Method and system for multi-modal quantitative evaluation of the degree of stray current corrosion of oil and gas pipelines

Abstract

The application discloses a kind of oil and gas pipeline stray current corrosion degree multi-modal quantitative evaluation method and system, including real-time acquisition of stray current parameter of oil and gas pipeline, potential gradient distribution and magnetic field intensity data;Dynamic current density characteristics are extracted from stray current parameter, and corrosion hotspot area is constructed according to dynamic current density characteristics and potential gradient distribution;Based on corrosion hotspot area, fusion magnetic field intensity data and soil resistivity parameter, calculate stray current corrosion intensity index;Based on the stray current corrosion intensity index and environmental variable of history, establish corrosion rate prediction model, input the stray current corrosion intensity index calculated currently into corrosion rate prediction model, output the corrosion degree and remaining safe life of pipeline.The application can comprehensively dynamic current density characteristics, potential gradient distribution and multi-modal information such as soil resistivity, accurately identify corrosion hotspot area, improve the accuracy of oil and gas pipeline stray current corrosion degree evaluation result.

Description

Technical Field

[0001] This invention relates to the field of oil and gas pipeline operation and maintenance technology, and in particular to a multimodal quantitative assessment method and system for the degree of stray current corrosion in oil and gas pipelines. Background Technology

[0002] Stray currents pose a persistent and serious threat to the corrosion of oil and gas pipelines, and current methods for addressing this issue have several shortcomings. Firstly, traditional monitoring methods struggle to comprehensively and in real-time acquire stray current parameters, potential gradient distribution, and magnetic field strength data. Previous methods often relied on single-point or limited, scattered detection points, failing to provide a holistic understanding of the pipeline's condition and easily overlooking critical corrosion-prone areas. Secondly, in corrosion assessment, most methods rely on only a single or limited set of parameters, lacking comprehensive analysis and fusion of multimodal data. For example, judging stray current strength solely based on the positive shift of the pipe-to-ground potential from the natural potential or the soil potential gradient is a one-sided approach that fails to accurately reflect the actual corrosion status of the pipeline, leading to biased identification of corrosion hotspots and inaccurate calculation of the stray current corrosion intensity index. Furthermore, existing technologies for predicting pipeline corrosion rates and remaining safe life largely depend on simple empirical models or considerations of single environmental factors, resulting in unreliable predictions and failing to provide a scientifically sound basis for pipeline maintenance. Summary of the Invention

[0003] To address at least one of the aforementioned technical problems, this invention provides a multimodal quantitative assessment method and system for the degree of stray current corrosion in oil and gas pipelines.

[0004] In a first aspect, the present invention provides a multimodal quantitative assessment method for the degree of stray current corrosion in oil and gas pipelines, characterized in that the method includes:

[0005] Stray current parameters, potential gradient distribution, and magnetic field strength data of oil and gas pipelines are collected in real time using a multimodal sensor array.

[0006] Dynamic current density characteristics are extracted from stray current parameters, and corrosion hotspot regions are constructed based on dynamic current density characteristics and potential gradient distribution;

[0007] Based on corrosion hotspots, the stray current corrosion intensity index is calculated by integrating magnetic field strength data and soil resistivity parameters.

[0008] A corrosion rate prediction model is established based on historical stray current corrosion intensity indices and environmental variables. The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion degree and remaining safe life of the pipeline are output.

[0009] Preferably, the real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data of oil and gas pipelines via a multi-modal sensor array includes:

[0010] Potential gradient probes, