation threshold is set to 2%, and areas exceeding the threshold are marked as abnormal deformation points.

[0038] Step S4: Locate the deformation area based on the pipeline inner wall deformation data and reconstruct the deformed pipeline frame; analyze the deformation response based on the pipeline continuous gas pressure field and gas flow velocity change data, and identify the pipeline deformation type.

[0039] In this embodiment, based on the location information of abnormal deformation points, all continuous deformation points are extracted to form a deformation region. The outline of the deformation region is reconstructed using cubic spline interpolation to establish a geometric framework for the deformed pipeline. Based on this framework, continuous pressure field data and velocity change data are remapped, and the data in the original coordinate system is projected into the deformation framework to correct the spatial coordinates of each point. The pressure distribution and velocity distribution within the deformation region are recalculated. The angle between the direction of maximum deformation and the direction of pressure gradient is used as the deformation response feature. If the angle is less than 15°, it is determined to be radial collapse deformation; if the angle is greater than 75°, it is determined to be corrosion-induced wall thickness weakening deformation; and other cases are determined to be comprehensive complex deformation.

[0040] Step S5: When the pipeline deformation type is pipeline corrosion deformation data, identify the corrosion evolution trend based on the pipeline corrosion deformation data, and conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, determine the gas flow level based on the pipeline physical deformation data, and conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

[0041] In this embodiment, when the identified pipeline deformation type is corrosion deformation data, the maximum, minimum, and average inner wall thinning rate of the corrosion area are statistically analyzed. Areas with a thinning rate exceeding 25% are recorded as high-risk corrosion points, and the corrosion area, corrosion depth, and distribution density are recorded. Based on a preset corrosion safety margin limit of 30%, if the fitted value of the trend line of the corrosion area thinning rate is expected to exceed this value within the next 12 months, it is determined that there is a risk, triggering a corrosion safety warning. If the identified pipeline deformation type is physical deformation data, the deformation value, deformation length, and deformation volume change rate are statistically analyzed. Based on a preset pressure bearing threshold of 0.4 MPa, if the deformation causes the local pressure in the area to increase beyond the threshold, it is determined that there is a physical safety hazard in the area, triggering a physical deformation warning. All evaluation data are output graphically.

[0042] Preferably, step S1 includes the following steps:

[0043] Step S11: Deploy multiple pressure sensor arrays along the inside of the gas pipeline to collect gas pressure data. The number of collection points is set to 5 to 50, the gas pressure sampling frequency is set to 1 Hz to 50 Hz, and the gas pressure value is set to 0.05 MPa to 1.6 MPa.

[0044] Step S12: Timestamp the gas pressure data of the gas pipeline to obtain time-synchronized gas pressure, wherein the time synchronization accuracy is set to 0.01s~0.5s;

[0045] Step S13: Perform distributed node position correction on the gas pipeline pressure data to generate spatially mapped pressure, wherein the spatial matching distance error range is set to 0m~5m; integrate time-synchronized pressure and spatially mapped pressure to generate aligned pressure data;

[0046] Step S14: Eliminate random noise interference in the aligned air pressure data and perform multidimensional spatiotemporal interpolation to obtain complete coverage air pressure data; construct a pipeline continuous air pressure field based on the complete coverage air pressure data, with the air pressure field grid cell size set to 0.5m~5m and the continuous interpolation smoothness parameter set to 0.1~1.0.

[0047] In this embodiment, a pressure sensor array is deployed at equal intervals along the gas pipeline, with 20 measuring points. The selected sensor model is the GE UNIK5000 pressure sensor, with a range of 0MPa to 2MPa. The gas pressure sampling frequency is set to 10Hz, collecting 10 sets of gas pressure values ​​per second. The collected gas pressure data is transmitted in real-time to the edge computing unit via RS485 bus protocol. The collected gas pressure values ​​are limited to the range of 0.05MPa to 1.6MPa; values ​​below 0.05MPa and above 1.6MPa are automatically discarded and not included in the dataset. The sensor deployment length covers a 50m gas pipeline section, ensuring uniform distribution along the pipeline and a minimum of 5 and a maximum of 50 sensors. A laser rangefinder is used for on-site positioning, and the coordinate error of each sensor is controlled within ±0.5m. For each data point in the collected gas pipeline pressure data, a Unix timestamp is automatically appended to the data packet with a time accuracy of 0.01s. The timestamp is provided by a GPS synchronization module, specifically the Ublox model. The NEO-M8N module has a timing error of less than 0.01s. All data is connected to the same clock reference. During time synchronization, the timestamps in the data packets of each measuring point are extracted and uniformly converted into standard UTC format. If the error exceeds 0.5s, the data set is discarded. After synchronization, the data is arranged according to the order of the timestamps to generate the time-synchronized air pressure dataset. Based on the actual spatial coordinates of the sensors recorded during deployment, the corresponding air pressure data is bound to the spatial position. A distributed position correction method is used to map the position of each sensor point to the standard pipeline coordinate system. The spatial matching error is limited to 0m to 5m. By correcting deployment errors, pipeline curvature, and coordinate measurement errors, a three-dimensional interpolation method is used to adjust the actual position of the sensors. The pipeline centerline measured by the lidar scan is used as the reference, and the interpolation method uses inverse... The DistanceWeighting method is used to remove measurement point data with a matching error greater than 5m, with the removal rate controlled within 5%. This completes the spatially mapped barometric pressure dataset. Time-synchronized barometric pressure data and spatially mapped barometric pressure data are integrated. Based on a unified time series and spatial coordinates, the barometric pressure values ​​of all measurement points are matched to generate an aligned barometric pressure dataset. The aligned barometric pressure data uses a local extremum removal method to eliminate random noise interference. First, the standard deviation of the barometric pressure value within 5 seconds for each measurement point is calculated. Data with a standard deviation exceeding a set threshold of 0.02MPa is considered random noise, and noisy data is replaced by the average of two adjacent seconds. Then, Kriging interpolation is used for multidimensional spatiotemporal interpolation, with the interpolation interval set from 0.5m to 5m, generating complete coverage barometric pressure data. The interpolation smoothness parameter is set to 0.5, and the interpolation smoothness coefficient is controlled between 0.1 and 1.In the 0 range, by adjusting this coefficient to balance the smoothness of the interpolation surface and the fitting accuracy, a continuous gas pressure field for the gas pipeline is finally constructed based on the complete coverage of gas pressure data. The grid cell size of the pressure field is set to 1m, and the pressure field results are stored in the form of a three-dimensional matrix. The coordinate axes correspond to the spatial X, Y, and Z positions and the gas pressure values, respectively. Each grid cell in the three-dimensional matrix corresponds to a single-point gas pressure value, and the data unit is uniformly set to MPa.

[0048] Preferably, step S2 includes the following steps:

[0049] Step S21: Decompose the pressure gradient of the continuous gas pressure field in the pipeline to obtain the pipeline pressure gradient field, wherein the pressure gradient decomposition step size is set to 0.5m~3m;

[0050] Step S22: Perform local region clustering analysis on the pipeline pressure gradient field and identify pressure anomaly regions; perform physical constraint decoupling processing on the pressure anomaly regions and identify differences in pipeline pressure distribution;

[0051] Step S23: Extract the pressure deviation characteristics of the pipeline pressure distribution differences; divide the deviation pipeline sections according to the pressure deviation characteristics, wherein the deviation characteristic threshold range is set to 0.01MPa~0.3MPa, and determine the deviation change rate;

[0052] Step S24: Estimate the transient gas flow velocity based on the deviation pipeline section and the deviation change rate.

[0053] In this embodiment, the three-dimensional matrix data of the continuous pressure field of the pipeline is segmented and the difference is calculated according to the spatial coordinate order. The pressure difference is calculated based on the position of the center point of adjacent grid cells. A fixed splitting step size is adopted, with the step size range limited to 0.5m to 3m. In this embodiment, the splitting step size is set to 1m. The pressure difference between the centers of adjacent grid cells is calculated based on the splitting step size. The pressure difference between adjacent points is divided by the step size using the central difference method to obtain the pressure gradient value in the corresponding direction. The pressure change rate is calculated along the axial, radial, and vertical directions of the pipeline to generate a three-dimensional pressure gradient field. The gradient field data structure is consistent with the original continuous pressure field, the gradient unit is uniformly set to MPa / m, all data are in floating-point format, and three decimal places are retained. Based on the generated pressure gradient field, a density-based spatial clustering method is used. (Noise, DBSCAN) performs cluster analysis on pressure gradient data within a local area. The minimum number of cluster units is set to 5, the radius threshold is set to 1.5m, and the threshold for the pressure gradient difference between adjacent points is limited to 0.05MPa / m. During the clustering algorithm execution, the number of neighbors within the set radius for each point is first calculated, and then it is determined whether the required number of cluster units is met. After clustering, abnormal gradient regions are marked. Local neighborhoods are extracted from the identified abnormal regions based on the gradient spatial distribution. The neighborhood range is set to a 2m cube area around the center point. Pressure components in each direction within the abnormal region are independently decoupled, and the superimposed effects of physical factors such as pipe bending, branching, and valve disturbances are eliminated. Physical constraints are determined by the actual pipeline layout and on-site structural parameters. After confirming the absence of physical constraint interference, residual pressure distribution difference areas are identified. For the pressure distribution difference areas after physical constraint decoupling, the pressure deviation within the area is calculated based on the air pressure values ​​at each measuring point. The deviation value is determined by the difference between the pressure at that point and the mean of the neighborhood, with a deviation threshold limited. In the range of 0.01 MPa to 0.3 MPa, this embodiment sets the deviation threshold to 0.05 MPa, filters all measuring points with deviations greater than 0.05 MPa, extracts their deviation values ​​and spatial locations, and divides the pipeline into multiple deviation sections based on the spatial distribution characteristics of the deviation values. Each deviation section is at least 1 m long and no longer than the total pipeline length. The deviation change rate is calculated based on the change in deviation values ​​of adjacent measuring points. The change rate is calculated by dividing the difference in deviation between adjacent points by the distance between the two points, in MPa / m. The critical value for the deviation change rate is set to 0.03 MPa / m. Sections exceeding this value are marked as abnormal deviation zones. Based on the divided deviation pipeline sections and the corresponding deviation change rates, a one-dimensional unsteady flow theory model is used to estimate the transient flow velocity of the gas. The model input parameters include the length of the deviation section, the gas pressure values ​​at both ends, the pipeline diameter, the gas density, and the pipeline friction coefficient. The gas density is determined based on the gas composition and ambient temperature and pressure conditions. In this embodiment, the gas density is set to 0.72 kg / m³, and the pipeline diameter is set to 0.With a depth of 3m and a friction coefficient of 0.015, the gas velocity within the deviation section is estimated based on the velocity change formula. The transient gas velocity value is then output in m / s, with the estimation accuracy controlled within 5%.

[0054] Preferably, step S3, which involves continuously recording the transient flow velocity of the gas at multiple time points and inferring its flow velocity changes, includes:

[0055] The transient flow velocity of the gas is continuously recorded and the recording timestamp is marked. The continuous recording time interval is set to 0.1s~2s, and the recording time period is set to 30s~300s.

[0056] Integrate the transient gas flow rates at multiple time points into a multi-time period transient gas flow rate set;

[0057] The range of transient gas flow rate changes with concentrated transient gas flow rates over multiple time periods was selected, and the difference inference range was limited to 0.05 m / s to 5 m / s.

[0058] Inferring gas velocity variation data based on the transient velocity variation range of gas.

[0059] In this embodiment, transient gas velocity data is continuously recorded using a high-frequency data acquisition device. The data acquisition device is a differential pressure velocity sensor with a sampling frequency supporting 0.5Hz to 10Hz. The recording time interval is set within the range of 0.1s to 2s; in this embodiment, the time interval is set to 0.5s, and the recording time period is set to 120s. During the acquisition process, all transient velocity data are appended with timestamps accurate to milliseconds. The timestamp format adopts the standard YYYY-MM-DD HH:MM:SS:ms format. After continuous acquisition, all data are sorted by timestamp and organized into a two-dimensional matrix format. The first column is the timestamp, and the second column is the corresponding transient velocity value in m / s. The data format uniformly retains three decimal places. The transient velocity values ​​are derived from the fusion of the velocity estimation result from the previous stage and the real-time measurement data. The continuously recorded transient velocity data are grouped according to time periods to form multi-time-period gas transient velocity sets. Each velocity set contains 240 consecutive data points. The naming rule for the velocity sets is "segment number + start time", for example, "A01_202". "50415123000", each velocity set is arranged in chronological order, recording the corresponding velocity value and timestamp. After integrating all velocity sets for all time periods, data verification is performed to remove missing data, abnormal jump points, and duplicate timestamp data. The threshold for abnormal jump points is set to a difference greater than 20 m / s between two consecutive times. After removing all abnormal data, missing values ​​are inserted using linear interpolation. During interpolation, the median value is calculated based on the adjacent valid values ​​before and after the missing point. The intervals with significant velocity changes in the multi-time period gas transient velocity set are selected, and the difference between any two adjacent transient velocities is calculated using the difference method. The unit for values ​​is uniformly set to m / s. The filtering criteria are limited to the range of 0.05 m / s to 5 m / s. A continuous difference interval that meets this range is defined as the transient flow velocity variation interval of the gas. The length of the continuous interval is set to be no less than 1 second. If a value exceeds the range within the interval, it will be automatically truncated. After filtering, the start time, end time, maximum flow velocity, minimum flow velocity, average flow velocity, and variation range of each variation interval are recorded. All records are saved to the variation interval table. The table structure includes seven fields: "segment number, start time, end time, maximum value, minimum value, average value, and variation range". The data unit is uniformly set to m. Based on the filtered transient flow velocity change intervals of the gas, the gas flow velocity change data is inferred. The moving average method is used to calculate the flow velocity change rate within each interval. The rate is defined as the difference in flow velocity between intervals divided by the duration of the interval, with the unit being m / s². The rate values ​​of all intervals are summarized to obtain gas flow velocity change data over multiple time periods. The data format is CSV file, with fields including "segment number, change interval number, rate value, interval start time, and interval end time". All values ​​are rounded to three decimal places, and the rate value is in m / s². Finally, the inference of gas flow velocity change data over multiple time periods is completed.

[0060] Preferably, the step S3 of determining the pipe inner wall deformation data through the continuous gas pressure field and gas flow velocity change data includes:

[0061] Identify pressure fluctuations in the continuous pressure field of a pipeline;

[0062] Extract abnormal pressure fluctuations from pressure change fluctuations;

[0063] The gas flow rate change data is converted into a flow rate change curve, and the flow rate change fluctuation characteristics are identified based on the flow rate change curve.

[0064] Stress tensor reconstruction is performed based on the characteristics of flow velocity fluctuations, and changes in fluid shear properties are identified.

[0065] Analyze fluid shear distribution data based on changes in fluid shear characteristics;

[0066] Segment correlation matching was performed on the data of abnormal pressure fluctuations and fluid shear distribution, and a flow-pressure correlation mapping was established to obtain the flow-pressure coupling characteristics;

[0067] Analysis of pipe wall fluid interaction data based on flow-pressure coupling characteristics;

[0068] Based on the fluid interaction data of the pipe wall, the wall roughness is inverted to generate pipe wall roughness data;

[0069] Standard deviation ellipsoid fitting was performed on the pipe wall roughness data, and local abrupt change regions were identified to obtain the pipe inner wall deformation data.

[0070] In this embodiment, pressure data continuously collected by a high-frequency pressure sensor built into the pipeline is sampled at a frequency of 20Hz, with a recording time period of 300s and the unit of measurement being Pa. All data are arranged in chronological order to construct a one-dimensional pressure field data sequence. A five-point smoothing averaging method is used to initially filter the original pressure sequence, removing high-frequency noise interference. The filtering window size is set to 5 data points. After smoothing, the pressure sequence is used to calculate the continuous pressure change rate using the second-order difference method. Based on the alternation of positive and negative rates of change and abrupt changes in amplitude, pressure fluctuations are identified. A threshold of 200 Pa / s is set for abrupt changes in pressure change rate. Fluctuations exceeding this threshold are marked as pressure fluctuation intervals. The start and end times of the fluctuations and the maximum rate of change are recorded, generating a list of pressure fluctuation intervals. The list format includes four fields: start time, end time, maximum rate of change, and average rate of change. All fields are uniformly measured in Pa / s. Based on the identified pressure fluctuation intervals, abnormal pressure fluctuations are extracted. Abnormal fluctuations are defined as a maximum pressure change rate exceeding 300 Pa / s within a single segment or a fluctuation duration exceeding 10s. The fluctuation duration is determined by the fluctuation interval... The start and end time difference between intervals is determined. If any condition is met, it is considered an abnormal fluctuation. The abnormal fluctuation interval number, start and end time, maximum rate of change, and duration are recorded. An abnormal fluctuation data table is constructed, containing six fields: interval number, abnormal number, start time, end time, maximum rate of change, and duration. All values ​​are rounded to two decimal places, and the units are uniformly Pa / s and seconds. The previously obtained gas flow rate change data are used to plot flow rate change curves in chronological order. Missing values ​​are filled using linear interpolation with an interpolation interval of 0.5 s. The unit for the vertical axis of the curve is meters. / s, with the horizontal axis in seconds. Curve smoothing uses the moving average method, with a sliding window of 5 data points. After smoothing, the velocity curve is calculated using the difference between adjacent points, with a velocity fluctuation threshold of 0.5 m / s². Sections with velocity fluctuations exceeding the threshold are selected, and the start and end times of the fluctuations and the extreme values ​​of the change rates are recorded to complete the extraction of velocity change fluctuation features. The data table includes five fields: section number, start time, end time, maximum velocity change, and average velocity change. Based on the extracted velocity change fluctuation features, the stress tensor distribution within the pipe cross-section is reconstructed using the finite volume method. The velocity fluctuation rate data and the pipe cross-sectional area parameter are input, with the pipe cross-sectional area set to 0.A transient shear stress distribution of 1256 m² was calculated based on the Newtonian fluid hypothesis. The unit of shear stress distribution was Pa. Based on the transient shear stress distribution, sections with shear stress change rates exceeding 5 Pa / s were identified. The start and end times of shear change, the maximum shear change rate, and the average shear change rate were recorded to construct a shear change dataset. The data format included section number, start time, end time, maximum shear change rate, and average shear change rate. Based on the shear change data, the transient shear stress distribution at various locations on the pipe cross-section was analyzed. The shear stress distribution was projected onto the cross-section using polar coordinates, with a polar angle resolution of 10°. The cross-section radius was divided into equidistant rings of 0.01 m. The average shear stress value within each ring was calculated, and the θ value was recorded. The upward shear stress distribution is analyzed, and a shear stress pole figure is plotted. The transient pole figure is compared with the steady-state pole figure to identify the direction and magnitude of shear stress abrupt changes. The abrupt change criterion is that the transient shear stress at a single point exceeds 30% of the steady-state average. The location and value of the abrupt change point are recorded. A shear distribution feature table is constructed, with data fields including segment number, polar angle, radius, and shear stress value. The aforementioned abnormal pressure fluctuation data and the shear distribution feature table are compared using the Pearson correlation coefficient matching method for multi-level correlation matching. The correlation coefficient threshold is set to 0.7. The Pearson correlation coefficient between the pressure change rate sequence and the shear stress change rate sequence within the corresponding time segment is calculated. Segments that meet the threshold are extracted as flow-pressure coupling segments, and the segment number and starting point are recorded. A flow-pressure coupling characteristic table was constructed using start time, end time, and correlation coefficient. Data fields included segment number, start time, end time, and correlation coefficient. Based on this table, the fluid interaction behavior within the pipe wall was analyzed. Using the Couette flow model, the pressure gradient and shear stress distribution values ​​within the coupling characteristic segments were substituted into the model to calculate the force per unit area of ​​the pipe wall (in Pa). The force distribution corresponding to each polar angle and radius position on the calculated cross-section was also analyzed. An abnormal interaction criterion was defined as the force per unit area exceeding 80% of the pipe's rated pressure (set to 1.6 MPa). The abnormal interaction segment number, start time, end time, maximum force value, and abnormal area ratio were recorded to construct the fluid interaction data. Based on the interactive data table, to avoid directly inverting the roughness data, a residual analysis-based method is used to calculate the distribution of pressure and shear stress residuals in each section. The residual values ​​are in Pa, and a residual threshold of 200 Pa is set. Sections with residuals exceeding the threshold are screened, and the section number, start time, end time, maximum residual value, and average residual are recorded. The corresponding pipe wall roughness value is inferred from the residual variation range. The roughness value is in mm, and the criterion is that the higher the residual value, the greater the roughness. A linear fitting model is used to map the residual values ​​to roughness, and the estimated roughness value and spatial location are recorded. For the pipe wall roughness data, the standard deviation ellipsoid fitting method is applied, with the deviation threshold set at 0.2 mm and the deviation ellipsoid axis length ratio at 1:1.A 5:2 ratio was used to fit the roughness point cloud data using the least squares method. After fitting, the residual value from each point to the ellipsoidal surface was calculated in mm. Abrupt change was defined as a residual greater than 0.3 mm. The spatial coordinates of the abrupt change points, the abrupt change residual value, and the corresponding roughness value were recorded. An inner wall deformation data table was constructed, with the table structure including segment number, spatial location, roughness value, residual value, and abrupt change determination. This completed the identification of the inner wall deformation data.

[0071] Preferably, step S4, which involves locating the deformation region based on the pipe's inner wall deformation data and reconstructing the deformed pipe frame, includes:

[0072] Deformation region location by extracting deformation data of the inner wall of the pipeline;

[0073] Obtain initial gas pipeline data;

[0074] Structural topology analysis was performed on the initial gas pipeline data to obtain the initial pipeline topology.

[0075] Construct a 3D initial pipeline based on the initial pipeline topology;

[0076] Based on the location of the deformation region, a stereoscopic projection is performed on the initial three-dimensional pipeline to determine the deformation region of the three-dimensional pipeline.

[0077] Based on the deformation data of the inner wall of the pipe, the deformation shape is tracing to obtain the tracing deformation shape;

[0078] Deformation simulation of the initial three-dimensional pipeline is performed based on the tracing deformation shape and the three-dimensional pipeline deformation region to generate a deformable pipeline framework.

[0079] In this embodiment, when locating the deformation region by extracting the deformation data of the inner wall of the pipe, the pipe wall roughness data and fluid interaction data obtained from the previous steps are first called, and a roughness change threshold is set. The fluid shear stress variation threshold is 0.3 mm. With a roughness abrupt change zone and shear anomaly zone of 5 Pa, the position data of the two in the pipeline length direction are selected and spatially superimposed. The superimposed area is processed by linear interpolation and finally the coordinate range data of the continuous deformation area is output. The coordinate range is defined as the deformation interval from the start point x1 to the end point x2, in meters. The interpolation step size is 0 during the superposition process.0.01 meters. The interpolation formula is based on piecewise linear interpolation, uniformly mapping the deformation area values ​​to a three-dimensional spatial coordinate system. When obtaining initial gas pipeline data, the gas pipeline as-built archive data provided by the pipeline design institute is called. The data content includes pipeline segment number, start and end mileage, pipe diameter, wall thickness, material, layout, burial depth, weld location, elbow and flange node location. The data format is CSV file, and the fields include segment_id, start_km, end_km, diameter, wall_thickness, material, buried_depth, weld_position, elbow_position, and flange_position. The file encoding format is UTF-8. Each row of records corresponds to a standard pipeline unit. All data is imported into a PostgreSQL database and a database named pipeline_segment is created. The structured table in nt uses the same field types as the CSV file, with length units uniformly in meters and pipe diameter and wall thickness units in millimeters. When performing structural topology analysis on the initial gas pipeline data, an undirected pipeline topology graph is constructed using NetworkX (a Python graph theory library) based on the imported `pipeline_segment` table data. Pipeline segment numbers (`segment_id`) are set as nodes, and connection points such as welds, elbows, and flanges are set as edges. Edge weights are calculated based on pipeline length, with the weight unit set to meters. A depth-first search (DFS) method is used to extract the main pipeline path, identify all branch paths, and record the start and end mileage and branch positions. The topology graph result is saved as a GraphML format file for easy reference in subsequent 3D modeling. The topology data includes the number of nodes, edges, path distribution, and the length attribute of each edge. When constructing the 3D initial pipeline based on the initial pipeline topology, the 3D modeling platform ANSYS is used. SpaceClaim imports a GraphML format topology map and automatically generates the spatial coordinate relationships between nodes and edges. Based on the start and end mileage and burial depth of the nodes, they are distributed along the Z-axis. The X and Y axes are mapped to the actual geographic coordinate system according to the layout. The pipe diameter is defined based on the `diameter` field, and the wall thickness is assigned based on the `wall_thickness` field. The positions of welds, elbows, and flanges are marked in the model based on the `weld_position`, `elbow_position`, and `flange_position` fields. Adjacent nodes are connected using NURBS curve fitting to generate a continuous and smooth 3D pipe entity. When performing stereoscopic projection on the initial 3D pipe based on the deformation region location, the built-in spatial projection module of the 3D modeling platform is called. The initial 3D pipe model is used as a reference, with the range from the start point x1 to the end point x2 of the deformation interval as the baseline. The corresponding pipe segment is then orthogonally projected onto the XY, XZ, and YZ planes, with the projection accuracy set to 0.The projection results are saved as three sets of two-dimensional vector graphics in .dxf format, named top_view, front_view, and side_view respectively. The projection maps are used to analyze the contour change characteristics of the deformation area in various directions. When performing deformation shape imprinting based on the pipe inner wall deformation data, discrete points corresponding to the x1 to x2 intervals in the pipe wall roughness data and fluid shear distribution data are selected. The locations of fluid shear force abrupt changes are extracted at 0.01-meter intervals. Combined with the inner wall roughness change curve, the deformation curvature change points are determined. A continuous curve is fitted between each abrupt change point using the Cubic Spline interpolation method. The fitted curve is projected onto the aforementioned top_view, front_view, and side_view three views and superimposed to generate a deformation imprint map. The imprint shape contour is marked on the three views to form two-dimensional imprint vector data. When performing deformation simulation on the three-dimensional initial pipe based on the imprint deformation shape and the three-dimensional pipe deformation area, ANSYS Workbench is called, the three-dimensional initial pipe .stp model is imported, the aforementioned deformation imprint map is imported as boundary conditions, and the pipe material property is set to API 5L. X70 steel, with a density of 7850 kg / m³, an elastic modulus of 210 GPa, and a Poisson's ratio of 0.3, was used to map the outline of the rubbing onto the corresponding positions in the 3D model. A static analysis module was employed, applying external and internal forces distributed along the rubbing shape. External forces were based on fluid shear stress values, and internal forces were based on abnormal pressure fluctuations within the inner wall. Boundary conditions were applied with both ends fixed. An equivalent internal pressure was applied inside the pipe, set based on the average value of the gas pressure field data. Deformation simulation used tetrahedral elements with a unit size of 5 mm. The final solution yielded a 3D deformed pipe frame model, and the deformed pipe .stp file and deformation distribution field diagram were exported. Deformation values ​​were in millimeters.

[0080] Preferably, step S4 involves analyzing the deformation response of the pipeline's continuous pressure field and gas velocity variation data based on the deformable pipeline frame, and identifying the pipeline deformation type, including:

[0081] Pressure-driven mapping is performed based on the continuous gas pressure field and gas flow velocity variation data of the pipeline to obtain the pipeline pressure response distribution;

[0082] Based on the pipeline pressure response distribution, a pipeline curvature reduction mapping is performed to generate the pipeline deformation distribution;

[0083] The distribution of pipeline deformation is clustered and reorganized according to the degree of deformation to generate pipeline strain clustering characteristics;

[0084] Based on the strain accumulation characteristics of pipelines and the location of deformation response sections in deformed pipeline frames;

[0085] The differential pressure response amplitude is extracted from the continuous gas pressure field of the pipeline based on the deformation response section, and the differential pressure response amplitude is generated.

[0086] Based on the deformation response section, the velocity disturbance amplitude is extracted from the gas velocity change data to obtain the velocity disturbance amplitude.

[0087] The deformation type of the deformable pipe frame is classified by the differential pressure response amplitude and the flow velocity disturbance amplitude.

[0088] In this embodiment, when performing pressure-driven mapping based on continuous gas pressure field and gas flow velocity change data in the pipeline, a high-frequency pressure sensor array is first deployed inside the gas pipeline to acquire gas pressure field data recorded at 1-second intervals. The data covers the entire length of the pipeline, with at least 3 sampling points per meter. The gas pressure data range is set to 0.1 MPa to 0.8 MPa. Simultaneously, a gas flow velocity sensor is deployed to record the gas flow velocity change data at the same location and corresponding time points. The gas flow velocity sampling frequency is 1 Hz, and the measurement range is 0.5 m / s to 12 m / s. The aforementioned gas pressure and flow velocity data are input into the pressure response mapping unit. Based on the finite element pressure-driven mapping method, the gas pressure field data and gas flow velocity change data are mapped according to time... Synchronously corresponding to spatial location, a two-dimensional pressure mapping matrix is ​​constructed, with the pressure value at each location used as the nodal load input. The nodal location corresponds to the three-dimensional coordinates within the gas pipeline. A three-dimensional pressure-driven model is used to obtain the pressure response distribution under the corresponding three-dimensional deformed pipeline frame. When performing pipeline curvature reduction mapping based on the pipeline pressure response distribution, the spatial curvature value of each node is first calculated using the nodal coordinates of the three-dimensional deformed pipeline frame and based on the third-order B-spline curve fitting method. The distance between adjacent points of the node in the curvature calculation formula is set to 0.5m to calculate the initial curvature value. Based on the pressure response distribution, a location with pressure higher than 0.6MPa is selected, and the curvature value at this location is multiplied by a reduction factor, which is set to 0.7 to 0.Between 9, the curvature is linearly adjusted according to the pressure gradient to form a curvature reduction matrix. The reduced curvature values ​​are then mapped back to three-dimensional spatial coordinates to form a pipeline deformation distribution dataset. When clustering and reorganizing the pipeline deformation distribution according to the degree of deformation, a deformation threshold range is set based on the deformation values ​​in the distribution data. This range is divided into three levels: 0mm to 2mm, 2mm to 5mm, and above 5mm. Using a K-means clustering method, all node deformation values ​​are classified into these three levels, with a cluster center count of 3. Nodes of the same deformation level are clustered in the three-dimensional coordinate system, and the volume and maximum deformation value of each cluster are calculated to form a pipeline strain clustering feature set. The number of strain clustering blocks is set to be no less than 10 to ensure... To cover all high-deformation areas, when locating deformation response sections based on pipeline strain accumulation characteristics, the center point coordinates of each strain accumulation block are extracted based on its location coordinates. Following the pipeline length direction, the center point coordinates are projected onto the pipeline axis, and the projection length is calculated. The deformation response section length is set to 1m to the left and right of the center point projection position, forming the deformation response section range. All strain accumulation blocks are numbered, and the start and end coordinates of each section are recorded, forming a deformation response section database. The number of sections is determined based on the number of accumulation blocks, ensuring at least coverage of all strain accumulation blocks. When extracting the differential pressure response amplitude of the pipeline's continuous pressure field based on the deformation response sections, all pressure sensor data within the deformation response section are extracted, and adjacent sampling data are calculated in chronological order. The pressure difference at sampling points is calculated at 1-second intervals to form a pressure difference time series. The maximum and minimum values ​​in this time series are extracted, and their difference is calculated as the pressure difference response amplitude. The unit of the pressure difference response amplitude is set to MPa. The pressure difference response amplitude corresponding to each deformation response segment is recorded, and a pressure difference response amplitude table is constructed. The range of the pressure difference response amplitude is set to 0.01 MPa to 0.2 MPa. When extracting the velocity disturbance amplitude based on the gas flow velocity change data of the deformation response segment, the flow velocity data recorded by all flow velocity sensors within the deformation response segment is extracted. The velocity difference between adjacent sampling points is calculated in chronological order, with the unit of the velocity difference being m / s. The velocity calculation interval is 1 second, forming a velocity disturbance time series. The maximum and minimum values ​​in this time series are extracted, and their difference is calculated as... The velocity disturbance amplitude, measured in m / s, is recorded for each deformation response segment. The velocity disturbance amplitude range is set from 0.1 m / s to 2.5 m / s. When classifying the deformation type of the deformable pipe frame based on the pressure difference response amplitude and the velocity disturbance amplitude, three types of pipe deformation are defined: compression, expansion, and torsion. Judgment thresholds are set based on the pressure difference response amplitude and the velocity disturbance amplitude. The criteria for compression deformation are a pressure difference response amplitude greater than 0.12 MPa and a velocity disturbance amplitude less than 0.5 m / s; for expansion deformation, a pressure difference response amplitude less than 0.05 MPa and a velocity disturbance amplitude greater than 1.5 m / s; and for torsion deformation, a pressure difference response amplitude of 0.For pressures between 0.05 MPa and 0.12 MPa, and flow velocity disturbance amplitudes between 0.5 m / s and 1.5 m / s, the differential pressure response amplitude and flow velocity disturbance amplitude of each deformation response segment are substituted into the discrimination criteria. Based on the discrimination results, the deformation type of the corresponding deformation response segment is numbered, forming a deformation type classification result table. The deformation type numbers are set as Type-1, Type-2, and Type-3, corresponding to extrusion, expansion, and torsion types, respectively. Finally, a deformation type classification map and distribution dataset for the entire pipeline are output.

[0089] Of particular importance is the generation of pipe deformation distributions by performing pipe curvature reduction mapping based on the pipe pressure response distribution, including:

[0090] Extract the high-pressure and low-pressure points at each location in the pipeline pressure response distribution, and determine the location of abnormal pressure response based on the high-pressure and low-pressure points at each location.

