Oil and gas pipeline corrosion evaluation method and system based on potential detection

Abstract

The present application relates to the field of oil and gas pipeline corrosion detection and evaluation, a kind of oil and gas pipeline corrosion evaluation method and system based on potential detection, comprising: confirming multiple potential detection points on oil and gas pipeline, based on multiple original potential node time sequence, obtain multiple calibration potential sequence, execute potential gradient analysis operation, obtain potential gradient sequence, carry out missing data interpolation processing to potential gradient sequence, obtain potential gradient distribution sequence set, carry out abnormal screening, obtain comprehensive potential gradient set, utilize comprehensive potential gradient set in oil and gas pipeline to confirm high-risk pipeline section, based on high-frequency detection point sequence, obtain multiple detection parameter node sequence, utilize corrosion rate correlation model to obtain steady-state corrosion rate sequence, carry out corrosion risk level division to high-risk pipeline section in oil and gas pipeline, obtain oil and gas pipeline corrosion risk evaluation report.The present application can solve the problem of isolated analysis of detection parameters and poor early corrosion warning capability in existing oil and gas pipeline corrosion evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of oil and gas pipeline corrosion detection and evaluation, and particularly relates to an oil and gas pipeline corrosion evaluation method and system based on potential detection. BACKGROUND

[0002] With the continuous expansion of oil and gas pipeline transportation scale, the pipeline corrosion problem is increasingly prominent, which poses a serious threat to the safe operation of the pipeline. At present, pipeline corrosion monitoring mainly relies on potential detection technology, which evaluates the corrosion risk by measuring the potential distribution of the pipeline.

[0003] The traditional corrosion evaluation method is usually based on static potential data or periodic manual detection, and lacks comprehensive analysis of multiple source parameters. Therefore, optimizing the oil and gas pipeline corrosion evaluation method is of great significance to improve the accuracy, real-time performance and reliability of pipeline corrosion monitoring. SUMMARY

[0004] The present application provides an oil and gas pipeline corrosion evaluation method based on potential detection and a computer readable storage medium, which mainly aims to solve the problems of isolated analysis of detection parameters and poor early corrosion warning capability in existing oil and gas pipeline corrosion evaluation.

[0005] To achieve the above purpose, the present application provides an oil and gas pipeline corrosion evaluation method based on potential detection, which comprises:

[0006] A plurality of potential detection points are identified on the pre-identified oil and gas pipeline. The pipeline potential, soil potential and environmental parameters are collected at the plurality of potential detection points using a pre-set collection time interval to obtain a plurality of original potential node time sequences. The original potential node includes pipeline potential, soil potential and environmental parameters, and the original potential node time sequence corresponds to the potential detection point one by one.

[0007] A plurality of calibration potential sequences are obtained based on the plurality of original potential node time sequences.

[0008] The following operations are performed on each calibration potential sequence in the plurality of calibration potential sequences:

[0009] The calibration potential sequence is subjected to a potential gradient analysis operation to obtain a potential gradient sequence.

[0010] The potential gradient sequence is subjected to missing data interpolation processing to obtain a potential gradient distribution sequence.

[0011] The potential gradient distribution sequences are summarized to obtain a potential gradient distribution sequence set.

[0012] The plurality of potential gradient distribution sequences in the potential gradient distribution sequence set are subjected to abnormal screening to obtain a comprehensive potential gradient set.

[0013] High-risk pipeline segments in oil and gas pipelines were identified using a comprehensive potential gradient set.

[0014] A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0015] A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods;

[0016] A steady-state corrosion rate sequence is obtained by using multiple detection parameter node sequences and a corrosion rate correlation model. Based on the preset grading threshold and the steady-state corrosion rate sequence, the corrosion risk level of high-risk pipeline sections in oil and gas pipelines is classified, resulting in an oil and gas pipeline corrosion risk assessment report.

[0017] Optionally, obtaining multiple calibration potential sequences based on the timing of multiple original potential nodes includes:

[0018] For each of the multiple original potential node timing sequences, the following operation is performed:

[0019] A three-dimensional correlation matrix is ​​constructed using the original potential node timing, and the noise coupling coefficient matrix is ​​obtained based on the three-dimensional correlation matrix.

[0020] A noise prediction model is constructed using a pre-built time series analysis method and a noise coupling coefficient matrix, and a noise prediction sequence is obtained using the noise prediction model.

[0021] Based on the original potential node timing sequence and noise prediction sequence, the pipeline potential at each potential detection point is filtered to obtain the filtered potential sequence.

[0022] By summarizing the filtered potential sequences, multiple filtered potential sequences are obtained, in which each filtered potential sequence corresponds one-to-one with the timing sequence of the original potential nodes.

[0023] Based on the time of the original potential nodes, multiple filtered potential sequences are reconstructed to obtain multiple calibration potential sequences.

[0024] Optionally, performing a potential gradient analysis operation on the calibration potential sequence to obtain a potential gradient sequence includes:

[0025] A first potential detection point is identified in the calibration potential sequence. Based on the first potential detection point, a second potential detection point is identified in the calibration potential sequence. The second potential detection point is adjacent to the first potential detection point and is placed after the first potential detection point.

[0026] Extract the first potential value and the second potential value corresponding to the first potential detection point and the second potential detection point from the calibration potential sequence;

[0027] The potential gradient is obtained using the first potential value and the second potential value;

[0028] Using the second potential detection point as the first potential detection point, and returning to the step of confirming the second potential detection point in the calibration potential sequence based on the first potential detection point, the potential gradients are summarized to obtain the potential gradient sequence.

[0029] Optionally, the step of interpolating missing data in the potential gradient sequence to obtain the potential gradient distribution sequence includes:

[0030] The integrity of the potential gradient sequence is checked to obtain the missing tag vector set;

[0031] Once it is confirmed that the set of missing label vectors includes at least one missing label vector, the following operation is performed on each missing label vector in the set of missing label vectors:

[0032] The missing label vector and the potential gradient sequence are input into a pre-constructed missing data labeling unit to obtain a labeled potential gradient matrix;

[0033] A local feature matrix is ​​constructed using the environmental parameters of each potential detection point in the time series of multiple original potential nodes;

[0034] The labeled potential gradient matrix and local feature matrix are input into the pre-constructed interpolation prediction model to obtain the predicted potential gradient value set, which includes multiple predicted potential gradient values, and the predicted potential gradient values ​​correspond one-to-one with the missing label vector.

[0035] The potential gradient matrix is ​​filled with the predicted potential gradient values ​​from the predicted potential gradient value set and the missing label vectors from the missing label vector set to obtain the potential gradient distribution sequence.

[0036] Optionally, the anomaly screening of multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set includes:

[0037] Based on the potential gradients from multiple potential gradient distribution sequences obtained by summarizing the potential gradients at potential detection points, multiple categorical potential gradient sequences are obtained. The following operation is performed on each of the multiple categorical potential gradient sequences:

[0038] The classification potential gradients are extracted sequentially from the classification potential gradient sequence, and the following operations are performed on the extracted classification potential gradients:

[0039] The extracted classification potential gradients are removed from the classification potential gradient sequence to obtain a reference classification potential gradient set. The potential gradient deviation value is calculated using a pre-constructed deviation formula, a pre-constructed potential gradient deviation function, the extracted classification potential gradients, and the reference classification potential gradient set. The deviation formula is shown below:

[0040]

[0041] Where D represents the potential gradient deviation, M j Let f(t0,t) represent the j-th reference classification potential gradient in the reference classification potential gradient set, M0 represent the extracted classification potential gradient, and f(t0,t) represent the j-th reference classification potential gradient in the reference classification potential gradient set. j ) represents the prediction bias of the potential gradient predicted using the potential gradient bias function, and t0 represents the time corresponding to the extracted classification potential gradient. j This represents the time corresponding to the j-th reference classification potential gradient in the reference classification potential gradient set, and k represents the total number of k reference classification potential gradients in the reference classification potential gradient set.

[0042] The potential gradient deviation value is compared with a preset potential gradient deviation threshold. If the potential gradient deviation value is greater than or equal to the potential gradient deviation threshold, the extracted potential gradient is removed from the classification potential gradient sequence to obtain an updated classification potential gradient sequence.

[0043] The updated classification potential gradient sequences are summarized to obtain multiple updated classification potential gradient sequences. Using a pre-constructed reference mapping sequence, the multiple updated classification potential gradient sequences are mapped to the reference mapping sequence to obtain multiple mapped potential gradient sequences.

[0044] Extract the mapped potential gradient sequence sequentially from multiple mapped potential gradient sequences, and perform the following operations on the extracted mapped potential gradient sequences:

[0045] Searching is performed in the extracted mapping potential gradient sequence. If a preset missing value is found in the mapping potential gradient sequence, multiple reference potential gradients are extracted from multiple mapping potential gradient sequences using the missing value and the position corresponding to the extracted mapping potential gradient sequence.

[0046] Calculate the mean of multiple reference potential gradients to obtain the reference potential mean. Use the reference potential mean to replace the missing value to obtain the target potential gradient sequence.

[0047] By summing the target potential gradient sequences, multiple target potential gradient sequences are obtained;

[0048] A comprehensive potential gradient set is obtained from multiple target potential gradient sequences.

[0049] Optionally, the identification of high-risk pipeline segments in oil and gas pipelines using a comprehensive potential gradient set includes:

[0050] Based on the multiple potential detection points, multiple gradient intervals are identified in the comprehensive potential gradient set, and the number of gradient intervals plus one equals the number of multiple potential detection points.

[0051] The potential gradient rate of change for each gradient interval is obtained by using multiple gradient intervals, resulting in a potential gradient rate of change sequence.

[0052] A potential gradient change curve is constructed based on the potential gradient change rate sequence. The potential gradient change curve is truncated using a preset abrupt change threshold. If an abrupt change curve is extracted from the potential gradient change curve, a high-risk pipeline segment is identified in the oil and gas pipeline using the gradient interval corresponding to the abrupt change curve.

[0053] Optionally, obtaining multiple detection parameter node sequences based on the high-frequency detection point sequence includes:

[0054] Using preset high-frequency detection intervals, multiple high-frequency potential detection points were identified on high-risk pipeline sections.

[0055] High-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters are collected at multiple high-frequency potential detection points using a preset high-frequency acquisition time interval. This yields high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence. The high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence are then merged to obtain multiple detection parameter node sequences.

[0056] Optionally, the step of constructing a corrosion rate correlation model using multiple detection parameter node sequences and pre-built machine learning methods includes:

[0057] Statistical features are extracted from the high-frequency pipeline potential sequence in multiple detection parameter node sequences to obtain a high-frequency pipeline potential feature set, which includes multiple high-frequency pipeline potential features.

[0058] Feature construction is performed on the coating state sequence and soil physicochemical parameter sequence in multiple detection parameter node sequences to obtain a high-frequency environmental feature set, which includes multiple high-frequency environmental features.

[0059] A fusion feature matrix is ​​constructed using multiple high-frequency pipeline potential characteristics and multiple high-frequency environmental characteristics.

[0060] The corrosion rate label vector is obtained from a pre-constructed historical detection database, and the corrosion rate label vector includes high-frequency pipeline potential characteristics, high-frequency environmental characteristics and measured corrosion rate values.

[0061] A corrosion rate correlation model is constructed by fusing feature matrices, corrosion rate label vectors, and pre-built machine learning methods.

[0062] Optionally, the steady-state corrosion rate sequence is obtained by utilizing multiple detection parameter node sequences and a corrosion rate correlation model. Based on a preset grading threshold and the steady-state corrosion rate sequence, the corrosion risk level of high-risk pipeline sections in the oil and gas pipeline is classified to obtain an oil and gas pipeline corrosion risk assessment report, including:

[0063] Multiple detection parameter node sequences are input into the corrosion rate correlation model to obtain the corrosion rate prediction sequence;

[0064] The corrosion rate prediction sequence was processed by time window moving average to obtain the steady-state corrosion rate sequence;

[0065] The steady-state corrosion rate sequence is compared step by step with the preset multi-level corrosion risk thresholds to obtain the risk level label sequence;

[0066] The corrosion risk level of high-risk pipeline sections is marked by risk level label sequence to obtain corrosion risk assessment report for oil and gas pipelines.

[0067] To achieve the above objectives, the present invention also provides a corrosion evaluation system for oil and gas pipelines based on potential detection, comprising:

[0068] The original potential acquisition module is used to identify multiple potential detection points on a pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, the pipeline potential, soil potential and environmental parameters are collected at the multiple potential detection points to obtain multiple original potential node timing sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node timing sequence corresponds one-to-one with the potential detection points.

[0069] Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes;

[0070] The potential gradient processing module performs the following operations on each of the multiple calibration potential sequences:

[0071] Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence;

[0072] The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence;

[0073] By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained.

[0074] Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set;

[0075] The rate model building module is used to identify high-risk pipeline segments in oil and gas pipelines using a comprehensive potential gradient set.

[0076] A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0077] A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods;

[0078] The corrosion pipeline evaluation module is used to obtain a steady-state corrosion rate sequence by using multiple detection parameter node sequences and corrosion rate correlation models. Based on the preset classification threshold and steady-state corrosion rate sequence, the module classifies the corrosion risk level of high-risk pipeline sections in oil and gas pipelines and obtains an oil and gas pipeline corrosion risk evaluation report.

[0079] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0080] Memory, storing at least one instruction; and

[0081] The processor executes the instructions stored in the memory to implement the above-described method for evaluating corrosion of oil and gas pipelines based on potential detection.

[0082] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the above-described method for evaluating corrosion of oil and gas pipelines based on potential detection.

[0083] To address the problems described in the background art, this invention identifies multiple potential detection points on a pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential, and environmental parameters are collected at these multiple potential detection points to obtain multiple original potential node time sequences. These original potential nodes include pipeline potential, soil potential, and environmental parameters, and each original potential node time sequence corresponds one-to-one with a potential detection point. Based on these multiple original potential node time sequences, multiple calibration potential sequences are obtained. It is evident that this invention significantly improves the accuracy and reliability of corrosion monitoring through simultaneous acquisition of multiple parameters. Furthermore, by integrating potential and environmental data, it effectively reduces the false alarm rate caused by stray currents and other interference. Based on this, the present invention performs the following operations on each calibration potential sequence in multiple calibration potential sequences: performing potential gradient analysis on the calibration potential sequence to obtain a potential gradient sequence; performing missing data interpolation on the potential gradient sequence to obtain a potential gradient distribution sequence; summarizing the potential gradient distribution sequences to obtain a potential gradient distribution sequence set; and performing anomaly screening on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set. It can be seen that the embodiments of the present invention achieve accurate extraction of corrosion signals and noise suppression through multi-level processing of potential gradient analysis, intelligent interpolation, and anomaly screening, significantly improving the integrity of potential gradient data. Next, this invention utilizes a comprehensive potential gradient set to identify high-risk pipeline segments in oil and gas pipelines, and identifies a high-frequency detection point sequence within these high-risk segments. This high-frequency detection point sequence includes multiple high-frequency detection points. Based on this sequence, multiple detection parameter node sequences are obtained, each including multiple detection parameter nodes. These detection parameter nodes include high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters. A corrosion rate correlation model is constructed using these multiple detection parameter node sequences and a pre-built machine learning method. It is evident that this embodiment of the invention significantly improves the sensitivity and timeliness of corrosion risk identification by focusing on high-risk pipeline segments and implementing high-frequency multi-parameter monitoring. Combined with the corrosion rate correlation model constructed using machine learning, it achieves a precise mapping from potential characteristics to corrosion rate, providing decision support for pipeline integrity management. Furthermore, this invention utilizes multiple detection parameter node sequences and a corrosion rate correlation model to obtain a steady-state corrosion rate sequence. Based on preset grading thresholds and the steady-state corrosion rate sequence, high-risk pipeline sections in oil and gas pipelines are classified into corrosion risk levels, resulting in an oil and gas pipeline corrosion risk assessment report. It is evident that this invention, through establishing a steady-state corrosion rate sequence and intelligent grading assessment, achieves dynamic quantification and precise early warning of pipeline corrosion risk, significantly improving the timeliness and accuracy of risk management. It can automatically identify high-risk pipeline sections and generate visual reports, providing data support for pipeline maintenance decisions and effectively reducing the probability of sudden corrosion accidents. Therefore, this invention can solve the problems of isolated analysis of detection parameters and poor early corrosion warning capabilities in existing oil and gas pipeline corrosion assessments. Attached Figure Description

[0084] Figure 1 This is a schematic flowchart of a corrosion evaluation method for oil and gas pipelines based on potential detection provided in an embodiment of the present invention.

[0085] Figure 2 A functional block diagram of an oil and gas pipeline corrosion evaluation system based on potential detection provided in an embodiment of the present invention;

[0086] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the potential detection-based corrosion evaluation method for oil and gas pipelines, as provided in an embodiment of the present invention.

[0087] Explanation of reference numerals in the attached figures:

[0088] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0089] 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

[0090] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0091] This application provides a method for evaluating corrosion of oil and gas pipelines based on potential detection. The executing entity of the method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0092] Reference Figure 1 The diagram shown is a flowchart illustrating a potential-based corrosion evaluation method for oil and gas pipelines according to an embodiment of the present invention. In this embodiment, the potential-based corrosion evaluation method for oil and gas pipelines includes:

[0093] S1. Multiple potential detection points are identified on the pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential and environmental parameters are collected at the multiple potential detection points to obtain multiple original potential node time sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node time sequences correspond one-to-one with the potential detection points.

[0094] It should be explained that the pre-confirmed oil and gas pipeline is a section of oil and gas transportation pipeline that has been confirmed through engineering surveys before corrosion evaluation is conducted. Potential detection points are detection locations arranged at preset intervals (e.g., every 100 meters) along the pre-confirmed oil and gas pipeline, and data acquisition equipment is deployed at these points to collect pipeline potential, soil potential, and environmental parameters. The preset data acquisition time interval is a pre-set time interval, for example, 24 hours. Pipeline potential is the potential difference between the pipeline and the soil measured by a reference electrode, used to reflect the electrochemical corrosion state of the pipeline. Soil potential is the potential difference between the soil and the reference electrode at the potential detection point, used to analyze soil corrosivity and its impact on pipeline potential. Environmental parameters include, but are not limited to, soil moisture, pH value, soil resistivity, chloride ion concentration, and soil temperature. The environmental parameters to be detected can be set empirically to assist in analyzing potential environmental factors causing potential changes. Optionally, pre-confirmed hygrometers, in-situ pH sensors, quadrupole resistivity meters, and chloride ion selective electrodes can be used to collect environmental parameters. The raw potential node time series is a sequence of data composed of raw potential nodes arranged in chronological order of acquisition time, with each detection point corresponding to a time series, used for subsequent calibration and analysis. The raw potential node is a node composed of the corresponding pipeline potential, soil potential, and environmental parameters at each potential detection point.

[0095] For example, on a 5-kilometer-long pre-confirmed oil and gas pipeline, a potential detection point is set up every 100 meters, for a total of 51 potential detection points. Pipeline potential, soil potential, and environmental parameters are collected once a day at 10:00 AM for 30 consecutive days, resulting in 51 sets of raw potential node time series, each set containing data records of 30 time nodes.

[0096] S2. Obtain multiple calibration potential sequences based on the timing of multiple original potential nodes.

[0097] It should be explained that the acquisition of multiple calibration potential sequences based on the timing of multiple original potential nodes includes:

[0098] For each of the multiple original potential node timing sequences, the following operation is performed:

[0099] A three-dimensional correlation matrix is ​​constructed using the original potential node timing, and the noise coupling coefficient matrix is ​​obtained based on the three-dimensional correlation matrix.

[0100] A noise prediction model is constructed using a pre-built time series analysis method and a noise coupling coefficient matrix, and a noise prediction sequence is obtained using the noise prediction model.

[0101] Based on the original potential node timing sequence and noise prediction sequence, the pipeline potential at each potential detection point is filtered to obtain the filtered potential sequence.

[0102] By summarizing the filtered potential sequences, multiple filtered potential sequences are obtained, in which each filtered potential sequence corresponds one-to-one with the timing sequence of the original potential nodes.

[0103] Based on the time of the original potential nodes, multiple filtered potential sequences are reconstructed to obtain multiple calibration potential sequences.

[0104] Furthermore, the three-dimensional correlation matrix is ​​constructed from the original potential node time series according to the three dimensions of pipeline potential, soil potential, and environmental parameters in terms of acquisition time and parameter type. It is used to analyze the spatiotemporal correlation between parameters, such as potential fluctuations at the same detection point at different times. The noise coupling coefficient matrix is ​​the principal component coefficient matrix extracted from the three-dimensional correlation matrix using principal component analysis. The noise coupling coefficient matrix characterizes the coupling relationship of noise in time and space, such as the influence of environmental interference (e.g., soil moisture) on the potential of multiple detection points.

[0105] For ease of understanding, we will take soil moisture and soil temperature as an example and construct a three-dimensional correlation matrix as shown below: In the three-dimensional correlation matrix, the first row corresponds to 8:00 AM, the second row to 11:00 AM, the third row to 2:00 PM, and the fourth row to 5:00 PM. The first column represents pipe potential, the second column represents soil potential, the third column represents soil temperature, and the fourth column represents soil moisture. Understandably, time series analysis is a method used to extract noise time-series features; optionally, a long short-term memory artificial neural network is used as the time series analysis method. The noise prediction model is a model obtained by training the time series analysis method using a noise coupling coefficient matrix, capable of predicting noise values ​​at future time nodes. The noise prediction sequence is a set of noise-affected potential estimates output by the noise prediction model, where each noise-affected potential estimate corresponds one-to-one with the pipe potential in the original potential node time series. Filtering removes interference components from the noise prediction sequence from the original potential nodes; understandably, the pipe potential in the original potential node time series is subtracted from the corresponding noise-affected potential estimate to obtain the denoised pipe potential. The filtered potential sequence is a sequence of pipe potentials after noise reduction, where the pipe potentials in the filtered potential sequence correspond one-to-one with the pipe potentials in the original potential node time series. Based on the acquired original potential node times, multiple filtered potential sequences are reconstructed to obtain multiple calibration potential sequences. This involves rearranging multiple filtered potential sequences according to the acquisition time intervals to obtain multiple calibration potential sequences aligned with the acquisition time.

[0106] S3. Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence.

[0107] It should be explained that the step of performing a potential gradient analysis operation on the calibration potential sequence to obtain a potential gradient sequence includes:

[0108] A first potential detection point is identified in the calibration potential sequence. Based on the first potential detection point, a second potential detection point is identified in the calibration potential sequence. The second potential detection point is adjacent to the first potential detection point and is placed after the first potential detection point.

[0109] Extract the first potential value and the second potential value corresponding to the first potential detection point and the second potential detection point from the calibration potential sequence;

[0110] The potential gradient is obtained using the first potential value and the second potential value;

[0111] Using the second potential detection point as the first potential detection point, and returning to the step of confirming the second potential detection point in the calibration potential sequence based on the first potential detection point, the potential gradients are summarized to obtain the potential gradient sequence.

[0112] Furthermore, the first potential detection point is the starting detection point selected in the calibration potential sequence according to spatial order (e.g., pipeline mileage from the beginning to the end). The second potential detection point is a detection point adjacent to and following the first potential detection point, with a preset detection point interval (e.g., 100 meters) between them. The first potential value is the pipeline potential corresponding to the first potential detection point. The second potential value is the pipeline potential corresponding to the second potential detection point.

[0113] Understandably, obtaining the potential gradient using the first and second potential values ​​refers to calculating the potential gradient using the first and second potential values ​​and the distance between the two potential detection points. The potential gradient is the rate of change of potential per unit length between adjacent detection points, directly reflecting the corrosion or insulation status of the pipeline. The potential gradient sequence is a sequence composed of the potential gradients corresponding to all adjacent detection points.

[0114] S4. Perform missing data interpolation on the potential gradient sequence to obtain the potential gradient distribution sequence.

[0115] It should be explained that the process of interpolating missing data in the potential gradient sequence to obtain the potential gradient distribution sequence includes:

[0116] The integrity of the potential gradient sequence is checked to obtain the missing tag vector set;

[0117] Once it is confirmed that the set of missing label vectors includes at least one missing label vector, the following operation is performed on each missing label vector in the set of missing label vectors:

[0118] The missing label vector and the potential gradient sequence are input into a pre-constructed missing data labeling unit to obtain a labeled potential gradient matrix;

[0119] A local feature matrix is ​​constructed using the environmental parameters of each potential detection point in the time series of multiple original potential nodes;

[0120] The labeled potential gradient matrix and local feature matrix are input into the pre-constructed interpolation prediction model to obtain the predicted potential gradient value set, which includes multiple predicted potential gradient values, and the predicted potential gradient values ​​correspond one-to-one with the missing label vector.

[0121] The potential gradient matrix is ​​filled with the predicted potential gradient values ​​from the predicted potential gradient value set and the missing label vectors from the missing label vector set to obtain the potential gradient distribution sequence.

[0122] Further, integrity verification is a method that scans the potential gradient sequence to identify the presence of missing potential gradients. Optionally, anomaly detection can be used as the integrity verification method. The missing label vector set is a collection of multiple missing label vectors. The missing label vector is a vector that identifies the location of missing potential gradients using integrity verification, and is used to locate the missing data positions in the potential gradient sequence. The pre-constructed missing data marking unit receives the potential gradient sequence and missing label vectors, marks the missing positions in the sequence with specific identifiers, and outputs a potential gradient matrix with missing labels. Optionally, null value marking can be used as the missing data marking unit. The marked potential gradient matrix is ​​a matrix obtained by constructing a potential gradient matrix using the potential gradient sequence and marking the positions of missing data in the potential gradient matrix using missing label vectors.

[0123] For example, suppose a pipeline section has 6 detection points (P1-P6), and the marked potential gradient matrix at 4 time points is shown below: In the marked potential gradient matrix, the first row corresponds to 8:00 AM, the second row to 11:00 AM, the third row to 2:00 PM, and the fourth row to 5:00 PM. The first column represents the potential gradient between detection points P1 and P2, the second column between P2 and P3, the third column between P3 and P4, the fourth column between P4 and P5, and the fifth column between P5 and P6. Understandably, constructing a local feature matrix using the environmental parameters of each potential detection point in the time series of multiple original potential nodes refers to a numerical matrix constructed by summarizing the environmental parameters of each potential detection point, used to quantify the corrosion influencing factors of the environment surrounding oil and gas pipelines. A pre-constructed interpolation prediction model is used to predict missing data in the marked potential gradient matrix; optionally, the Kriging method is used as the interpolation prediction model. The predicted potential gradient value set is a collection containing multiple predicted potential gradient values. These predicted potential gradient values ​​are the results of predicting missing potential gradient data using an interpolation prediction model. The potential gradient distribution sequence is the complete sequence obtained by filling the labeled potential gradient matrix with the predicted potential gradient values.

[0124] For example, the potential gradient distribution sequence is shown below:

[0125] 08:00 - [0.02, 0.021, 0.03, 0.01, 0.02]

[0126] 11:00 - [0.01, 0.02, 0.024, 0.015, 0.03]

[0127] 14:00 - [0.022, 0.01, 0.02, 0.02, 0.019]

[0128] 17:00 - [0.03, 0.02, 0.01, 0.01, 0.02]

[0129] S5. Perform anomaly screening on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set.

[0130] It should be explained that the process of screening multiple potential gradient distribution sequences in the potential gradient distribution sequence set for anomalies to obtain a comprehensive potential gradient set includes:

[0131] Based on the potential gradients from multiple potential gradient distribution sequences obtained by summarizing the potential gradients at potential detection points, multiple categorical potential gradient sequences are obtained. The following operation is performed on each of the multiple categorical potential gradient sequences:

[0132] The classification potential gradients are extracted sequentially from the classification potential gradient sequence, and the following operations are performed on the extracted classification potential gradients:

[0133] The extracted classification potential gradients are removed from the classification potential gradient sequence to obtain a reference classification potential gradient set. The potential gradient deviation value is calculated using a pre-constructed deviation formula, a pre-constructed potential gradient deviation function, the extracted classification potential gradients, and the reference classification potential gradient set. The deviation formula is shown below:

[0134]

[0135] Where D represents the potential gradient deviation, M j Let f(t0,t) represent the j-th reference classification potential gradient in the reference classification potential gradient set, M0 represent the extracted classification potential gradient, and f(t0,t) represent the j-th reference classification potential gradient in the reference classification potential gradient set. j ) represents the prediction bias of the potential gradient predicted using the potential gradient bias function, and t0 represents the time corresponding to the extracted classification potential gradient. j This represents the time corresponding to the j-th reference classification potential gradient in the reference classification potential gradient set, and k represents the total number of k reference classification potential gradients in the reference classification potential gradient set.

[0136] The potential gradient deviation value is compared with a preset potential gradient deviation threshold. If the potential gradient deviation value is greater than or equal to the potential gradient deviation threshold, the extracted potential gradient is removed from the classification potential gradient sequence to obtain an updated classification potential gradient sequence.

[0137] The updated classification potential gradient sequences are summarized to obtain multiple updated classification potential gradient sequences. Using a pre-constructed reference mapping sequence, the multiple updated classification potential gradient sequences are mapped to the reference mapping sequence to obtain multiple mapped potential gradient sequences.

[0138] Extract the mapped potential gradient sequence sequentially from multiple mapped potential gradient sequences, and perform the following operations on the extracted mapped potential gradient sequences:

[0139] Searching is performed in the extracted mapping potential gradient sequence. If a preset missing value is found in the mapping potential gradient sequence, multiple reference potential gradients are extracted from multiple mapping potential gradient sequences using the missing value and the position corresponding to the extracted mapping potential gradient sequence.

[0140] Calculate the mean of multiple reference potential gradients to obtain the reference potential mean. Use the reference potential mean to replace the missing value to obtain the target potential gradient sequence.

[0141] By summing the target potential gradient sequences, multiple target potential gradient sequences are obtained;

[0142] A comprehensive potential gradient set is obtained from multiple target potential gradient sequences.

[0143] Furthermore, the potential gradient distribution sequence set is a collection of potential gradient distribution sequences from multiple different time points. The categorical potential gradient sequence is obtained by sorting the potential gradients at potential detection points according to the acquisition time interval. The categorical potential gradient is the potential gradient at a potential detection point at a single acquisition time point. The reference categorical potential gradient set is a collection of multiple categorical potential gradients from the categorical potential gradient sequence after removing the extracted categorical potential gradients. The reference categorical potential gradient is the categorical potential gradient in the categorical potential gradient sequence after removing the extracted categorical potential gradients. The potential gradient deviation value is the comprehensive degree of deviation between the categorical potential gradient and the reference categorical potential gradient. The potential gradient deviation function is a function constructed based on the ARIMA time series model, used to characterize the change of the potential gradient corresponding to the same potential detection point over time. This allows for the correction of the potential prediction deviation corresponding to the extracted categorical potential gradient by combining different time points, thereby improving the identification of abnormal categorical potential gradients in the categorical potential gradient sequence and ensuring the accuracy of subsequently obtaining the comprehensive potential gradient set. The potential gradient prediction bias is the prediction deviation from the classification potential gradient calculated using the potential gradient deviation function at the time corresponding to the classification potential gradient. This prediction bias is represented by the difference between the predicted value and the classification potential gradient. The predicted potential gradient is the predicted value of the potential gradient obtained using the potential deviation function and the selected potential gradient. For example, by using the potential gradient deviation function in conjunction with the time points corresponding to different classification potential gradients, the value of another classification potential gradient is predicted using the time points corresponding to two classification potential gradients and one classification potential gradient. The predicted value is then calculated, and the absolute difference between the predicted value and the other classification potential gradient is obtained, thus yielding the potential gradient prediction bias.

[0144] It should be explained that the updated classification potential gradient sequence is the potential gradient sequence obtained after removing classification potential gradients whose potential gradient deviation value is greater than or equal to the potential gradient deviation threshold. The pre-constructed reference mapping sequence is a standardized index sequence used to map the potential gradients in the updated classification potential gradient sequence to the index sequence, so as to retrieve missing values ​​in the updated classification potential gradient sequence. The mapped potential gradient sequence is a sequence generated by aligning the updated classification potential gradient sequence according to the reference mapping sequence, and there may be gaps in the sequence. The preset missing value refers to the missing position in the mapped potential gradient sequence. The reference potential gradient is the potential gradient extracted from other mapped potential gradient sequences at the same position corresponding to the missing value. The target potential gradient sequence is the complete mapped potential gradient sequence obtained by replacing the missing values ​​in the mapped potential gradient with the mean of the reference potential. The comprehensive potential gradient set is the set obtained by averaging the potential gradients at the same potential detection point in multiple target potential gradient sequences. Optionally, when summarizing the potential gradient using potential detection points, the first potential detection point in the potential detection point sequence can be removed, the first potential gradient in the potential gradient distribution sequence can be summed to the second potential detection point, and the second potential gradient in the potential gradient distribution sequence can be summed to the third potential detection point, and so on, so as to summarize the potential gradients in multiple potential gradient distribution sequences based on the potential detection points and obtain multiple classified potential gradient sequences.

[0145] S6. High-risk pipeline sections in oil and gas pipelines are identified by using the comprehensive potential gradient set.

[0146] It should be explained that the identification of high-risk pipeline segments in oil and gas pipelines using comprehensive potential gradient sets includes:

[0147] Based on the multiple potential detection points, multiple gradient intervals are identified in the comprehensive potential gradient set, and the number of gradient intervals plus one equals the number of multiple potential detection points.

[0148] The potential gradient rate of change for each gradient interval is obtained by using multiple gradient intervals, resulting in a potential gradient rate of change sequence.

[0149] A potential gradient change curve is constructed based on the potential gradient change rate sequence. The potential gradient change curve is truncated using a preset abrupt change threshold. If an abrupt change curve is extracted from the potential gradient change curve, a high-risk pipeline segment is identified in the oil and gas pipeline using the gradient interval corresponding to the abrupt change curve.

[0150] Furthermore, the gradient interval is the pipe segment region between two adjacent potential detection points. The rate of change of potential gradient is a numerical value representing the drastic change in potential gradient, and the formula for calculating the rate of change of potential gradient is as follows:

[0151]

[0152] Where, γ i This represents the rate of change of the potential gradient per unit length between the potential gradients of the i-th gradient interval and the (i+1)-th gradient interval. This represents the potential gradient in the (i+1)th gradient interval. Let L represent the potential gradient of the i-th gradient interval, and L represent the length of the gradient interval (the distance between two adjacent potential detection points).

[0153] Understandably, the potential gradient rate of change sequence is a sequence of potential gradient rates arranged in order of potential detection points from front to back. The potential gradient rate of change curve is a line graph with the potential detection points on the x-axis and the potential rate of change on the y-axis. The preset abrupt change threshold is a threshold for the potential gradient rate of change set based on historical data statistics, for example, 0.4V / m. 2 High-risk pipeline sections are those potentially corroded in oil and gas pipelines identified using the gradient intervals corresponding to abrupt change curves.

[0154] For example, suppose a pipeline section has 7 potential detection points (P1-P7) with a spacing of L = 100 meters, and 6 potential gradients are identified: P1-P2, P2-P3, P3-P4, P4-P5, P5-P6, and P6-P7, where the corresponding potential gradient change rates are 0.5, 0.3, -0.45, -0.45, and -0.1. By comparing the potential gradient change rates with the abrupt change threshold, P1-P3 and P4-P7 are identified as high-risk pipeline sections.

[0155] S7. A high-frequency detection point sequence is identified in the high-risk pipeline section, and the high-frequency detection point sequence includes multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences are obtained, and the detection parameter node sequences include multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0156] It should be explained that obtaining multiple detection parameter node sequences based on the high-frequency detection point sequence includes:

[0157] Using preset high-frequency detection intervals, multiple high-frequency potential detection points were identified on high-risk pipeline sections.

[0158] High-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters are collected at multiple high-frequency potential detection points using a preset high-frequency acquisition time interval. This yields high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence. The high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence are then merged to obtain multiple detection parameter node sequences.

[0159] Furthermore, the preset high-frequency detection interval is the spacing between detection points set within the high-risk pipeline section, for example, 10 meters. High-frequency potential detection points are potential detection points deployed within the high-risk pipeline section according to the high-frequency detection interval. The preset high-frequency acquisition time interval is the preset time interval for acquiring detection parameter nodes, for example, 3 hours. The high-frequency pipeline potential sequence is the pipeline potential recorded in chronological order of acquisition time. The coating state sequence is the pipeline anti-corrosion layer state data recorded in chronological order of acquisition time. The soil physicochemical parameter sequence is the soil parameter sequence recorded in chronological order of acquisition time. The high-frequency pipeline potential is the pipeline potential measured at high-frequency potential detection points at high-frequency acquisition time intervals (e.g., 3 hours), used to monitor the corrosion electrochemical state of high-risk pipeline sections. The pipeline coating state data is a parameter characterizing the integrity of the pipeline anti-corrosion layer; optionally, the DC ground potential gradient method is used as the method for obtaining the pipeline coating state data. Soil physicochemical parameters include, but are not limited to, soil moisture, pH, soil resistivity, chloride ion concentration, and soil temperature. The specific soil physicochemical parameters to be measured can be empirically determined to aid in the analysis of potential environmental factors contributing to potential changes. Optionally, pre-validated thermometers, hygrometers, in-situ pH sensors, quadrupole resistivity meters, and chloride ion-selective electrodes can be used to collect environmental parameters. The detection parameter node sequence is a sequence obtained by merging the high-frequency pipeline potential sequence, coating state sequence, and soil physicochemical parameter sequence, with each detection point corresponding to one detection parameter node sequence.

[0160] S8. Construct a corrosion rate correlation model using multiple detection parameter node sequences and pre-built machine learning methods.

[0161] It should be explained that the construction of the corrosion rate correlation model using multiple detection parameter node sequences and pre-built machine learning methods includes:

[0162] Statistical features are extracted from the high-frequency pipeline potential sequence in multiple detection parameter node sequences to obtain a high-frequency pipeline potential feature set, which includes multiple high-frequency pipeline potential features.

[0163] Feature construction is performed on the coating state sequence and soil physicochemical parameter sequence in multiple detection parameter node sequences to obtain a high-frequency environmental feature set, which includes multiple high-frequency environmental features.

[0164] A fusion feature matrix is ​​constructed using multiple high-frequency pipeline potential characteristics and multiple high-frequency environmental characteristics.

[0165] The corrosion rate label vector is obtained from a pre-constructed historical detection database, and the corrosion rate label vector includes high-frequency pipeline potential characteristics, high-frequency environmental characteristics and measured corrosion rate values.

[0166] A corrosion rate correlation model is constructed by fusing feature matrices, corrosion rate label vectors, and pre-built machine learning methods.

[0167] Furthermore, the high-frequency pipeline potential feature set is a collection containing multiple high-frequency pipeline potential features. High-frequency pipeline potential features are statistical features extracted from multiple high-frequency pipeline potential sequences, such as the rate of change of high-frequency pipeline potential at the same position in different high-frequency pipeline potential sequences at the same time point. Statistical feature extraction is a method for extracting high-frequency pipeline potential features from high-frequency pipeline potential sequences; optionally, wavelet transform can be used as the statistical feature extraction method.

[0168] Understandably, feature construction is a method of generating new features by combining pipeline coating state data and soil physicochemical parameters. Optionally, statistical feature construction can be used as the feature construction method. The high-frequency environmental feature set is a collection of multiple high-frequency environmental features used to characterize the external environmental state of the pipeline. High-frequency environmental features are features constructed using pipeline coating state data and soil physicochemical parameters. The fusion feature matrix is ​​a matrix obtained by merging high-frequency pipeline potential features and high-frequency environmental features, used as input to the machine learning model. The pre-built historical detection database is a database storing the potential features, environmental features, and measured corrosion rate values ​​from excavation detections over the years. The corrosion rate label vector is a vector representation of the high-frequency pipeline potential features, high-frequency environmental features, and measured corrosion rate values. The measured corrosion rate values ​​are corrosion rate data directly obtained from the historical detection database. The pre-built machine learning method is a method used to train the corrosion rate prediction model. Optionally, the random forest algorithm can be used as the machine learning method. The corrosion rate association model is a trained machine learning model capable of obtaining predicted corrosion rates based on the input detection parameter node sequence.

[0169] S9. Obtain a steady-state corrosion rate sequence by using multiple detection parameter node sequences and corrosion rate correlation models. Based on the preset grading threshold and steady-state corrosion rate sequence, classify the corrosion risk level of high-risk pipeline sections in oil and gas pipelines to obtain an oil and gas pipeline corrosion risk assessment report.

[0170] It should be explained that the process of obtaining a steady-state corrosion rate sequence using multiple detection parameter node sequences and a corrosion rate correlation model, and classifying the corrosion risk level of high-risk pipeline sections in oil and gas pipelines according to preset grading thresholds and the steady-state corrosion rate sequence, results in an oil and gas pipeline corrosion risk assessment report, including:

[0171] Multiple detection parameter node sequences are input into the corrosion rate correlation model to obtain the corrosion rate prediction sequence;

[0172] The corrosion rate prediction sequence was processed by time window moving average to obtain the steady-state corrosion rate sequence;

[0173] The steady-state corrosion rate sequence is compared step by step with the preset multi-level corrosion risk thresholds to obtain the risk level label sequence;

[0174] The corrosion risk level of high-risk pipeline sections is marked by risk level label sequence to obtain corrosion risk assessment report for oil and gas pipelines.

[0175] Furthermore, the corrosion rate prediction sequence is a sequence of multiple predicted corrosion rate values ​​output after the detection parameter node sequence is input into the corrosion rate association model. Each predicted corrosion rate value represents the predicted corrosion rate of the pipe segment between two adjacent detection points. The time window moving average processing is an operation to smooth the corrosion rate prediction sequence and eliminate instantaneous fluctuation noise. Optionally, exponential smoothing can be used as the time window moving average processing method. The steady-state corrosion rate sequence is the corrosion rate prediction sequence after time window moving average processing.

[0176] It should be explained that the preset multi-level corrosion risk thresholds are critical values ​​for corrosion rate risk levels based on historical data. Comparing the steady-state corrosion rate sequence with the preset multi-level corrosion risk thresholds level by level to obtain the risk level label sequence means assigning a corresponding risk level label to each steady-state corrosion rate in the sequence, comparing it with the multi-level corrosion risk threshold. Risk level labels include low-corrosion-risk pipe sections, medium-corrosion-risk pipe sections, and high-corrosion-risk pipe sections. For example, pipe sections with a corrosion rate less than 0.1 mm / year are identified as low-corrosion-risk sections, pipe sections with a corrosion rate greater than or equal to 0.1 mm / year and less than or equal to 0.3 mm / year are identified as medium-corrosion-risk sections, and pipe sections with a corrosion rate greater than 0.3 mm / year are identified as high-corrosion-risk sections. The risk level label sequence is a sequence composed of multiple risk level labels. Corrosion risk level labeling maps the risk level labels to pipe sections, forming spatial-risk correlation data. The oil and gas pipeline corrosion risk assessment report is a report containing the corrosion risk level of each pipe section within the high-risk pipeline segment. For example, an oil and gas pipeline is divided into three pipeline segments: the first pipeline segment, the second pipeline segment, and the third pipeline segment. The risk level label sequence obtained from the three pipeline segments is: low corrosion risk segment, high corrosion risk segment, and low corrosion risk segment. The high-risk pipeline segment is then labeled with its corrosion risk level using the risk level label sequence. Thus, the second pipeline segment is labeled as the high corrosion risk segment - the second pipeline segment.

[0177] To address the problems described in the background art, this invention identifies multiple potential detection points on a pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential, and environmental parameters are collected at these multiple potential detection points to obtain multiple original potential node time sequences. These original potential nodes include pipeline potential, soil potential, and environmental parameters, and each original potential node time sequence corresponds one-to-one with a potential detection point. Based on these multiple original potential node time sequences, multiple calibration potential sequences are obtained. It is evident that this invention significantly improves the accuracy and reliability of corrosion monitoring through simultaneous acquisition of multiple parameters. Furthermore, by integrating potential and environmental data, it effectively reduces the false alarm rate caused by stray currents and other interference. Based on this, the present invention performs the following operations on each calibration potential sequence in multiple calibration potential sequences: performing potential gradient analysis on the calibration potential sequence to obtain a potential gradient sequence; performing missing data interpolation on the potential gradient sequence to obtain a potential gradient distribution sequence; summarizing the potential gradient distribution sequences to obtain a potential gradient distribution sequence set; and performing anomaly screening on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set. It can be seen that the embodiments of the present invention achieve accurate extraction of corrosion signals and noise suppression through multi-level processing of potential gradient analysis, intelligent interpolation, and anomaly screening, significantly improving the integrity of potential gradient data. Next, this invention utilizes a comprehensive potential gradient set to identify high-risk pipeline segments in oil and gas pipelines, and identifies a high-frequency detection point sequence within these high-risk segments. This high-frequency detection point sequence includes multiple high-frequency detection points. Based on this sequence, multiple detection parameter node sequences are obtained, each including multiple detection parameter nodes. These detection parameter nodes include high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters. A corrosion rate correlation model is constructed using these multiple detection parameter node sequences and a pre-built machine learning method. It is evident that this embodiment of the invention significantly improves the sensitivity and timeliness of corrosion risk identification by focusing on high-risk pipeline segments and implementing high-frequency multi-parameter monitoring. Combined with the corrosion rate correlation model constructed using machine learning, it achieves a precise mapping from potential characteristics to corrosion rate, providing decision support for pipeline integrity management. Furthermore, this invention utilizes multiple detection parameter node sequences and a corrosion rate correlation model to obtain a steady-state corrosion rate sequence. Based on preset grading thresholds and the steady-state corrosion rate sequence, high-risk pipeline sections in oil and gas pipelines are classified into corrosion risk levels, resulting in an oil and gas pipeline corrosion risk assessment report. It is evident that this invention, through establishing a steady-state corrosion rate sequence and intelligent grading assessment, achieves dynamic quantification and precise early warning of pipeline corrosion risk, significantly improving the timeliness and accuracy of risk management. It can automatically identify high-risk pipeline sections and generate visual reports, providing data support for pipeline maintenance decisions and effectively reducing the probability of sudden corrosion accidents. Therefore, this invention can solve the problems of isolated analysis of detection parameters and poor early corrosion warning capabilities in existing oil and gas pipeline corrosion assessments.

