Gas system physical examination evaluation and regulation method based on total factor digital twinning

Abstract

The present application relates to the field of gas system safety monitoring, and specifically discloses a gas system physical examination evaluation and regulation method based on full-element digital twinning, comprising: constructing and calibrating a pipeline wall thickness attenuation dynamic twinning model to predict gas pipeline network wall thickness distribution; using ultrasonic guided wave scanning to identify corrosion hot spot pipe sections, adaptively adjusting the monitoring node density to obtain the measured wall thickness distribution; identifying suspicious nodes through statistical outlier analysis, double verifying the deviation of the measured and predicted wall thickness, determining abnormal pipe sections and evaluating the remaining life; optimizing the operating parameters of the cathodic protection system, developing multiple regulation strategies, running in the pipeline wall thickness attenuation dynamic twinning model and predicting the corrosion inhibition effect, and outputting in descending order of effect. The present application integrates multi-source heterogeneous data and dynamic twinning model, realizes double identification and verification of corrosion anomalies and closed-loop optimization of regulation strategies, and improves the intelligent level of gas system physical examination evaluation and the initiative and reliability of corrosion prevention and control.

Description

Technical Field

[0001] This invention relates to the field of gas system safety monitoring, specifically to a gas system health check, assessment, and control method based on full-element digital twins. Background Technology

[0002] The gas system encompasses multiple stages, including gas sources, pipelines, pressure regulating stations, gas storage facilities, and end users. Among these, buried gas pipelines, as the core carrier of gas transmission, are characterized by their wide distribution, high degree of concealment, and complex operating environments, making them a critical and vulnerable link in the safe operation of the gas system. Leaks or ruptures in the pipeline network can easily trigger major safety accidents such as fires and explosions, seriously threatening public safety and urban operational order.

[0003] Among the various risks faced by gas pipeline networks, corrosion and aging are among the most significant and insidious. Corrosion is a major cause of pipe wall thinning, structural strength reduction, and even leakage. Its development is slow, gradual, and irreversible, making it difficult to detect through routine inspections and easily overlooked for extended periods. Current technologies for gas pipeline corrosion detection and assessment primarily rely on periodic manual sampling or single-sensor monitoring, which has the following shortcomings: On the one hand, existing methods often rely solely on single detection data or fixed thresholds to determine anomalies, resulting in a high false alarm rate and an inability to distinguish between actual corrosion acceleration and measurement noise or localized random fluctuations.

[0004] On the other hand, the assessment report only provides a description of the corrosion state and does not form a closed-loop linkage with the active control strategy. It also lacks a mechanism for predicting the effectiveness of the control strategy based on the prediction of the digital twin model.

[0005] Therefore, there is an urgent need for a gas system health assessment and control method that can intelligently identify abnormally corroded pipe sections and proactively output optimized control strategies. Summary of the Invention

[0006] To address the aforementioned problems, this invention proposes a gas system health check, assessment, and control method based on full-element digital twins. The specific technical solution is as follows: The gas system health check, assessment, and control method based on full-element digital twins includes the following steps: Step S1: Based on the multi-source heterogeneous data of the buried gas pipeline network, construct a dynamic twin model of pipeline wall thickness attenuation, and calibrate the dynamic twin model of pipeline wall thickness attenuation using historical corrosion detection data. Based on the calibrated dynamic twin model of pipeline wall thickness attenuation, predict the wall thickness distribution of the gas pipeline network at the current moment.

[0007] Step S2: Use ultrasonic guided wave scanning to scan the gas pipeline network, identify corrosion hotspots in the pipeline sections, adjust the spatial density of monitoring nodes along the pipeline section based on the identification results, detect the wall thickness at each monitoring node, and obtain the measured wall thickness distribution of the gas pipeline network.

[0008] Step S3: Perform statistical outlier analysis on the measured wall thickness distribution to initially identify suspicious nodes with accelerated corrosion. Verify the results by combining the deviation between the measured and predicted wall thicknesses at the suspicious nodes. Finally, determine the abnormal pipe sections with accelerated corrosion, assess their remaining lifespan, and generate a health check report.

[0009] Step S4: By optimizing the operating parameters of the forced current cathodic protection system, formulate various control strategies to mitigate the corrosion of abnormal pipe sections, test the control strategies in the dynamic twin model of pipe wall thickness attenuation and predict the corrosion suppression effect, and output each control strategy in descending order of effect.

[0010] Based on the above embodiments, the method for constructing and calibrating a dynamic twin model of pipe wall thickness attenuation includes: Collect multi-source heterogeneous data on buried gas pipeline networks, including pipeline network structural design information, dynamic operation information, and corrosive environment information.

[0011] Spatiotemporal alignment and normalization are performed on multi-source heterogeneous data to construct a unified data fusion framework, and a pipe wall thickness attenuation function with time and space as independent variables is established.

[0012] Based on the electrochemical corrosion kinetics model, structural mechanics attenuation law, and machine learning regression algorithm, a dynamic twin model of pipe wall thickness attenuation is established. The dynamic twin model of pipe wall thickness attenuation integrates the mapping relationship between corrosion rate and operating conditions and environmental parameters, and introduces time attenuation factor and spatial heterogeneity coefficient to characterize the non-uniform attenuation characteristics of wall thickness caused by local environmental differences and operating load fluctuations in different pipe sections.

[0013] By using historical corrosion detection data, the parameters of the dynamic twin model of pipe wall thickness attenuation are calibrated and the error is corrected in reverse, so that the instantaneous pipe wall thickness prediction value output by the dynamic twin model of pipe wall thickness attenuation approaches the actual detection value within a preset confidence interval.

[0014] Based on the above embodiments, the method for identifying corrosion hotspot pipe sections includes: Based on the effective propagation distance of ultrasonic guided waves, the gas pipeline network is divided into several pipe sections.

[0015] An ultrasonic guided wave excitation device is installed at the detection starting point of each pipe section, and the frequency, bandwidth and propagation mode of the ultrasonic guided wave signal along the pipe wall are set.

[0016] Along the length of the pipe section, reflected echo signals are received point by point at a set spatial sampling interval. The received echo signals are then denoised and enhanced to obtain the measured echo signals at each sampling point. The amplitude and phase of the measured echo signals are recorded.

[0017] Based on the pre-stored normal echo signals of each section of the gas pipeline network under healthy conditions, the amplitude and phase of the normal echo signals at the same locations as each sampling point are extracted.

[0018] Compare the amplitude and phase of the measured echo at each sampling point with the normal echo, and calculate the amplitude deviation and phase deviation of the echo.

[0019] If the echo amplitude or phase deviation at a certain sampling point exceeds the corresponding preset threshold, the echo at that sampling point is determined to be abnormal.

[0020] All sampling points with echo anomalies were counted, and spatial cluster analysis was performed to mark the continuous pipe sections where the echo anomaly sampling points were located as corrosion hotspots.

[0021] Based on the above embodiments, the method for adjusting the spatial density of monitoring nodes along the pipeline segment includes: The gas pipeline network is divided into corrosion hotspots and non-corrosion hotspots. Based on the preset monitoring node spacing for the corresponding corrosion hotspots and non-corrosion hotspots, the spatial density of monitoring nodes along each pipeline segment is adaptively adjusted. The monitoring node spacing for corrosion hotspots is smaller than that for non-corrosion hotspots.

[0022] Based on the above embodiments, the method for initially identifying suspicious nodes with accelerated corrosion includes: Based on the measured wall thickness distribution of the gas pipeline network, the measured wall thickness values ​​at all monitoring nodes on the same pipe section are used to construct a spatial dimension wall thickness dataset, resulting in multiple wall thickness datasets.

[0023] For each wall thickness dataset, the elements in the wall thickness dataset are sorted in ascending order of value to obtain an ordered sequence. The first quartile and the third quartile of the ordered sequence are determined, and the difference between the two is calculated as the interquartile range. Based on the first quartile and the interquartile range, combined with the preset outlier coefficient, the lower bound threshold of the low outlier is calculated.

[0024] The measured wall thickness values ​​of each monitoring node in the wall thickness dataset are compared one by one with the lower bound threshold for low outliers. Monitoring nodes with measured wall thickness values ​​less than the lower bound threshold for low outliers are marked as low outlier points.

[0025] We summarize the low wall thickness outliers in all wall thickness datasets as suspected nodes for accelerated corrosion.

[0026] Based on the above embodiments, the method for verifying suspected nodes with accelerated corrosion includes: Based on the predicted wall thickness distribution of the gas pipeline network, the predicted wall thickness at suspicious nodes is obtained.

[0027] The measured wall thickness at the suspicious node is compared with the predicted wall thickness. If the measured wall thickness of the suspicious node is lower than the predicted wall thickness, and the deviation of the wall thickness calculated from the measured wall thickness and the predicted wall thickness is greater than the preset deviation threshold, then the suspicious node is determined to be a corrosion-accelerated node. Otherwise, the suspicious node is removed from the label.

[0028] Based on the above embodiments, the method for identifying abnormal pipe sections with accelerated corrosion includes: Identify the location of all corrosion-accelerating nodes in the gas pipeline network, and mark the pipe sections with continuous corrosion-accelerating nodes and a density greater than a preset density threshold as abnormal pipe sections with accelerated corrosion.

[0029] Based on the above embodiments, the method for evaluating the remaining life of pipe sections with abnormal corrosion acceleration includes: Based on the measured wall thickness of each corrosion-accelerated node on the abnormal corrosion-accelerated pipe section, combined with the initial wall thickness and service life of the gas pipeline, the corrosion rate of each corrosion-accelerated node on the abnormal pipe section is calculated.

[0030] The maximum value among all corrosion-accelerating nodes on the abnormal pipe section is taken as the corrosion rate of the abnormal pipe section, and the minimum value among all measured wall thicknesses of corrosion-accelerating nodes is taken as the current wall thickness of the abnormal pipe section.

[0031] Based on the current wall thickness, corrosion rate, and safety threshold of the gas pipeline wall thickness of the abnormal pipe section, the remaining life of the corrosion-accelerated abnormal pipe section is calculated.

[0032] Based on the above embodiments, the methods for formulating various control strategies to mitigate abnormal pipe section corrosion include: The forced current cathodic protection system of the gas pipeline network uses multiple polarization probes and intelligent test piles installed in each pipeline section to detect the current polarization potential data of the pipeline section and extract the polarization potential data of abnormal pipeline sections.

[0033] Based on the polarization potential data of the abnormal pipe section, the polarization potential range of the abnormal pipe section is obtained. The detected polarization potential range is compared with the pre-stored protection potential safety range to determine the adjustable range of the polarization potential of the abnormal pipe section.

[0034] Within the adjustable range, select several potential values ​​according to a preset step size to obtain each candidate polarization potential.

[0035] By adjusting the output current of the potentiostat in the cathodic protection system, the polarization potential of the abnormal pipe section is shifted sequentially to each candidate polarization potential, thus obtaining various control strategies for mitigating corrosion of the abnormal pipe section.

[0036] Based on the above embodiments, the method for predicting the corrosion suppression effect of the control strategy includes: The control strategy was piloted in the dynamic twin model of pipeline wall thickness attenuation. The corrosion rate of abnormal pipe sections after control was predicted and compared with the corrosion rate of abnormal pipe sections before control to obtain the corrosion rate reduction.

[0037] Calculate the ratio of the corrosion rate reduction to the corrosion rate before regulation to determine the effectiveness of the regulation strategy in suppressing corrosion.

[0038] Compared with existing technologies, the gas system health check, assessment and control method based on full-element digital twins described in this invention has the following beneficial effects: 1. This invention identifies corrosion hotspots in pipe sections using ultrasonic guided wave scanning and strategically deploys monitoring nodes accordingly. Based on the wall thickness data detected at each monitoring node, a measured wall thickness distribution of the gas pipeline network is generated. Statistical outlier analysis is then performed on this distribution to initially identify suspected nodes with accelerated corrosion. Furthermore, a double verification is conducted by combining the deviation between the measured and predicted wall thicknesses at the suspected nodes, ultimately determining the abnormal pipe sections with accelerated corrosion. This method significantly improves the accuracy and robustness of identifying corrosion hotspots and abnormal pipe sections, effectively reducing false alarm rates and the risk of missed detections.

[0039] 2. This invention adaptively adjusts the spatial density of monitoring nodes according to the different needs of corroded and non-corroded pipe sections, achieving a differentiated node deployment strategy: denser deployment in high-risk areas and sparse deployment in low-risk areas. This strategy effectively reduces redundant detection workload and improves detection efficiency and data quality while ensuring data integrity in key areas.

[0040] 3. Based on measured polarization potential data and the safe range of protection potential in abnormal pipe sections, this invention generates multiple candidate control strategies. Each strategy is then tested in a digital twin model to predict its corrosion inhibition effect. Finally, strategies are output in descending order of effectiveness. This establishes a closed-loop linkage mechanism from initial assessment to implementation of control strategies. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0043] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the gas system health assessment and control method based on full-element digital twin proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0045] The specific scheme of the gas system health check, evaluation and control method based on full-element digital twin provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Please see Figure 1 As shown, the gas system health check and control method based on full-element digital twin provided by the present invention includes the following steps S1 to S4.

[0047] Step S1: Based on the multi-source heterogeneous data of the buried gas pipeline network, construct a dynamic twin model of pipeline wall thickness attenuation, and calibrate the dynamic twin model of pipeline wall thickness attenuation using historical corrosion detection data. Based on the calibrated dynamic twin model of pipeline wall thickness attenuation, predict the wall thickness distribution of the gas pipeline network at the current moment.

[0048] In a preferred embodiment of the present invention, the method for constructing and calibrating the dynamic twin model of pipe wall thickness attenuation in step S1 includes: Collect multi-source heterogeneous data on buried gas pipeline networks, including pipeline network structural design information, dynamic operation information, and corrosive environment information.

[0049] Spatiotemporal alignment and normalization are performed on multi-source heterogeneous data to construct a unified data fusion framework, and a pipe wall thickness attenuation function with time and space as independent variables is established.

[0050] Based on the electrochemical corrosion kinetics model, structural mechanics attenuation law, and machine learning regression algorithm, a dynamic twin model of pipe wall thickness attenuation is established. The dynamic twin model of pipe wall thickness attenuation integrates the mapping relationship between corrosion rate and operating conditions and environmental parameters, and introduces time attenuation factor and spatial heterogeneity coefficient to characterize the non-uniform attenuation characteristics of wall thickness caused by local environmental differences and operating load fluctuations in different pipe sections.

[0051] By using historical corrosion detection data, the parameters of the dynamic twin model of pipe wall thickness attenuation are calibrated and the error is corrected in reverse, so that the instantaneous pipe wall thickness prediction value output by the dynamic twin model of pipe wall thickness attenuation approaches the actual detection value within a preset confidence interval.

[0052] It should be noted that the calibrated dynamic twin model of pipe wall thickness attenuation can dynamically output the instantaneous pipe wall thickness prediction value at any point in time for any spatial point in the gas pipeline network, based on the input real-time operating conditions and environmental parameters.

[0053] It should be noted that, in order to illustrate the technical implementation path of constructing and calibrating a dynamic twin model of pipe wall thickness attenuation in this invention, the following specific embodiments will be used for explanation.

[0054] For example, in one specific embodiment, the process of constructing the dynamic twin model of pipe wall thickness attenuation is as follows: A1. Multi-source heterogeneous data acquisition: Acquiring multi-source heterogeneous data of all elements of the buried gas pipeline network, specifically including: (1) Structural design information: including pipe material (such as X52, X65 steel), initial wall thickness (unit: mm), pipe diameter, design pressure, laying period, welding method, anti-corrosion layer type and thickness, etc.

[0055] (2) Dynamic operation information: including pipeline pressure, instantaneous flow rate, gas temperature, pressure fluctuation frequency, running time, shutdown and maintenance records, etc., which are mainly obtained in real time through the SCADA system.

[0056] (3) Corrosion environment information: including soil resistivity (Ω·m), water content (%), pH value, redox potential (mV), sulfate-reducing bacteria content, stray current interference intensity (mA / m²), etc., which are obtained through on-site testing and geographic information system (GIS).

[0057] The data sources are diverse, the formats vary, and the sampling frequencies differ, so unified preprocessing is required.

[0058] A2. Data preprocessing and fusion, specifically including the following steps: A2-1, Spatiotemporal Alignment: (1) Time alignment: All dynamic data are linearly interpolated with 1 hour as the basic time unit to unify the time axis.

[0059] (2) Spatial alignment: The gas pipeline network is discretized into several spatial units of fixed length (e.g., 10 meters), each unit is assigned a unique number, and all data is mapped to the corresponding unit.

[0060] A2-2. Normalization: For each type of characteristic parameter (such as pressure, soil resistivity, etc.), a normalization method is used to scale it to the [0,1] interval to eliminate the influence of dimensions.

[0061] A2-3. Data Fusion Framework Construction: Establish a fusion data table with "spatial unit ID + timestamp" as the primary key, and store the three types of data (structure, operation, and environment) in a unified manner to form a standardized dataset that can be used for model training and prediction.

[0062] A3. Establishment of the wall thickness attenuation function: Based on the above fused data, a pipe wall thickness attenuation function is established with time and space as independent variables. Let's assume a certain spatial location of the pipe... place, time The wall thickness is Its decay process can be expressed as: .

[0063] in, For position place, time Wall thickness; For position Initial wall thickness at the location (unit: mm). For position place, time Instantaneous corrosion rate (unit: mm / year); integral term represents the rate from the initial moment. up to the current moment The cumulative wall thickness loss.

[0064] A4. Structural Design of the Dynamic Twin Model: The dynamic twin model constructed in this embodiment adopts a hybrid modeling method, integrating an electrochemical corrosion kinetic model, structural mechanical attenuation laws, and machine learning regression algorithms, as detailed below: A4-1. Electrochemical Corrosion Kinetic Model: Based on Faraday's law, establish the fundamental relationship between corrosion current density and wall thickness loss rate: .

[0065] in, For position place, time The wall thickness loss rate, The molar mass of iron, For electron transfer number, It is Faraday's constant. For the density of the pipe material, This represents the corrosion current density (calculated from soil environmental parameters).

[0066] A4-2, Structural Mechanical Attenuation Law: Introducing Stress Corrosion and Fatigue Load Correction Factors This is used to characterize the accelerating effect of pressure fluctuations inside the pipe on the corrosion rate: .

[0067] in, This represents the amplitude of pressure fluctuation. To design for maximum pressure, This is an empirical coefficient, specifically ranging from 0.2 to 0.5. It can be obtained by statistically analyzing the corrosion rate growth rate under different pressure fluctuation amplitudes based on accelerated corrosion tests of buried pipelines, and then regressing the values. value.

[0068] A4-3. Machine Learning Regression Algorithm: Random Forest or XGBoost algorithms are used, with historical corrosion detection data as labels and operating conditions and environmental parameters as features, to learn residual terms. This is used to compensate for the prediction bias of the above mechanism model.

[0069] Finally, the output expression of the dynamic twin model is: .

[0070] in, Position after compensation for prediction bias place, time The wall thickness, The spatial heterogeneity coefficient reflects the difference in corrosion rate caused by differences in soil type, anti-corrosion layer quality, local damage, etc. in different pipe sections, and its value ranges from 0.5 to 2.0.

[0071] This is the machine learning residual term, output by the regression algorithm.

[0072] This is the time decay factor, reflecting time-related effects such as aging of the anti-corrosion layer and decline in cathodic protection effectiveness. Its typical form is: .

[0073] in, and For the parameters to be calibrated, Service life (in years).

[0074] A5. Discretization of the model: To facilitate computer programming, the above continuous model is discretized: A5-1, Spatial Discreteness: Dividing the gas pipeline network into... Each spatial unit has a length of [number] units. It is 10 meters.

[0075] A5-2, Time Discreteness: Using months as the time step The total number of time steps is .

[0076] A5-3, The formula for updating the discretized wall thickness is: .

[0077] Among them, subscript Indicates the spatial unit number, superscript Indicates a time step. Indicates the first The spatial unit in the first Predicted wall thickness at each time step Indicates the next time step Predicted wall thickness.

[0078] After the dynamic twin model of pipeline wall thickness attenuation is constructed, the model parameters need to be calibrated using historical corrosion detection data to ensure prediction accuracy. In a specific embodiment, the calibration process for the dynamic twin model of pipeline wall thickness attenuation is as follows: B1. Prepare calibration data: Collect historical corrosion detection data of gas pipeline networks, including: intelligent internal detection data (such as wall thickness values ​​obtained by magnetic flux leakage detection or ultrasonic detection), ultrasonic guided wave detection data, and manual ultrasonic thickness measurement point data.

[0079] The above detection data is mapped to the corresponding spatial units and time points according to the spatiotemporal alignment rules, forming a set of "spatial unit ID, detection time, and measured wall thickness" triplets, which serve as calibration labels.

[0080] B2. Determine the parameters to be calibrated: The parameters that need to be calibrated in this model include: (1) Spatial heterogeneity coefficient (Each spatial unit is independent).

[0081] (2) Time decay factor parameter and (Globally shared).

[0082] (3) Hyperparameters and weights of machine learning models (random forest / XGBoost).

[0083] B3. Calibration Method and Procedure: Parameter calibration is performed using a method combining backpropagation and Bayesian optimization. The specific steps are as follows: B3-1. Initial parameter assignment: (Initial assumption: no spatial heterogeneity); , Machine learning models use default hyperparameters.

[0084] B3-2, Forward Propagation: Run the dynamic twin model with initial parameters and output the predicted wall thickness for each calibration data point. .

[0085] B3-3, Loss Function Calculation: The root mean square error (RMSE) is used as the loss function. .

[0086] in, To calibrate the total number of data points, To predict wall thickness for the model, This refers to the actual wall thickness measured.

[0087] B3-4. Error Reverse Correction: B3-4-1. Update using gradient descent method .

[0088] B3-4-2. Use Bayesian optimization to update the hyperparameters of the machine learning model.

[0089] B3-4-3, Machine Learning Sub-models with "Prediction Error" Using labels and operating conditions and environmental parameters as characteristics, the residual terms are retrained. .

[0090] B3-5. Iterative Convergence: Repeat the forward propagation and backward correction process until the loss function is satisfied. The value is less than a preset threshold (e.g., 0.1 mm) or the number of iterations reaches the upper limit (e.g., 500 times).

[0091] B3-6. Confidence Interval Verification: After calibration, testing was performed using a validation dataset that was not used in the training.

[0092] The model's output wall thickness prediction should have an error of no more than ±5% from the actual detected value within a 95% confidence interval.

[0093] If the requirements are met, the model calibration is complete; if not, the model structure is adjusted or the amount of calibration data is increased, and the model is recalibrated.

[0094] B4. Output and Application of Calibration Results: After calibration, the model can output: the calibrated spatial heterogeneity coefficient of each spatial unit. Time decay factor curve The current predicted wall thickness of each spatial unit and its confidence interval.

[0095] The output of the calibrated dynamic twin model has reached a high degree of consistency with the actual detection data, and can be reliably used for the prediction of gas pipeline wall thickness distribution in subsequent inspection cycles, providing accurate basic data support for subsequent abnormal pipe section identification and control strategy formulation.

[0096] Step S2: Use ultrasonic guided wave scanning to scan the gas pipeline network, identify corrosion hotspots in the pipeline sections, adjust the spatial density of monitoring nodes along the pipeline section based on the identification results, detect the wall thickness at each monitoring node, and obtain the measured wall thickness distribution of the gas pipeline network.

[0097] In a preferred embodiment of the present invention, the method for identifying corrosion hotspot pipe sections in step S2 includes: Based on the effective propagation distance of ultrasonic guided waves, the gas pipeline network is divided into several pipe sections.

[0098] An ultrasonic guided wave excitation device is installed at the detection starting point of each pipe section, and the frequency, bandwidth and propagation mode of the ultrasonic guided wave signal along the pipe wall are set.

[0099] Along the length of the pipe section, reflected echo signals are received point by point at a set spatial sampling interval. The received echo signals are then denoised and enhanced to obtain the measured echo signals at each sampling point. The amplitude and phase of the measured echo signals are recorded.

[0100] Based on the pre-stored normal echo signals of each section of the gas pipeline network under healthy conditions, the amplitude and phase of the normal echo signals at the same locations as each sampling point are extracted.

[0101] Compare the amplitude and phase of the measured echo at each sampling point with the normal echo, and calculate the amplitude deviation and phase deviation of the echo.

[0102] If the echo amplitude or phase deviation at a certain sampling point exceeds the corresponding preset threshold, the echo at that sampling point is determined to be abnormal.

[0103] All sampling points with echo anomalies were counted, and spatial cluster analysis was performed to mark the continuous pipe sections where the echo anomaly sampling points were located as corrosion hotspots.

[0104] It should be noted that the effective propagation distance of ultrasonic guided waves is determined by the guided wave mode, frequency, and pipeline characteristics. As an example, for a DN200 steel gas pipeline, using an ultrasonic guided wave in the T(0,1) mode at a frequency of 50kHz, the effective propagation distance is 50m. Based on this, the gas pipeline network is divided into several pipe sections with a length not exceeding 50m.

[0105] It should be noted that this invention employs a single excitation and full-segment acquisition method to obtain measured echo signals at each sampling point within a pipe segment. Specifically, an ultrasonic guided wave excitation device is installed at the detection starting point of each pipe segment. This device emits an ultrasonic guided wave signal once and continuously receives echo signals reflected back from the entire length of the pipe segment at the same location, thereby obtaining an original echo sequence containing complete time axis information. Subsequently, this original echo sequence is converted into a distance domain signal according to the propagation speed of the guided wave and discretized at a preset spatial sampling interval, thus obtaining the echo signal at each sampling point. It should be clarified that receiving reflected echo signals point by point along the length of the pipe segment does not refer to physically moving the ultrasonic guided wave excitation device point by point, but rather to discretizing the continuously received echo signals containing a complete time history in the digital domain at a set spatial interval according to the time-distance conversion relationship, thereby obtaining the echo data corresponding to each spatial sampling point.

[0106] It should be noted that the normal echo signal of each section of the gas pipeline network under healthy conditions can be obtained in any of the following ways: the measured echo obtained by the first scan when no obvious corrosion occurs in the early stage of pipeline operation, or the theoretical echo waveform generated by numerical simulation calculation based on the structural design parameters and material properties of the pipeline.

[0107] It should be noted that the echo amplitude deviation threshold is used to determine whether the deviation of the measured echo signal amplitude from the normal echo reference amplitude exceeds the reasonable fluctuation range. Its setting can be based on the signal-to-noise ratio of the ultrasonic guided wave detection system, the attenuation characteristics of the guided wave amplitude due to pipe wall corrosion defects, and the statistical standard deviation of historical normal echo amplitudes under healthy conditions in the same pipe section. In one specific embodiment, the echo amplitude deviation threshold is set to 15%. That is, when the measured echo amplitude is more than 15% lower than the reference amplitude, it is determined to meet the echo anomaly condition. Furthermore, the threshold can be adaptively adjusted according to the pipe wall thickness and the operating frequency of the guided wave: for example, when the pipe wall thickness is less than 6 mm, the guided wave in the thin-walled pipe is more sensitive to corrosion defects, and the amplitude attenuation response is enhanced. In this case, the threshold can be appropriately reduced (e.g., adjusted from 15.0% to 10.0%) to improve detection sensitivity; when the operating frequency of the guided wave is higher than 100 kHz, the propagation attenuation of high-frequency signals increases and it is easily affected by local geometric features. The threshold can be appropriately increased (e.g., adjusted from 15.0% to 18.0%) to suppress misjudgments caused by non-corrosive factors.

[0108] It should be noted that the echo phase deviation threshold is used to determine whether the deviation of the measured echo signal phase from the normal echo reference phase has exceeded the reasonable fluctuation range. Its setting can be based on the phase measurement accuracy of the ultrasonic guided wave detection system, the change law of guided wave phase velocity caused by corrosion, and the statistical standard deviation of historical normal echo phases under healthy conditions in the same pipe section. In a specific embodiment, the echo phase deviation threshold is set to 10°. That is, when the absolute value of the deviation between the measured echo phase and the reference phase exceeds 10°, it is determined to meet the echo anomaly condition. Furthermore, this threshold can be adaptively adjusted according to the ambient temperature and pipeline burial conditions: for example, when the ambient temperature is below 0°C, the change in the elastic modulus of the pipeline material leads to changes in the guided wave propagation characteristics and a decrease in phase stability. In this case, the threshold can be appropriately increased (e.g., adjusted from 10.0° to 12.0°) to avoid misjudgments caused by environmental factors; when the pipeline is buried in highly corrosive soil, the accelerated corrosion rate makes the phase shift more significant, and the threshold can be appropriately decreased (e.g., adjusted from 10.0° to 8.0°) to improve the detection sensitivity for early corrosion.

[0109] It should be noted that the spatial clustering analysis uses sliding window clustering or the DBSCAN algorithm to mark pipe sections with consecutive abnormal sampling points and an abnormal sampling point density exceeding a preset density threshold as corrosion hotspots.

[0110] In a preferred embodiment of the present invention, the method for adjusting the spatial density of monitoring nodes along the pipeline segment in step S2 includes: The gas pipeline network is divided into corrosion hotspots and non-corrosion hotspots. Based on the preset monitoring node spacing for the corresponding corrosion hotspots and non-corrosion hotspots, the spatial density of monitoring nodes along each pipeline segment is adaptively adjusted. The monitoring node spacing for corrosion hotspots is smaller than that for non-corrosion hotspots.

[0111] It should be noted that the spacing between monitoring nodes for corroded and non-corroded hotspot pipe sections is used to adaptively adjust the spatial density of monitoring nodes along the pipeline. This spacing can be based on the spatial distribution characteristics of corrosion defects, the positioning accuracy of ultrasonic guided wave detection, and the data resolution required for assessing the remaining life of the pipe section. In one specific embodiment, the spacing between monitoring nodes for non-corroded hotspot pipe sections is set to 5.0m, and the spacing between monitoring nodes for corroded hotspot pipe sections is set to 1.0m. That is, the monitoring node density for corroded hotspot pipe sections is five times that for non-corroded hotspot pipe sections, to achieve refined wall thickness monitoring in the accelerated corrosion zone. In addition, when a non-corrosion hotspot pipe section is adjacent to a corrosion hotspot pipe section (e.g., the distance is less than 10m), considering the possibility of corrosion spreading to the surrounding area, the monitoring node spacing of the non-corrosion hotspot pipe section can be appropriately reduced (e.g., from 5.0m to 3.0m) as a transitional monitoring area. When no corrosion hotspots are detected in the non-corrosion hotspot pipe section after two consecutive inspection cycles and the pipeline operating pressure is lower than 60% of the design pressure, the monitoring node spacing can be appropriately increased (e.g., from 5.0m to 8.0m) to reduce redundant monitoring nodes.

[0112] It should be noted that the monitoring nodes are densely deployed in the corrosion hotspot pipe sections and sparsely deployed in the non-corrosion hotspot pipe sections. This differentiated deployment method can, on the one hand, prevent the omission of key data due to too few monitoring nodes in severely corroded areas, and on the other hand, prevent the redundancy of detection resources and increased workload due to too many monitoring nodes in slightly corroded areas.

[0113] It should be noted that the method for obtaining the measured wall thickness distribution of the gas pipeline network includes: using an ultrasonic thickness gauge to detect the remaining wall thickness at each monitoring node, and taking the minimum remaining wall thickness detected at each monitoring node as the wall thickness at that monitoring node. Based on the wall thickness data at all monitoring nodes on each pipeline, the measured wall thickness distribution of the gas pipeline network is generated.

[0114] Step S3: Perform statistical outlier analysis on the measured wall thickness distribution to initially identify suspicious nodes with accelerated corrosion. Verify the results by combining the deviation between the measured and predicted wall thicknesses at the suspicious nodes. Finally, determine the abnormal pipe sections with accelerated corrosion, assess their remaining lifespan, and generate a health check report.

[0115] In a preferred embodiment of the present invention, the method for initially identifying suspicious nodes with accelerated corrosion in step S3 includes: Based on the measured wall thickness distribution of the gas pipeline network, the measured wall thickness values ​​at all monitoring nodes on the same pipe section are used to construct a spatial dimension wall thickness dataset, resulting in multiple wall thickness datasets.

[0116] For each wall thickness dataset, the elements in the wall thickness dataset are sorted in ascending order of value to obtain an ordered sequence. The first quartile and the third quartile of the ordered sequence are determined, and the difference between the two is calculated as the interquartile range. Based on the first quartile and the interquartile range, combined with the preset outlier coefficient, the lower bound threshold of the low outlier is calculated.

[0117] The measured wall thickness values ​​of each monitoring node in the wall thickness dataset are compared one by one with the lower bound threshold for low outliers. Monitoring nodes with measured wall thickness values ​​less than the lower bound threshold for low outliers are marked as low outlier points.

[0118] We summarize the low wall thickness outliers in all wall thickness datasets as suspected nodes for accelerated corrosion.

[0119] It should be noted that the outlier coefficient is used to control the sensitivity of identifying low-thickness outliers. A larger outlier coefficient results in a lower lower threshold for low outliers, and stricter outlier identification conditions, with only monitoring nodes with significantly thinner walls being marked as suspicious. Conversely, a smaller outlier coefficient results in a higher lower threshold for low outliers, and looser outlier identification conditions, allowing even monitoring nodes with slightly thinner walls to be marked as suspicious. In one specific embodiment, the outlier coefficient is set to 1.5, a value suitable for routine corrosion detection scenarios, striking a balance between identifying genuine corrosion anomalies and controlling the false alarm rate. Furthermore, the outlier coefficient can be adaptively adjusted based on the pipeline's operating years, historical corrosion rates, and the dispersion of the detection data. For example, when the dispersion of the detection data is small (e.g., the interquartile range is less than 5% of the average wall thickness) and the pipeline operating environment is stable, the outlier coefficient can be appropriately increased (e.g., from 1.5 to 2.0) to raise the identification threshold and suppress misjudgments caused by measurement errors or minor local fluctuations.

[0120] It should be noted that low wall thickness outliers indicate that the measured wall thickness at this monitoring node is statistically significantly thinner than that at other monitoring nodes on the same pipe section, and are therefore identified as suspected local corrosion anomalies.

[0121] As an example, the process for identifying outliers with low wall thickness is as follows: Taking a corrosion hotspot pipe section as an example, the section is 10m long, with 11 monitoring nodes evenly spaced along the line, numbered N1, N2, ... N11. An ultrasonic thickness gauge was used to measure the remaining wall thickness at each monitoring node. The minimum wall thickness at each node was taken as the measured wall thickness for that node, resulting in the following wall thickness dataset D (unit: mm) for the pipe section: D={7.2, 7.0, 6.8, 7.1, 5.2, 6.9, 7.0, 6.7, 7.0, 6.9, 7.1}; First, sort the elements in the dataset in ascending order of their values ​​to obtain an ordered sequence S: S={5.2, 6.7, 6.8, 6.9, 6.9, 7.0, 7.0, 7.0, 7.1, 7.1, 7.2}; The sequence has 11 elements, with position indices from 1 to 11. The first quartile, Q1, is located in the 3rd element, i.e., Q1 = 6.8 mm; the third quartile, Q3, is located in the 9th element, i.e., Q3 = 7.1 mm. The interquartile range (IQR) is calculated as IQR = Q3 − Q1 = 0.3 mm.

[0122] In this embodiment, the outlier coefficient is set to 1.5. The formula for calculating the lower outlier threshold (LB) is as follows: LB = Q1 − 1.5 × IQR = 6.35 mm; Subsequently, the measured wall thickness values ​​of each monitoring node in the wall thickness dataset D were compared one by one with the lower bound threshold of 6.35 mm for low outliers. The comparison showed that the measured wall thickness of monitoring node N5 was 5.2 mm, which is less than 6.35 mm, thus meeting the criteria for low outlier determination; the measured wall thicknesses of the remaining monitoring nodes were all not less than 6.35 mm, and therefore did not meet the criteria.

[0123] Therefore, monitoring node N5 was identified as an outlier with low wall thickness, indicating significant local thinning at this node, and was marked as a suspected node with accelerated corrosion.

[0124] In a preferred embodiment of the present invention, the method for verifying suspected nodes with accelerated corrosion in step S3 includes: Based on the predicted wall thickness distribution of the gas pipeline network, the predicted wall thickness at suspicious nodes is obtained.

[0125] The measured wall thickness at the suspicious node is compared with the predicted wall thickness. If the measured wall thickness of the suspicious node is lower than the predicted wall thickness, and the deviation of the wall thickness calculated from the measured wall thickness and the predicted wall thickness is greater than the preset deviation threshold, then the suspicious node is determined to be a corrosion-accelerated node. Otherwise, the suspicious node is removed from the label.

[0126] It should be noted that the wall thickness deviation is used to characterize the degree of relative deviation between the measured wall thickness and the predicted wall thickness. The calculation method is as follows: divide the difference between the predicted wall thickness and the measured wall thickness by the predicted wall thickness to obtain the wall thickness deviation.

[0127] It should be noted that the wall thickness deviation threshold is used to determine whether the relative deviation between the measured wall thickness and the predicted wall thickness at a suspicious node exceeds a reasonable range, thereby confirming the true corrosion acceleration node. A threshold that is too small may misjudge normal measurement fluctuations or minor local thinning as corrosion acceleration, leading to a higher false alarm rate; a threshold that is too large may miss actual corrosion acceleration nodes, reducing the reliability of the assessment. The threshold can be set based on the measurement accuracy of the ultrasonic thickness gauge, the historical statistical distribution of the annual corrosion rate of the pipeline wall, the uncertainty of the predicted wall thickness model, and the acceptable safety margin for remaining life assessment in engineering. In one specific embodiment, the wall thickness deviation threshold is set to 10%. That is, when the measured wall thickness at a suspicious node is lower than the predicted wall thickness, and the ratio of the difference to the predicted wall thickness (i.e., the wall thickness deviation) is greater than 10%, the suspicious node is determined to be a corrosion acceleration node; otherwise, its suspicious node label is removed. Furthermore, this threshold can be adaptively adjusted according to the pipeline operating pressure level, wall thickness measurement uncertainty, and pipeline remaining life requirements. For example, when the pipeline operating pressure is higher than 80% of the design pressure, it indicates that the pipeline is under high stress, and accelerated local corrosion may lead to rapid failure. In this case, the deviation threshold should be appropriately reduced (e.g., from 10% to 6%) to improve the sensitivity of identifying corrosion acceleration nodes and prevent missed detections that could lead to safety accidents. When the pipeline operating pressure is lower than 50% of the design pressure, the pipeline has a larger safety margin, and the deviation threshold can be appropriately increased (e.g., from 10% to 12%) to reduce unnecessary maintenance interventions. As another example, when the measurement uncertainty of the ultrasonic thickness gauge is large (e.g., the measurement error exceeds ±0.3 mm) or the uncertainty of the predicted wall thickness model in the area of ​​suspicious nodes is high, the deviation threshold can be appropriately increased (e.g., from 10% to 15%) to suppress misjudgments caused by measurement errors or model deviations. Conversely, when the measurement system has high accuracy and the model prediction results are reliable, the deviation threshold can be appropriately reduced (e.g., from 10% to 8%) to improve identification sensitivity.

[0128] In a preferred embodiment of the present invention, the method for determining the abnormal pipe section with accelerated corrosion in step S3 includes: Identify the location of all corrosion-accelerating nodes in the gas pipeline network, and mark the pipe sections with continuous corrosion-accelerating nodes and a density greater than a preset density threshold as abnormal pipe sections with accelerated corrosion.

[0129] It should be noted that the density of corrosion-accelerating nodes is used to characterize the number of corrosion-accelerating nodes per unit length of pipe segment. It is calculated as the ratio of the total number of corrosion-accelerating nodes in the pipe segment to the total length of the pipe segment.

[0130] It should be noted that the density threshold is used to determine whether a continuous pipe segment constitutes an abnormal segment with accelerated corrosion due to the aggregation of corrosion-accelerating nodes. A density threshold that is too small may misclassify scattered, isolated corrosion-accelerating nodes as abnormal segments, leading to over-maintenance; a density threshold that is too large may miss segments with actual aggregated accelerated corrosion, reducing the comprehensiveness of the assessment. The density threshold can be set based on the spatial clustering characteristics of corrosion-accelerating nodes, the spatial resolution of ultrasonic guided wave detection, the continuity of pipeline corrosion, and economic considerations for engineering maintenance. In a specific embodiment, the density threshold is set to 0.5 nodes / m, meaning that when consecutive corrosion-accelerating nodes appear within a pipe segment and their density is greater than 0.5 nodes / m, the pipe segment is marked as an abnormal segment with accelerated corrosion. Furthermore, this density threshold can be adaptively adjusted according to the pipe segment diameter, pipeline operating pressure, and the spatial distribution characteristics of the corrosion-accelerating nodes. For example, when the pipe section diameter is large (e.g., DN300 and above), the spacing between corrosion defects within the same length is usually large. The density threshold can be appropriately reduced (e.g., from 0.5 defects / m to 0.3 defects / m) to improve the sensitivity of identifying corrosion accumulation zones in large-diameter pipes. When the pipe section diameter is small (e.g., DN100 and below), the density threshold can be appropriately increased (e.g., from 0.5 defects / m to 0.7 defects / m) to avoid over-marking due to denser node spacing. Furthermore, when corrosion-accelerating nodes exhibit a uniform, discrete distribution rather than continuous clustering within the pipe section, even with a high node density, the lack of continuity may prevent the fulfillment of the abnormal pipe section judgment criteria. In this case, there is no need to adjust the density threshold; instead, the condition of continuous occurrence should be used as a constraint.

[0131] In a preferred embodiment of the present invention, the method for evaluating the remaining life of the corrosion-accelerated pipe section in step S3 includes: Based on the measured wall thickness of each corrosion-accelerated node on the abnormal corrosion-accelerated pipe section, combined with the initial wall thickness and service life of the gas pipeline, the corrosion rate of each corrosion-accelerated node on the abnormal pipe section is calculated.

[0132] The maximum value among all corrosion-accelerating nodes on the abnormal pipe section is taken as the corrosion rate of the abnormal pipe section, and the minimum value among all measured wall thicknesses of corrosion-accelerating nodes is taken as the current wall thickness of the abnormal pipe section.

[0133] Based on the current wall thickness, corrosion rate, and safety threshold of the gas pipeline wall thickness of the abnormal pipe section, the remaining life of the corrosion-accelerated abnormal pipe section is calculated.

[0134] It should be noted that the safe threshold for gas pipeline wall thickness is determined according to ASME B31.G standard. As an example, for a steel gas pipeline with an operating pressure of 0.4 MPa, an outer diameter of 219 mm, and material L245, the calculated safe threshold for wall thickness is 3.5 mm.

[0135] It should be noted that the method for calculating the corrosion rate is as follows: based on the difference between the measured wall thickness and the initial wall thickness, the wall thickness reduction is obtained, and the ratio of this wall thickness reduction to the service life is calculated, which is the corrosion rate.

[0136] It should be noted that the method for calculating the remaining life is as follows: based on the difference between the current wall thickness and the safe threshold for the gas pipeline wall thickness, the wall thickness reduction margin is obtained, and the ratio of this wall thickness reduction margin to the corrosion rate is calculated, which is the remaining life.

[0137] It should be noted that, based on the measured wall thickness distribution cloud map of the gas pipeline network, the abnormally corroded pipe sections and their remaining lifespan are marked, and a health check report is generated.

[0138] It should be noted that the abnormal pipe section information and its measured wall thickness data identified in this inspection are used as new historical corrosion detection data and input into the pipeline wall thickness attenuation dynamic twin model. The pipeline wall thickness attenuation dynamic twin model is incrementally calibrated and iteratively optimized to update the parameters of the pipeline wall thickness attenuation dynamic twin model for predicting the wall thickness distribution in the next inspection cycle.

[0139] Step S4: By optimizing the operating parameters of the forced current cathodic protection system, formulate various control strategies to mitigate the corrosion of abnormal pipe sections, test the control strategies in the dynamic twin model of pipe wall thickness attenuation and predict the corrosion suppression effect, and output each control strategy in descending order of effect.

[0140] In a preferred embodiment of the present invention, the method for formulating various control strategies to mitigate abnormal pipe section corrosion in step S4 includes: The forced current cathodic protection system of the gas pipeline network uses multiple polarization probes and intelligent test piles installed in each pipeline section to detect the current polarization potential data of the pipeline section and extract the polarization potential data of abnormal pipeline sections.

[0141] Based on the polarization potential data of the abnormal pipe section, the polarization potential range of the abnormal pipe section is obtained. The detected polarization potential range is compared with the pre-stored protection potential safety range to determine the adjustable range of the polarization potential of the abnormal pipe section.

[0142] Within the adjustable range, select several potential values ​​according to a preset step size to obtain each candidate polarization potential.

[0143] By adjusting the output current of the potentiostat in the cathodic protection system, the polarization potential of the abnormal pipe section is shifted sequentially to each candidate polarization potential, thus obtaining various control strategies for mitigating corrosion of the abnormal pipe section.

[0144] It should be noted that the pre-stored protection potential safety range is determined according to the national standard GB / T21448-2017 "Technical Specification for Cathodic Protection of Buried Steel Pipelines". For buried steel gas pipelines with coating protection, the protection potential safety range is -0.85V to -1.20V relative to the saturated copper / copper sulfate reference electrode (CSE). Specifically, when the potential is below -0.85V (i.e., more negative), the pipeline enters an effective protection state; when the potential is below -1.20V (i.e., too negative), it may lead to coating peeling or hydrogen embrittlement risk.

[0145] It should be noted that the method for determining the adjustable range of the polarization potential of the abnormal pipe section is as follows: compare the minimum value of the current polarization potential of the abnormal pipe section with the lower limit of the safe range of the protection potential, and take the larger one as the lower limit of the adjustable range; compare the maximum value of the current polarization potential of the abnormal pipe section with the upper limit of the safe range of the protection potential, and take the smaller one as the upper limit of the adjustable range. If the lower limit is not greater than the upper limit, there is an effective adjustable range; if the lower limit is greater than the upper limit, there is no effective adjustable range, and the overall output of the cathodic protection system should be adjusted first to bring the pipe section potential into the safe range before further adjustment.

[0146] It should be noted that the gas pipeline network in this invention is a buried steel gas trunk line protected by a coating, and a forced current cathodic protection system is pre-embedded during the design and construction phase. This system suppresses electrochemical corrosion reactions by applying cathodic current to the protected pipeline. To achieve a closed-loop parameter acquisition and control, the cathodic protection system is equipped with devices such as a potentiostat, a reference electrode, an auxiliary anode bed, a polarization probe, and an intelligent test pile. The testing device for detecting polarization potential consists of a polarization probe and an intelligent test pile: the intelligent test pile contains a potential acquisition instrument and a wireless communication module; the polarization probe, buried near the test pile, is electrically connected to the pipeline and integrates a reference electrode and a polarization test piece made of the same material as the pipeline. Through the above device, the polarization potential data of the pipeline can be acquired in real time and uploaded to a remote monitoring platform.

[0147] It should be noted that there are several ways to optimize the operating parameters of a forced current cathodic protection system, including adjusting the anode ground bed arrangement, increasing the number of auxiliary anodes, and adjusting the output parameters of the potentiostat. This method prioritizes optimizing the potentiostat output parameters as the core control strategy due to its advantages such as fast response speed, low implementation cost, continuous and fine adjustment, ease of closed-loop control, and suitability for dynamic optimization. When this method cannot meet the adjustment requirements, physical optimization methods such as anode ground bed modification are used as supplementary measures.

[0148] In a preferred embodiment of the present invention, the method for predicting the corrosion suppression effect of the control strategy in step S4 includes: The control strategy was piloted in the dynamic twin model of pipeline wall thickness attenuation. The corrosion rate of abnormal pipe sections after control was predicted and compared with the corrosion rate of abnormal pipe sections before control to obtain the corrosion rate reduction.

[0149] Calculate the ratio of the corrosion rate reduction to the corrosion rate before regulation to determine the effectiveness of the regulation strategy in suppressing corrosion.

[0150] In one embodiment of the present invention, after predicting the corrosion suppression effect of each control strategy, it is also necessary to comprehensively evaluate the negative effects of each strategy, including but not limited to coating peeling risk, hydrogen embrittlement sensitivity, and energy consumption cost. Based on this, the candidate polarization potential that optimizes the overall benefit index is selected as the best control strategy and fed back to the cathodic protection system for execution.

[0151] In summary, this invention predicts the wall thickness distribution of gas pipeline networks by constructing and calibrating a dynamic twin model of pipeline wall thickness attenuation; it uses ultrasonic guided wave scanning to identify corrosion hotspots in pipe sections, adaptively adjusts the monitoring node density, and obtains the measured wall thickness distribution; it identifies suspicious nodes through statistical outlier analysis, and performs dual verification by combining measured and predicted wall thickness deviations to determine abnormal pipe sections and assess their remaining lifespan; it optimizes the operating parameters of the cathodic protection system, formulates multiple control strategies, and conducts trial runs in the dynamic twin model of pipeline wall thickness attenuation to predict the corrosion suppression effect, outputting the results in descending order of effectiveness. This invention integrates multi-source heterogeneous data with a dynamic twin model to achieve dual identification and verification of corrosion anomalies and closed-loop optimization of control strategies, significantly improving the intelligence level of gas system health assessment and the initiative and reliability of corrosion prevention and control.

[0152] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0153] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0154] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. The focus of each embodiment is on its differences from other embodiments. In particular, the apparatus embodiments are described simply because they are fundamentally based on the method embodiments; relevant details can be found in the descriptions of the method embodiments.

[0155] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0156] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0157] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0158] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0159] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0160] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0161] In conclusion, the above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for the health check, assessment, and control of a gas system based on full-element digital twins, characterized in that: Includes the following steps: S1. Based on multi-source heterogeneous data of buried gas pipeline network, construct a dynamic twin model of pipeline wall thickness attenuation, and use historical corrosion detection data to calibrate the dynamic twin model of pipeline wall thickness attenuation. Based on the calibrated dynamic twin model of pipeline wall thickness attenuation, predict the wall thickness distribution of gas pipeline network at the current moment. S2. Use ultrasonic guided wave scanning to scan the gas pipeline network, identify corrosion hotspots in the pipeline sections, adjust the spatial density of monitoring nodes along the pipeline section based on the identification results, detect the wall thickness at each monitoring node, and obtain the measured wall thickness distribution of the gas pipeline network. S3. Perform statistical outlier analysis on the measured wall thickness distribution to preliminarily identify suspicious nodes with accelerated corrosion. Verify the results by combining the deviation between the measured wall thickness and the predicted wall thickness at the suspicious nodes. Finally, determine the abnormal pipe section with accelerated corrosion and assess its remaining life, and generate a health check report. S4. By optimizing the operating parameters of the forced current cathodic protection system, formulate various control strategies to mitigate the corrosion of abnormal pipe sections, test the control strategies in the dynamic twin model of pipe wall thickness attenuation and predict the corrosion inhibition effect, and output each control strategy in descending order of effect.

2. The gas system health check, assessment, and control method based on full-element digital twins as described in claim 1, characterized in that, Methods for constructing and calibrating a dynamic twin model of pipe wall thickness attenuation include: Collect multi-source heterogeneous data on buried gas pipeline networks, including pipeline network structural design information, dynamic operation information, and corrosive environment information; Spatiotemporal alignment and normalization of multi-source heterogeneous data are performed to construct a unified data fusion framework and establish a pipe wall thickness attenuation function with time and space as independent variables. Based on the electrochemical corrosion kinetics model, structural mechanics attenuation law and machine learning regression algorithm, a dynamic twin model of pipe wall thickness attenuation is established. The dynamic twin model of pipe wall thickness attenuation integrates the mapping relationship between corrosion rate and operating conditions and environmental parameters, and introduces time attenuation factor and spatial heterogeneity coefficient to characterize the non-uniform attenuation characteristics of wall thickness caused by local environmental differences and operating load fluctuations in different pipe sections. By using historical corrosion detection data, the parameters of the dynamic twin model of pipe wall thickness attenuation are calibrated and the error is corrected in reverse, so that the instantaneous pipe wall thickness prediction value output by the dynamic twin model of pipe wall thickness attenuation approaches the actual detection value within a preset confidence interval.

3. The gas system health check, assessment, and control method based on full-element digital twins as described in claim 1, characterized in that, Methods for identifying corrosion hotspots in pipe sections include: Based on the effective propagation distance of ultrasonic guided waves, the gas pipeline network is divided into several pipe sections; An ultrasonic guided wave excitation device is set at the detection starting point of each pipe section, and the frequency, bandwidth and propagation mode of the ultrasonic guided wave signal along the pipe wall are set. Along the length of the pipe section, the reflected echo signal is received point by point at a set spatial sampling interval, and the received echo signal is denoised and enhanced to obtain the measured echo signal at each sampling point. The amplitude and phase of the measured echo signal are recorded. Based on the pre-stored normal echo signals of each section of the gas pipeline network under healthy conditions, extract the amplitude and phase of the normal echo signals at the same locations as each sampling point. Compare the amplitude and phase of the measured echo at each sampling point with the normal echo, and calculate the amplitude deviation and phase deviation of the echo. If the echo amplitude or phase deviation at a certain sampling point exceeds the corresponding preset threshold, the echo at that sampling point is determined to be abnormal. All sampling points with echo anomalies were counted, and spatial cluster analysis was performed to mark the continuous pipe sections where the echo anomaly sampling points were located as corrosion hotspots.

4. The gas system health check, assessment, and control method based on full-element digital twin as described in claim 1, characterized in that, Methods for adjusting the spatial density of monitoring nodes along the pipeline section include: The gas pipeline network is divided into corrosion hotspots and non-corrosion hotspots. Based on the preset monitoring node spacing for the corresponding corrosion hotspots and non-corrosion hotspots, the spatial density of monitoring nodes along each pipeline segment is adaptively adjusted. The monitoring node spacing for corrosion hotspots is smaller than that for non-corrosion hotspots.

5. The gas system health check, assessment, and control method based on full-element digital twin as described in claim 1, characterized in that, Methods for initially identifying suspicious nodes with accelerated corrosion include: Based on the measured wall thickness distribution of the gas pipeline network, the measured wall thickness values ​​at all monitoring nodes on the same pipe section are used to construct a spatial dimension wall thickness dataset, resulting in multiple wall thickness datasets. For each wall thickness dataset, the elements in the wall thickness dataset are sorted in ascending order of value to obtain an ordered sequence. The first quartile and the third quartile of the ordered sequence are determined, and the difference between the two is calculated as the interquartile range. Based on the first quartile and the interquartile range, combined with the preset outlier coefficient, the lower bound threshold of the low outlier is calculated. The measured wall thickness values ​​of each monitoring node in the wall thickness dataset are compared one by one with the lower bound threshold for low outliers. Monitoring nodes with measured wall thickness values ​​less than the lower bound threshold for low outliers are marked as low outlier points. We summarize the low wall thickness outliers in all wall thickness datasets as suspected nodes for accelerated corrosion.

6. The gas system health check, assessment, and control method based on full-element digital twin as described in claim 1, characterized in that, Methods for verifying suspected nodes with accelerated corrosion include: Based on the predicted wall thickness distribution of the gas pipeline network, the predicted wall thickness at suspicious nodes is obtained; The measured wall thickness at the suspicious node is compared with the predicted wall thickness. If the measured wall thickness of the suspicious node is lower than the predicted wall thickness, and the deviation of the wall thickness calculated from the measured wall thickness and the predicted wall thickness is greater than the preset deviation threshold, then the suspicious node is determined to be a corrosion-accelerated node. Otherwise, the suspicious node is removed from the label.

7. The gas system health check, assessment, and control method based on full-element digital twin as described in claim 6, characterized in that, Methods for identifying abnormal pipe sections with accelerated corrosion include: Identify the location of all corrosion-accelerating nodes in the gas pipeline network, and mark the pipe sections with continuous corrosion-accelerating nodes and a density greater than a preset density threshold as abnormal pipe sections with accelerated corrosion.

8. The gas system health check, assessment, and control method based on full-element digital twins according to claim 7, characterized in that, Methods for assessing the remaining life of pipe sections with accelerated corrosion include: Based on the measured wall thickness of each corrosion-accelerated node on the abnormal corrosion-accelerated pipe section, combined with the initial wall thickness and service life of the gas pipeline, the corrosion rate of each corrosion-accelerated node on the abnormal pipe section is calculated. The maximum value among all corrosion-accelerating nodes on the abnormal pipe section is taken as the corrosion rate of the abnormal pipe section, and the minimum value among all measured wall thicknesses of corrosion-accelerating nodes is taken as the current wall thickness of the abnormal pipe section. Based on the current wall thickness, corrosion rate, and safety threshold of the gas pipeline wall thickness of the abnormal pipe section, the remaining life of the corrosion-accelerated abnormal pipe section is calculated.

9. The gas system health check, assessment, and control method based on full-element digital twins according to claim 1, characterized in that, Methods for developing various control strategies to mitigate corrosion in abnormal pipe sections include: The forced current cathodic protection system of the gas pipeline network uses multiple polarization probes and intelligent test piles installed in each pipeline section to detect the current polarization potential data of the pipeline section and extract the polarization potential data of abnormal pipeline sections. Based on the polarization potential data of the abnormal pipe section, the polarization potential range of the abnormal pipe section is obtained. The detected polarization potential range is compared with the pre-stored protection potential safety range to determine the adjustable range of the polarization potential of the abnormal pipe section. Within the adjustable range, select several potential values ​​according to a preset step size to obtain each candidate polarization potential; By adjusting the output current of the potentiostat in the cathodic protection system, the polarization potential of the abnormal pipe section is shifted sequentially to each candidate polarization potential, thus obtaining various control strategies for mitigating corrosion of the abnormal pipe section.

10. The gas system health check, assessment, and control method based on full-element digital twin as described in claim 1, characterized in that, Methods for predicting and controlling the effectiveness of corrosion suppression strategies include: The control strategy was piloted in the dynamic twin model of pipe wall thickness attenuation to predict the corrosion rate of abnormal pipe sections after control and compare it with the corrosion rate of abnormal pipe sections before control to obtain the corrosion rate reduction. Calculate the ratio of the corrosion rate reduction to the corrosion rate before regulation to determine the effectiveness of the regulation strategy in suppressing corrosion.

Images