[0091] Plot the pressure difference distribution curve based on the location of abnormal pressure response;

[0092] Screening abnormal pressure zones based on pressure difference distribution curves;

[0093] Based on the pressure difference distribution curve, the curvature reduction ratio of the abnormal pressure section is mapped to generate the section reduction coefficient.

[0094] The section reduction coefficients are compared with the preset pipe section reduction coefficients, and the difference coefficients are extracted to obtain the deformation curvature distribution;

[0095] The distribution of pipe deformation location is determined by the deformation curvature distribution, thereby generating the pipe deformation distribution.

[0096] In this embodiment of the invention, during the pipeline pressure response distribution data acquisition stage, high-frequency dynamic strain gauges and an embedded pressure sensor array arranged on the outer wall of the pipeline are used to collect instantaneous pressure values ​​at equally spaced locations along the pipeline length. The arrangement spacing is limited to 0.5 meters, and the pressure value recorded at each sensing point is set to a range of 0 to 10 MPa, with a recording frequency of 1000 Hz. All collected instantaneous pressure values ​​are transmitted in real time to the pressure response data analysis module via the data acquisition module. Based on the pressure threshold judgment algorithm built into this module, points with pressure values ​​greater than 8 MPa are extracted as high-pressure points, and points with pressure values ​​less than 2 MPa are extracted as low-pressure points. All high-pressure points and low-pressure points meeting the above conditions are... A pressure anomaly point set is constructed based on spatial coordinates. Using the coordinate index values ​​within this set, the locations of abnormal pressure responses are marked, forming a pressure response anomaly identification matrix. Based on these locations, the anomaly positions are sorted along the pipeline length. The pressure difference between adjacent anomaly points is calculated sequentially. All differences are plotted as pressure difference distribution curves, with the horizontal axis corresponding to the pipeline location index and the vertical axis corresponding to the pressure difference. A piecewise linear interpolation method is used to fit the curves, with an interpolation point spacing of 0.05 meters. The interpolated curves retain all local maximum and minimum values. Abnormal pressure segments are then filtered on the pressure difference distribution curves based on a set threshold: a pressure difference greater than 4 MPa. For continuous pressure sections, all intervals continuously exceeding the threshold are selected, and the start and end positions of the intervals are recorded. All intervals meeting the conditions are marked as abnormal pressure sections, and an abnormal section index table is generated. For each abnormal pressure section, the curvature change of the pressure difference curve within the section is calculated. A second-order difference discrete curvature estimation method is used, with the curvature calculation interval set to 0.05 meters. The ratio of the calculated curvature value to the maximum curvature value within the section is used as the curvature reduction ratio, denoted as the section reduction coefficient. All reduction coefficients are summarized to form a reduction coefficient distribution table. The generated section reduction coefficients are compared one by one with the preset standard pipeline section reduction coefficients. The preset reduction coefficients are obtained through finite element simulation, with the simulation condition set to an internal pipe pressure of 9 MPa. The temperature is 50 degrees Celsius. The simulated pipeline is made of Q235 material, with a diameter of 500 mm and a wall thickness of 8 mm. Based on this condition, a standard reduction factor value is obtained, with the standard factor limited to a range of 0.85 to 1. The difference between the actual reduction factor and the standard value is calculated, and all coefficients with a difference greater than 0.05 are extracted to generate a set of difference coefficients. Using the set of difference coefficients, the coordinates of the abnormal pressure section are remapped. The deformation curvature value is calibrated according to the magnitude of the difference coefficient. The larger the difference coefficient within the section, the higher the corresponding curvature value, forming a complete deformation curvature distribution map. The distribution map uses a color gradient to represent the curvature value, from 0.1 m⁻¹ to 1 m⁻¹, with the color gradient from green to red. Finally, based on the curvature value greater than 0.05 in the deformation curvature distribution map...The 6-meter mark is used as the pipe deformation location. The coordinates of all deformation locations are extracted. Using a deformation estimation method, based on the relationship that deformation equals curvature multiplied by segment length, the deformation at each point is calculated, forming a deformation distribution matrix. The matrix dimension is twice the number of coordinate points; the first column is the coordinates, and the second column is the corresponding deformation value. This matrix is ​​used to generate the pipe deformation distribution.

[0097] Of particular importance is that, based on the strain accumulation characteristics of the pipeline and the location of the deformation response zone of the deformed pipeline frame, the following are included:

[0098] Multi-segment block differential trimming is performed on the strain aggregation characteristics of the pipeline to obtain response partition differential data;

[0099] The response partition difference data is filtered by segment disturbance density to obtain data of disturbance-dense segments;

[0100] The data of the densely disturbed section were processed by measuring the pressure-flow synchronization offset to obtain the pressure-flow offset characteristic data.

[0101] The pressure-flow offset characteristic data is processed by delay amplitude mapping to obtain delay amplitude characteristic data;

[0102] Based on the delay amplitude characteristic data, the response symmetry fitting process of the deformable pipe frame is performed to obtain the symmetry offset data;

[0103] The data of densely disturbed segments and symmetric offset data are labeled with segment responses to obtain deformation response segments.

[0104] In this embodiment, when performing multi-segment differential trimming on the pipeline strain aggregation characteristics, it is necessary to divide the entire strain characteristic matrix into multiple continuous and non-overlapping equal-length blocks along the pipeline length direction using a window segmentation method with fixed segment lengths, based on the pre-obtained pipeline strain aggregation characteristic matrix. Each block is set to a length of 5 meters. After trimming, the range of strain aggregation intensity values ​​within each block is calculated. The range is the difference between the maximum and minimum values ​​within that block, and the range threshold is set to 120. (Micro-strain) Blocks with a range greater than a threshold are marked as anomalous difference blocks. The original strain aggregation feature values ​​of these blocks are retained, and blocks that do not meet the range threshold condition are pruned. Finally, a response partition difference data matrix composed of multiple anomalous difference blocks is obtained. When performing segment perturbation density filtering on the response partition difference data, the perturbation density is defined as the number of perturbation events per unit length. The criterion for perturbation events is set as the strain value change amplitude of 5 or more consecutive sampling points within a block exceeding 80. A method was used to count the number of disturbance events within a 5-meter block. If the number of disturbance events in a block exceeded 3, the block was identified as a densely disturbed area. All blocks that met the criteria for densely disturbed areas were selected and retained, while low-density blocks were removed. The final data set of densely disturbed areas consisted of four dimensions: block number, start and end positions, number of disturbance events, and maximum disturbance amplitude. When performing pressure-flow synchronization offset measurement on the data of densely disturbed areas, continuous pressure field data and gas flow velocity change data at the corresponding positions within each densely disturbed area were extracted separately. The sampling frequency was set to 10Hz, and the pressure change and flow velocity disturbance waveforms at the same time stamp were synchronized. The offset between the two sets of data was calculated using the maximum cross-correlation method. The offset is defined as the time delay value corresponding to the maximum value of the cross-correlation function of two signals, in milliseconds. The offset calculation results for all densely disturbed sections are summarized to form pressure-flow offset characteristic data. This data includes section numbers, pressure and velocity offset values, and the maximum cross-correlation value. When performing delay amplitude mapping on the pressure-flow offset characteristic data, a two-dimensional scatter plot is constructed based on the aforementioned data, using the offset as the horizontal axis and the peak amplitude of the pressure response as the vertical axis. The offset increment step is set to 10ms. The offset data is grouped by increment, and the average peak amplitude of the pressure response within each offset increment is taken to obtain the delay amplitude corresponding to each increment. Finally, the offset increment and the corresponding delay amplitude are combined to form the delay amplitude. The value characteristic data table contains three columns: offset range, average differential pressure amplitude, and section number. When performing response symmetry fitting on the deformed pipe frame based on the delay amplitude characteristic data, the offset and delay amplitude corresponding to each densely disturbed section are paired. Following the central symmetry of the section, the difference between the offset and delay amplitude at each pair of symmetrical positions is calculated. A symmetry error threshold of 20% is set. If the difference between any pair of symmetrical points exceeds this threshold, its magnitude and direction are recorded. Data on asymmetrical point pairs within all sections are summarized, and the least squares method is used to fit the distribution curve of the offset and delay amplitude difference. The fitting result is output as symmetrical offset data. The data structure includes section number, fitting residual, maximum offset position, and pair number. When labeling the data of densely disturbed sections and symmetric offset data for segment response, the labeling rules are set according to the corresponding symmetric offset data based on the number of the densely disturbed section. If the maximum offset value in the symmetric offset data within the segment is less than 10%, it is marked as a Class I response segment; if the maximum offset value is between 10% and 20%, it is marked as a Class II response segment; and if it exceeds 20%, it is marked as a Class III response segment. After completing the response label assignment for all segments, the data of densely disturbed sections and the corresponding label numbers are combined to form a deformation response segment data table. The data table structure includes six columns: segment number, number of disturbance events, pressure-flow offset, delay amplitude, symmetric offset percentage, and segment response label.