[0178] like Figure 2 The diagram shown is a functional block diagram of an oil and gas pipeline corrosion evaluation system based on potential detection provided in an embodiment of the present invention.

[0179] The potential detection-based oil and gas pipeline corrosion evaluation system 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the potential detection-based oil and gas pipeline corrosion evaluation system 100 may include a raw potential acquisition module 101, a potential gradient processing module 102, a rate model construction module 103, and a corrosion pipeline evaluation module 104. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0180] The original potential acquisition module 101 is used to identify multiple potential detection points on the pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, it acquires pipeline potential, soil potential and environmental parameters at the multiple potential detection points to obtain multiple original potential node timing sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node timing sequence corresponds one-to-one with the potential detection points.

[0181] Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes;

[0182] The potential gradient processing module 102 is used to perform the following operation on each of the multiple calibration potential sequences:

[0183] Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence;

[0184] The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence;

[0185] By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained.

[0186] Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set;

[0187] The rate model construction module 103 is used to identify high-risk pipeline segments in oil and gas pipelines using a comprehensive potential gradient set.

[0188] A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0189] A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods;

[0190] The corrosion pipeline evaluation module 104 is used to obtain a steady-state corrosion rate sequence by using multiple detection parameter node sequences and corrosion rate correlation models, and to classify the corrosion risk level of high-risk pipeline sections in oil and gas pipelines according to preset classification thresholds and steady-state corrosion rate sequences, so as to obtain an oil and gas pipeline corrosion risk evaluation report.

[0191] In detail, the modules in the potential detection-based oil and gas pipeline corrosion evaluation system 100 described in this embodiment of the invention employ the same methods as described above during use. Figure 1 The method used is the same as the potential detection-based corrosion evaluation method for oil and gas pipelines described above, and it can produce the same technical effect, so it will not be repeated here.

[0192] like Figure 3 The diagram shown is a schematic diagram of an electronic device for implementing a potential detection-based corrosion evaluation method for oil and gas pipelines, according to an embodiment of the present invention.

[0193] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a program for evaluating corrosion of oil and gas pipelines based on potential detection.

[0194] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a potential detection-based oil and gas pipeline corrosion evaluation method program, but also to temporarily store data that has been output or will be output.

[0195] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a corrosion evaluation method program for oil and gas pipelines based on potential detection) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0196] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0197] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0198] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, thereby enabling functions such as charging management, discharging management, and power consumption management through the power management system. The power supply may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0199] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0200] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0201] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0202] The program for evaluating corrosion of oil and gas pipelines based on potential detection, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0203] Multiple potential detection points were identified on the pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential and environmental parameters were collected at the multiple potential detection points to obtain multiple original potential node time sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node time sequences correspond one-to-one with the potential detection points.

[0204] Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes;

[0205] For each of the multiple calibration potential sequences, perform the following operation:

[0206] Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence;

[0207] The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence;

[0208] By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained.

[0209] Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set;

[0210] High-risk pipeline segments in oil and gas pipelines were identified using a comprehensive potential gradient set.

[0211] A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0212] A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods;

[0213] A steady-state corrosion rate sequence is obtained by using multiple detection parameter node sequences and a corrosion rate correlation model. Based on the preset grading threshold and the steady-state corrosion rate sequence, the corrosion risk level of high-risk pipeline sections in oil and gas pipelines is classified, resulting in an oil and gas pipeline corrosion risk assessment report.

[0214] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0215] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0216] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0217] Multiple potential detection points were identified on the pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential and environmental parameters were collected at the multiple potential detection points to obtain multiple original potential node time sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node time sequences correspond one-to-one with the potential detection points.

[0218] Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes;

[0219] For each of the multiple calibration potential sequences, perform the following operation:

[0220] Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence;

[0221] The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence;

[0222] By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained.

[0223] Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set;

[0224] High-risk pipeline segments in oil and gas pipelines were identified using a comprehensive potential gradient set.

[0225] A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters.

[0226] A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods;

[0227] Steady-state corrosion rate sequence is obtained by using multiple detection parameter node sequences and corrosion rate correlation models. Based on preset classification thresholds and steady-state corrosion rate sequences, the corrosion risk level of high-risk pipeline sections in oil and gas pipelines is classified, resulting in an oil and gas pipeline corrosion risk assessment report.

[0228] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0229] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0230] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0231] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0232] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim may also be implemented by a single unit or system through software or hardware. The term "second class" is used to indicate names and does not indicate any specific order.

[0233] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for evaluating corrosion of oil and gas pipelines based on potential detection, characterized in that, The method includes: Multiple potential detection points were identified on the pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, pipeline potential, soil potential and environmental parameters were collected at the multiple potential detection points to obtain multiple original potential node time sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node time sequences correspond one-to-one with the potential detection points. Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes; The process of obtaining multiple calibration potential sequences based on the timing of multiple original potential nodes includes: For each of the multiple original potential node timing sequences, the following operation is performed: A three-dimensional correlation matrix is ​​constructed using the original potential node timing, and the noise coupling coefficient matrix is ​​obtained based on the three-dimensional correlation matrix. A noise prediction model is constructed using a pre-built time series analysis method and a noise coupling coefficient matrix, and a noise prediction sequence is obtained using the noise prediction model. Based on the original potential node timing sequence and noise prediction sequence, the pipeline potential at each potential detection point is filtered to obtain the filtered potential sequence. By summarizing the filtered potential sequences, multiple filtered potential sequences are obtained, in which each filtered potential sequence corresponds one-to-one with the timing sequence of the original potential nodes. Based on the time of the original potential nodes, multiple filtered potential sequences are reconstructed to obtain multiple calibration potential sequences. For each of the multiple calibration potential sequences, perform the following operation: Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence; The step of performing a potential gradient analysis operation on the calibration potential sequence to obtain a potential gradient sequence includes: A first potential detection point is identified in the calibration potential sequence. Based on the first potential detection point, a second potential detection point is identified in the calibration potential sequence. The second potential detection point is adjacent to the first potential detection point and is placed after the first potential detection point. Extract the first potential value and the second potential value corresponding to the first potential detection point and the second potential detection point from the calibration potential sequence; The potential gradient is obtained using the first potential value and the second potential value; By summing the potential gradients, a potential gradient sequence is obtained; The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence; The step of interpolating missing data in the potential gradient sequence to obtain the potential gradient distribution sequence includes: The integrity of the potential gradient sequence is checked to obtain the missing tag vector set; Once it is confirmed that the set of missing label vectors includes at least one missing label vector, the following operation is performed on each missing label vector in the set of missing label vectors: The missing label vector and the potential gradient sequence are input into a pre-constructed missing data labeling unit to obtain a labeled potential gradient matrix; A local feature matrix is ​​constructed using the environmental parameters of each potential detection point in the time series of multiple original potential nodes; The labeled potential gradient matrix and local feature matrix are input into the pre-constructed interpolation prediction model to obtain the predicted potential gradient value set, which includes multiple predicted potential gradient values, and the predicted potential gradient values ​​correspond one-to-one with the missing label vector. The potential gradient matrix is ​​filled with the predicted potential gradient values ​​from the predicted potential gradient value set and the missing label vectors from the missing label vector set to obtain the potential gradient distribution sequence. By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained. Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set; The process involves screening multiple potential gradient distribution sequences in the potential gradient distribution sequence set for anomalies to obtain a comprehensive potential gradient set, including: Based on the potential gradients from multiple potential gradient distribution sequences obtained by summarizing the potential gradients at potential detection points, multiple categorical potential gradient sequences are obtained. The following operation is performed on each of the multiple categorical potential gradient sequences: The classification potential gradients are extracted sequentially from the classification potential gradient sequence, and the following operations are performed on the extracted classification potential gradients: The extracted classification potential gradients are removed from the classification potential gradient sequence to obtain a reference classification potential gradient set. The potential gradient deviation value is calculated using a pre-constructed deviation formula, a pre-constructed potential gradient deviation function, the extracted classification potential gradients, and the reference classification potential gradient set. The deviation formula is shown below: ； in, This indicates the deviation of the potential gradient. Indicates the first reference classification potential gradient set A reference classification potential gradient, This represents the extracted classification potential gradient. This represents the prediction bias of the potential gradient using the potential gradient bias function. This indicates the time corresponding to the extracted classification potential gradient. Indicates the first reference classification potential gradient set The time corresponding to each reference classification potential gradient This indicates that the reference classification potential gradient set has a total of One reference classification potential gradient; The potential gradient deviation value is compared with a preset potential gradient deviation threshold. If the potential gradient deviation value is greater than or equal to the potential gradient deviation threshold, the extracted potential gradient is removed from the classification potential gradient sequence to obtain an updated classification potential gradient sequence. The updated classification potential gradient sequences are summarized to obtain multiple updated classification potential gradient sequences. Using a pre-constructed reference mapping sequence, the multiple updated classification potential gradient sequences are mapped to the reference mapping sequence to obtain multiple mapped potential gradient sequences. Extract the mapped potential gradient sequence sequentially from multiple mapped potential gradient sequences, and perform the following operations on the extracted mapped potential gradient sequences: Searching is performed in the extracted mapping potential gradient sequence. If a preset missing value is found in the mapping potential gradient sequence, multiple reference potential gradients are extracted from multiple mapping potential gradient sequences using the missing value and the position corresponding to the extracted mapping potential gradient sequence. Calculate the mean of multiple reference potential gradients to obtain the reference potential mean. Use the reference potential mean to replace the missing value to obtain the target potential gradient sequence. By summing the target potential gradient sequences, multiple target potential gradient sequences are obtained; A comprehensive potential gradient set is obtained based on multiple target potential gradient sequences; High-risk pipeline segments in oil and gas pipelines were identified using a comprehensive potential gradient set. A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters. A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods; A steady-state corrosion rate sequence is obtained by using multiple detection parameter node sequences and a corrosion rate correlation model. Based on the preset grading threshold and the steady-state corrosion rate sequence, the corrosion risk level of high-risk pipeline sections in oil and gas pipelines is classified, resulting in an oil and gas pipeline corrosion risk assessment report.

2. The corrosion evaluation method for oil and gas pipelines based on potential detection as described in claim 1, characterized in that, The method of identifying high-risk pipeline segments in oil and gas pipelines using integrated potential gradient sets includes: Based on the multiple potential detection points, multiple gradient intervals are identified in the comprehensive potential gradient set, and the number of gradient intervals plus one equals the number of multiple potential detection points. The potential gradient rate of change for each gradient interval is obtained by using multiple gradient intervals, resulting in a potential gradient rate of change sequence. A potential gradient change curve is constructed based on the potential gradient change rate sequence. The potential gradient change curve is truncated using a preset abrupt change threshold. If an abrupt change curve is extracted from the potential gradient change curve, a high-risk pipeline segment is identified in the oil and gas pipeline using the gradient interval corresponding to the abrupt change curve.

3. The corrosion evaluation method for oil and gas pipelines based on potential detection as described in claim 2, characterized in that, The process of obtaining multiple detection parameter node sequences based on the high-frequency detection point sequence includes: Using preset high-frequency detection intervals, multiple high-frequency potential detection points were identified on high-risk pipeline sections. High-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters are collected at multiple high-frequency potential detection points using a preset high-frequency acquisition time interval. This yields high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence. The high-frequency pipeline potential sequence, coating status sequence, and soil physicochemical parameter sequence are then merged to obtain multiple detection parameter node sequences.

4. The corrosion evaluation method for oil and gas pipelines based on potential detection as described in claim 3, characterized in that, The method of constructing a corrosion rate correlation model using multiple detection parameter node sequences and pre-built machine learning methods includes: Statistical features are extracted from the high-frequency pipeline potential sequence in multiple detection parameter node sequences to obtain a high-frequency pipeline potential feature set, which includes multiple high-frequency pipeline potential features. Feature construction is performed on the coating state sequence and soil physicochemical parameter sequence in multiple detection parameter node sequences to obtain a high-frequency environmental feature set, which includes multiple high-frequency environmental features. A fusion feature matrix is ​​constructed using multiple high-frequency pipeline potential characteristics and multiple high-frequency environmental characteristics. The corrosion rate label vector is obtained from a pre-constructed historical detection database, and the corrosion rate label vector includes high-frequency pipeline potential characteristics, high-frequency environmental characteristics and measured corrosion rate values. A corrosion rate correlation model is constructed by fusing feature matrices, corrosion rate label vectors, and pre-built machine learning methods.

5. The corrosion evaluation method for oil and gas pipelines based on potential detection as described in claim 4, characterized in that, The steady-state corrosion rate sequence is obtained by utilizing multiple detection parameter node sequences and a corrosion rate correlation model. Based on preset grading thresholds and the steady-state corrosion rate sequence, the corrosion risk level of high-risk pipeline sections in oil and gas pipelines is classified, resulting in an oil and gas pipeline corrosion risk assessment report, including: Multiple detection parameter node sequences are input into the corrosion rate correlation model to obtain the corrosion rate prediction sequence; The corrosion rate prediction sequence was processed by time window moving average to obtain the steady-state corrosion rate sequence; The steady-state corrosion rate sequence is compared step by step with the preset multi-level corrosion risk thresholds to obtain the risk level label sequence; The corrosion risk level of high-risk pipeline sections is marked by risk level label sequence to obtain corrosion risk assessment report for oil and gas pipelines.

6. A system for evaluating corrosion of oil and gas pipelines based on potential detection as described in any one of claims 1 to 5, characterized in that, The system includes: The original potential acquisition module is used to identify multiple potential detection points on a pre-confirmed oil and gas pipeline. Using a preset acquisition time interval, the pipeline potential, soil potential and environmental parameters are collected at the multiple potential detection points to obtain multiple original potential node timing sequences. The original potential nodes include pipeline potential, soil potential and environmental parameters, and the original potential node timing sequence corresponds one-to-one with the potential detection points. Multiple calibration potential sequences are obtained based on the timing of multiple original potential nodes; The potential gradient processing module performs the following operations on each of the multiple calibration potential sequences: Perform a potential gradient analysis operation on the calibration potential sequence to obtain the potential gradient sequence; The potential gradient sequence is interpolated for missing data to obtain the potential gradient distribution sequence; By summarizing the potential gradient distribution sequences, a set of potential gradient distribution sequences is obtained. Anomaly screening was performed on multiple potential gradient distribution sequences in the potential gradient distribution sequence set to obtain a comprehensive potential gradient set; The rate model building module is used to identify high-risk pipeline segments in oil and gas pipelines using a comprehensive potential gradient set. A high-frequency detection point sequence was identified in a high-risk pipeline section, and the high-frequency detection point sequence included multiple high-frequency detection points. Based on the high-frequency detection point sequence, multiple detection parameter node sequences were obtained, and the detection parameter node sequences included multiple detection parameter nodes, including high-frequency pipeline potential, pipeline coating status data, and soil physicochemical parameters. A corrosion rate correlation model is constructed using multiple detection parameter node sequences and pre-built machine learning methods; The corrosion pipeline evaluation module is used to obtain a steady-state corrosion rate sequence by using multiple detection parameter node sequences and corrosion rate correlation models. Based on the preset classification threshold and steady-state corrosion rate sequence, the module classifies the corrosion risk level of high-risk pipeline sections in oil and gas pipelines and obtains an oil and gas pipeline corrosion risk evaluation report.

Images