[0105] Preferably, in step S5, when the pipeline deformation type is pipeline corrosion deformation data, identifying the corrosion evolution trend based on the pipeline corrosion deformation data, and performing corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend, includes:

[0106] When the pipeline deformation type is pipeline corrosion deformation data, the pipeline corrosion deformation data is time-series tracked and the corrosion evolution process is recorded to obtain corrosion history data.

[0107] Inferring corrosion propagation trends from pipeline corrosion deformation data based on historical corrosion data;

[0108] Early warning boundaries are set based on preset corrosion safety margins;

[0109] A corrosion safety assessment is conducted based on the corrosion expansion trend according to the early warning boundary, and the level is determined based on the safety assessment data to obtain corrosion safety level data;

[0110] A multi-level early warning mechanism is constructed based on corrosion safety level data and early warning boundaries.

[0111] In this embodiment, when the pipeline deformation type is pipeline corrosion deformation data, the inner wall of the pipeline is first continuously and periodically inspected using an ultrasonic corrosion detector and a boréscope measurement device. The inspection cycle is set to 30 days. Corrosion pit depth, corrosion area, corrosion location coordinates, and corrosion type identifiers are collected. The corrosion pit depth detection resolution is set to 0.1 mm, and the corrosion area measurement range is set to 10 mm² to 5000 mm². All corrosion parameters are recorded according to the inspection timestamp, constructing a time-series corrosion evolution data table. The data table includes five parameters: inspection time, pit depth, area, location, and type. Multiple cycles of data are continuously recorded, with a cumulative recording period of no less than 12 cycles, forming corrosion history data covering one year. When inferring the corrosion expansion trend of pipeline corrosion deformation data based on the corrosion history data, linear interpolation and locally weighted regression (LOESS) are used to fit the time-varying trends of corrosion pit depth and area data. The interpolation step size is set to 1 day, and the fitting confidence interval is... The interval is set to 95%. The annual growth rate of corrosion pit depth is determined based on the slope of the fitted curve. If the absolute value of the slope is greater than 0.5 mm / year, it is recorded as a high expansion trend; otherwise, it is recorded as a low expansion trend. At the same time, the annual growth rate of corrosion area is processed in the same way. If the annual growth rate of area is greater than 400 mm² / year, it is recorded as a high expansion trend of area. Finally, the corrosion expansion trend level is determined based on the combination of pit depth and area trend. When setting the warning boundary based on the preset corrosion safety margin, the corrosion safety margin value is set as the difference between the wall thickness margin and the safe wall thickness required for the maximum operating pressure of the pipeline, according to the gas pipeline design wall thickness standard. The minimum safety margin threshold is taken as 2 mm. When the actual corrosion pit depth reaches the wall thickness minus 2 mm, it is defined as the warning boundary. At the same time, based on the corrosion area and expansion trend, an area warning value is set. When the area of ​​a single corrosion pit exceeds 2000 mm² or the annual growth rate of corrosion pit depth exceeds 0.When the corrosion rate is 5 mm / year, it is classified as an area expansion warning boundary. The warning boundary data consists of three parameters: corrosion pit number, warning pit depth, warning area, and expansion rate threshold, forming a complete warning boundary threshold library. When conducting corrosion safety assessments based on the corrosion expansion trend according to the warning boundary, the current corrosion expansion trend data is compared with the warning boundary value item by item. If the predicted pit depth trend exceeds the warning pit depth or the predicted area expansion exceeds the area warning value, it is recorded as an unsafe section. The safety assessment level is divided according to the number of exceeding items. If both exceed the standard, it is marked as Level I; if any one exceeds the standard, it is marked as Level II; if no item exceeds the standard, it is marked as Level III. Finally, the safety levels of all corrosion sections are summarized to generate a corrosion safety level data table. The data table includes the section number, pit depth exceeding the standard, and area exceeding the standard. The multi-level early warning mechanism, based on corrosion safety level data and warning boundaries, addresses situations where the corrosion rate exceeds the standard and the final level. It sets three warning trigger conditions according to the distribution of Level I, II, and III corrosion safety levels in the data: Level I triggers a red warning, Level II triggers an orange warning, and Level III triggers a yellow warning. Each warning level corresponds to a different response strategy. In red warning zones, gas transmission operations must be immediately stopped and excavation and repair arranged. In orange warning zones, high-frequency monitoring is implemented, with the monitoring cycle adjusted to 7 days. In yellow warning zones, the original 30-day monitoring cycle is maintained. The multi-level early warning mechanism generates a warning response list by mapping corrosion zone numbers to warning levels. The list includes four items: zone number, current corrosion level, recommended treatment measures, and next monitoring time, achieving graded response management of corrosion risk.

[0112] Preferably, when the pipeline deformation type in step S5 is pipeline physical deformation data, the gas flow level is determined based on the pipeline physical deformation data, and a physical safety assessment and early warning are performed based on a preset pressure tolerance threshold, including:

[0113] When the pipeline deformation type is pipeline physical deformation data, calculate the cross-sectional area change rate of the pipeline physical deformation data.

[0114] Gas flow obstruction is simulated based on the rate of change of cross-sectional area, and the degree of obstruction is evaluated based on the simulated gas flow obstruction data to generate the degree of gas flow.

[0115] The gas flow rate is compared with the preset standard flow rate, and the flow rate reduction ratio is calculated to obtain the gas flow rate impact data.

[0116] Gas volume pressure was simulated based on the gas flow rate influence data to obtain simulated gas volume pressure data;

[0117] Stress analysis was performed based on simulated gas volume pressure data and pipeline physical deformation data, and the results were quantified into gas volume pressure stress values.

[0118] Physical stress safety assessment is conducted by using preset pressure tolerance thresholds and gas volume pressure stress values.

[0119] Based on physical stress safety assessment data and pipeline physical deformation data, the coordinates of rupture risk are located;

[0120] The coordinates of the rupture risk are transmitted to a visualization page for real-time early warning.

[0121] In this embodiment, when the pipeline deformation type is physical deformation data, a laser caliper and a boréscope device are first used to perform a full-circumference scan of the internal cross-sectional profile of the gas pipeline. The scanning interval is set to 50 mm, and the caliper accuracy is not less than 0.1 mm. Based on the scanning results, the maximum and minimum diameters of each cross-section are extracted. The actual cross-sectional area is calculated using a cross-sectional fitting algorithm. The cross-sectional area calculation formula is π multiplied by the product of the maximum and minimum radii. Subsequently, the cross-sectional area change rate is calculated based on the initial design cross-sectional area and the actual cross-sectional area obtained from the detection. The change rate calculation formula is the detected cross-sectional area minus the initial design area, divided by the initial design area, and then multiplied by 100%. All detected cross-sectional data are used to form a cross-sectional area change rate distribution table, which records three parameters: cross-sectional number, detected area, design area, and change rate. This ensures that the number of detected cross-sections is not less than 200. When simulating gas flow obstruction based on the cross-sectional area change rate, the finite volume method is used based on CFD (Computational Fluid Dynamics). The Dynamics (Computational Fluid Dynamics) platform was used to numerically simulate the flow field of a gas pipeline. The gas type was set to natural gas, with a density of 0.717 kg / m³, an inlet pressure of 0.4 MPa, and an outlet pressure of 0.35 MPa. A k-ε (turbulent kinetic energy-dissipation rate model) turbulence model was used, with the rate of change of cross-sectional area applied as a boundary condition to the pipeline's axial profile model. Simulation calculations were performed, extracting velocity distribution, pressure distribution, and eddy regions. Based on the velocity drop section and turbulence intensity changes, the percentage of obstruction was calculated. The percentage of obstruction was defined as the ratio of the velocity drop to the original velocity multiplied by 100%. All calculation results were recorded in a gas flow obstruction data table, which included the cross-section number, percentage of obstruction, eddy intensity, and pressure drop. When comparing the gas flow level with a preset standard flow level, the following parameters were set... The standard flow level is set at an inlet velocity of 7 m / s, with an allowable fluctuation range of ±0.5 m / s. Based on the simulated velocity values ​​of the obstructed sections, the velocity reduction ratio is calculated point by point. The reduction ratio is calculated by subtracting the actual velocity from the standard velocity, dividing by the standard velocity, and then multiplying by 100%. A reduction ratio exceeding 10% is marked as slightly obstructed, exceeding 20% ​​as moderately obstructed, and exceeding 30% as severely obstructed. All comparison results are summarized to generate a gas flow rate impact data table, which includes the cross-section number, standard velocity, simulated velocity, reduction ratio, and obstruction level. When performing gas volume pressure simulation based on the gas flow rate impact data, the same CFD simulation model is used, with the velocity reduction ratio of each section as the input variable. The pressure rise of the corresponding section is calculated. The pressure rise is obtained by extracting the simulated pressure distribution through simulation and calculating the pressure difference between the outlet pressure and the upstream pressure of the obstructed section. If the pressure difference is greater than 0...0.03 MPa is defined as the gas volume pressure. The number of gas volume pressure sections and the total accumulated pressure are accumulated. The accumulated pressure is calculated as the sum of the pressure differences between each section, resulting in a complete simulated gas volume pressure data table. The table records the section number, pressure difference, total accumulated pressure, and length of the accumulated section. When performing stress analysis based on the simulated gas volume pressure data and pipeline physical deformation data, a pipeline wall thickness of 8 mm and an inner diameter of 600 mm are selected. The radial and circumferential stresses generated by the accumulated gas on the pipeline wall are calculated according to the Lambert-Clapeyron formula. The radial stress... Force calculation is calculated as accumulated pressure multiplied by radius divided by wall thickness. Circumferential stress calculation is calculated as accumulated pressure multiplied by inner diameter divided by twice wall thickness. All calculated values ​​are summarized to generate a gas volumetric pressure stress value table, which includes section number, accumulated pressure, radial stress, circumferential stress, and maximum stress value. When conducting a physical stress safety assessment using preset pressure tolerance thresholds and gas volumetric pressure stress values, the maximum allowable stress threshold for the pipeline is set at 240 MPa. Based on the comparison between the maximum stress value and the tolerance threshold, a warning is marked when the stress value exceeds 80% of the threshold. When the stress exceeds the threshold by more than 90%, it is marked as severe; when it exceeds the threshold by more than 100%, it is marked as extreme instability. Based on the stress conditions of each section, sections are divided into three risk levels: Level I (extreme instability), Level II (severe), and Level III (alert). A physical stress safety assessment data table is generated, containing section number, stress value, threshold ratio, and risk level. When locating rupture risk coordinates based on the physical stress safety assessment data and pipeline physical deformation data, all assessed section numbers are mapped to the pipeline's 3D coordinate model. Sections are selected as Level I or Level II based on risk level, and the 3D coordinates of the section center point are recorded as rupture risk coordinates, generating a rupture risk coordinate table containing section number, X-coordinate, Y-coordinate, Z-coordinate, and risk level. When transmitting rupture risk coordinates to the visualization page for real-time early warning, a 3D pipeline visualization platform built on Cesium (a 3D geospatial visualization platform) is used. The rupture risk coordinates are imported into the platform via GeoJSON format files, utilizing a 3D solid sphere (Primitive)... The rupture point is marked with a Sphere, with red indicating Level I risk and orange indicating Level II risk. The real-time refresh frequency is set to 10 seconds. The risk level, stress value, and coordinate location are simultaneously displayed in a pop-up window above the pipe model, triggering an audible and visual alarm. The alarm level is set according to the risk level, with the beep intensity and flashing frequency adjusted accordingly: Level I risk has a beep intensity of 90dB and a flashing frequency of 2Hz; Level II risk has a beep intensity of 70dB and a flashing frequency of 1Hz.

[0122] This invention also provides a data analysis-based gas pipeline safety assessment and early warning system for executing the data analysis-based gas pipeline safety assessment and early warning method described above. The data analysis-based gas pipeline safety assessment and early warning system includes:

[0123] The multi-point gas pressure synchronization module is used to simultaneously collect gas pressure data from multiple points in the gas pipeline; it performs spatiotemporal synchronization enhancement on the gas pipeline pressure data to obtain aligned pressure data; and it constructs a continuous pressure field for the pipeline based on the aligned pressure data.

[0124] The gas flow velocity estimation module is used to identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; and to estimate the transient gas flow velocity through the differences in pipeline pressure distribution.

[0125] The deformation response detection module is used to continuously record the transient flow velocity of gas at multiple time points and infer the flow velocity change to obtain gas flow velocity change data; the deformation data of the inner wall of the pipeline is judged by the continuous gas pressure field and gas flow velocity change data of the pipeline.

[0126] The deformation type identification module is used to locate the deformation area based on the deformation data of the inner wall of the pipeline and reconstruct the deformed pipeline frame; based on the deformed pipeline frame, it performs deformation response analysis on the continuous gas pressure field and gas flow velocity change data of the pipeline and identifies the pipeline deformation type.

[0127] The safety early warning decision module is used to identify the corrosion evolution trend based on the pipeline corrosion deformation data when the pipeline deformation type is pipeline corrosion deformation data, and to conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, it determines the gas flow level based on the pipeline physical deformation data, and to conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

[0128] This invention, through the application of a multi-point gas pressure synchronization module, achieves comprehensive acquisition and spatiotemporal synchronization of gas pipeline pressure data, ensuring data consistency and accuracy, and providing a foundation for subsequent analysis. The constructed continuous gas pressure field of the pipeline can effectively reflect the pressure changes inside the pipeline, helping to identify potential pressure anomalies. The gas flow velocity estimation module, by analyzing pressure distribution differences, can accurately estimate the transient flow velocity of the gas, providing a deeper understanding of the gas flow characteristics. The deformation response detection module, by continuously recording the transient flow velocity of the gas, can monitor flow velocity changes in real time, providing important data support for judging the deformation of the pipeline inner wall. The introduction of the deformation type identification module enables precise location of the deformation area and reconstruction of the deformed pipeline framework, laying the foundation for subsequent pressure field and flow velocity change data analysis. The safety early warning decision module functions to promptly identify the corrosion evolution trend when pipeline corrosion deformation is detected, and conduct in-depth corrosion risk assessment in combination with preset safety margins to ensure safety. By taking measures before corrosion problems escalate and avoiding safety hazards, the system can assess gas flow based on physical deformation and conduct physical safety assessments based on pressure tolerance thresholds. This effectively reduces the risk of accidents caused by physical damage to pipelines. The entire system integrates multiple data analysis technologies, enabling comprehensive monitoring and dynamic assessment of gas pipelines. This enhances the intelligence level of pipeline safety management, improves the real-time monitoring capability of pipeline status, strengthens the early warning capability for potential risks, reduces the impact of human factors on safety assessments, promotes a data-driven safety management model, provides a scientific basis for the long-term stable operation of gas pipelines, and ultimately achieves comprehensive protection for gas pipeline safety. This promotes the intelligent development of the gas industry, forms a sustainable safety assessment and early warning mechanism, provides important references for the future application and promotion of related technologies, improves the overall safety and economy of the industry, creates conditions for a safe gas environment in society, and promotes the protection and improvement of public safety.

[0129] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0130] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A data analysis-based method for safety assessment and early warning of gas pipelines, characterized in that, Includes the following steps: Step S1: Collect gas pressure data from multiple points simultaneously; perform spatiotemporal synchronization enhancement on the gas pressure data from the gas pipeline to obtain aligned gas pressure data; Construct a continuous pressure field for the pipeline based on aligned pressure data; Step S2: Identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; estimate the transient flow velocity of the gas through the differences in pipeline pressure distribution; Step S3: Continuously record the transient gas velocity at multiple time points and infer its velocity changes to obtain gas velocity change data; determine the pipeline inner wall deformation data using the continuous gas pressure field and gas velocity change data; wherein, determining the pipeline inner wall deformation data using the continuous gas pressure field and gas velocity change data includes: Identify pressure fluctuations in the continuous pressure field of a pipeline; Extract abnormal pressure fluctuations from pressure change fluctuations; The gas flow rate change data is converted into a flow rate change curve, and the flow rate change fluctuation characteristics are identified based on the flow rate change curve. Stress tensor reconstruction is performed based on the characteristics of flow velocity fluctuations, and changes in fluid shear properties are identified. Analyze fluid shear distribution data based on changes in fluid shear characteristics; Segment correlation matching was performed on the data of abnormal pressure fluctuations and fluid shear distribution, and a flow-pressure correlation mapping was established to obtain the flow-pressure coupling characteristics; Analysis of pipe wall fluid interaction data based on flow-pressure coupling characteristics; Based on the fluid interaction data of the pipe wall, the wall roughness is inverted to generate pipe wall roughness data; The pipe wall roughness data were fitted with a standard deviation ellipsoid and local abrupt change regions were identified to obtain the pipe inner wall deformation data. Step S4: Locate the deformation area based on the pipeline inner wall deformation data and reconstruct the deformed pipeline frame; analyze the deformation response based on the pipeline continuous gas pressure field and gas flow velocity change data, and identify the pipeline deformation type. Step S5: When the pipeline deformation type is pipeline corrosion deformation data, identify the corrosion evolution trend based on the pipeline corrosion deformation data, and conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, determine the gas flow level based on the pipeline physical deformation data, and conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

2. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Deploy multiple pressure sensor arrays along the inside of the gas pipeline to collect gas pressure data. The number of collection points is set to 5 to 50, the gas pressure sampling frequency is set to 1 Hz to 50 Hz, and the gas pressure value is set to 0.05 MPa to 1.6 MPa. Step S12: Timestamp the gas pressure data of the gas pipeline to obtain time-synchronized gas pressure, wherein the time synchronization accuracy is set to 0.01s~0.5s; Step S13: Perform distributed node position correction on the gas pipeline pressure data to generate spatially mapped pressure, wherein the spatial matching distance error range is set to 0m~5m; integrate time-synchronized pressure and spatially mapped pressure to generate aligned pressure data; Step S14: Eliminate random noise interference in the aligned air pressure data and perform multidimensional spatiotemporal interpolation to obtain complete coverage air pressure data; construct a pipeline continuous air pressure field based on the complete coverage air pressure data, with the air pressure field grid cell size set to 0.5m~5m and the continuous interpolation smoothness parameter set to 0.1~1.

0.

3. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Decompose the pressure gradient of the continuous gas pressure field in the pipeline to obtain the pipeline pressure gradient field, wherein the pressure gradient decomposition step size is set to 0.5m~3m; Step S22: Perform local region clustering analysis on the pipeline pressure gradient field and identify pressure anomaly regions; perform physical constraint decoupling processing on the pressure anomaly regions and identify differences in pipeline pressure distribution; Step S23: Extract the pressure deviation characteristics of the pipeline pressure distribution differences; divide the deviation pipeline sections according to the pressure deviation characteristics, wherein the deviation characteristic threshold range is set to 0.01MPa~0.3MPa, and determine the deviation change rate; Step S24: Estimate the transient gas flow velocity based on the deviation pipeline section and the deviation change rate.

4. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, Step S3 involves continuously recording the transient flow velocity of the gas at multiple time points and inferring its velocity changes, including: The transient flow velocity of the gas is continuously recorded and the recording timestamp is marked. The continuous recording time interval is set to 0.1s~2s, and the recording time period is set to 30s~300s. Integrate the transient gas flow velocities at multiple time points into a multi-time period transient gas flow velocity set; The range of transient gas flow rate changes with concentrated transient gas flow rates over multiple time periods was selected, and the difference inference range was limited to 0.05 m / s to 5 m / s. Inferring gas velocity variation data based on the transient velocity variation range of gas.

5. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, Step S4, which involves locating the deformation region based on the pipe's inner wall deformation data and reconstructing the deformed pipe framework, includes: Deformation region location by extracting deformation data of the inner wall of the pipeline; Obtain initial gas pipeline data; Structural topology analysis was performed on the initial gas pipeline data to obtain the initial pipeline topology. Construct a 3D initial pipeline based on the initial pipeline topology; Based on the location of the deformation region, a stereoscopic projection is performed on the initial three-dimensional pipeline to determine the deformation region of the three-dimensional pipeline. Based on the deformation data of the inner wall of the pipe, the deformation shape is tracing to obtain the tracing deformation shape; Deformation simulation of the initial three-dimensional pipeline is performed based on the tracing deformation shape and the three-dimensional pipeline deformation region to generate a deformable pipeline framework.

6. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, Step S4 involves analyzing the deformation response of the pipeline's continuous pressure field and gas velocity variation data based on the deformable pipeline frame, and identifying the types of pipeline deformation, including: Pressure-driven mapping is performed based on the continuous gas pressure field and gas flow velocity variation data of the pipeline to obtain the pipeline pressure response distribution; Based on the pipeline pressure response distribution, a pipeline curvature reduction mapping is performed to generate the pipeline deformation distribution; The distribution of pipeline deformation is clustered and reorganized according to the degree of deformation to generate pipeline strain clustering characteristics; Based on the strain accumulation characteristics of pipelines and the location of deformation response sections in deformed pipeline frames; The differential pressure response amplitude is extracted from the continuous gas pressure field of the pipeline based on the deformation response section, and the differential pressure response amplitude is generated. Based on the deformation response section, the velocity disturbance amplitude is extracted from the gas velocity change data to obtain the velocity disturbance amplitude. The deformation type of the deformable pipe frame is classified by the differential pressure response amplitude and the flow velocity disturbance amplitude.

7. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, In step S5, when the pipeline deformation type is pipeline corrosion deformation data, the corrosion evolution trend is identified based on the pipeline corrosion deformation data, and a corrosion safety assessment and early warning are performed based on the preset corrosion safety margin and corrosion evolution trend, including: When the pipeline deformation type is pipeline corrosion deformation data, the pipeline corrosion deformation data is time-series tracked and the corrosion evolution process is recorded to obtain corrosion history data. Inferring corrosion propagation trends from pipeline corrosion deformation data based on historical corrosion data; Early warning boundaries are set based on preset corrosion safety margins; A corrosion safety assessment is conducted based on the corrosion expansion trend according to the early warning boundary, and the level is determined based on the safety assessment data to obtain corrosion safety level data; A multi-level early warning mechanism is constructed based on corrosion safety level data and early warning boundaries.

8. The gas pipeline safety assessment and early warning method based on data analysis according to claim 1, characterized in that, When the pipeline deformation type in step S5 is pipeline physical deformation data, the gas flow level is determined based on the pipeline physical deformation data, and a physical safety assessment and early warning are performed based on a preset pressure tolerance threshold, including: When the pipeline deformation type is pipeline physical deformation data, calculate the cross-sectional area change rate of the pipeline physical deformation data. Gas flow obstruction is simulated based on the rate of change of cross-sectional area, and the degree of obstruction is evaluated based on the simulated gas flow obstruction data to generate the degree of gas flow. The gas flow rate is compared with the preset standard flow rate, and the flow rate reduction ratio is calculated to obtain the gas flow rate impact data. Gas volume pressure was simulated based on the gas flow rate influence data to obtain simulated gas volume pressure data; Stress analysis was performed based on simulated gas volume pressure data and pipeline physical deformation data, and the results were quantified into gas volume pressure stress values. Physical stress safety assessment is conducted by using preset pressure tolerance thresholds and gas volume pressure stress values. Based on physical stress safety assessment data and pipeline physical deformation data, the coordinates of rupture risk are located; The coordinates of the rupture risk are transmitted to a visualization page for real-time early warning.

9. A gas pipeline safety assessment and early warning system based on data analysis, characterized in that, For executing the data analysis-based gas pipeline safety assessment and early warning method as described in claim 1, the data analysis-based gas pipeline safety assessment and early warning system includes: The multi-point gas pressure synchronization module is used to simultaneously collect gas pressure data from multiple points in the gas pipeline; it performs spatiotemporal synchronization enhancement on the gas pipeline pressure data to obtain aligned pressure data; and it constructs a continuous pressure field for the pipeline based on the aligned pressure data. The gas flow velocity estimation module is used to identify differences in pipeline pressure distribution based on the continuous gas pressure field in the pipeline; and to estimate the transient gas flow velocity through the differences in pipeline pressure distribution. The deformation response detection module is used to continuously record the transient flow velocity of gas at multiple time points and infer the flow velocity change to obtain gas flow velocity change data; the deformation data of the inner wall of the pipeline is judged by the continuous gas pressure field and gas flow velocity change data of the pipeline. The deformation type identification module is used to locate the deformation area based on the deformation data of the inner wall of the pipeline and reconstruct the deformed pipeline frame; based on the deformed pipeline frame, it performs deformation response analysis on the continuous gas pressure field and gas flow velocity change data of the pipeline and identifies the pipeline deformation type. The safety early warning decision module is used to identify the corrosion evolution trend based on the pipeline corrosion deformation data when the pipeline deformation type is pipeline corrosion deformation data, and to conduct corrosion safety assessment and early warning based on the preset corrosion safety margin and corrosion evolution trend; when the pipeline deformation type is pipeline physical deformation data, it determines the gas flow level based on the pipeline physical deformation data, and to conduct physical safety assessment and early warning based on the preset pressure bearing threshold.

Images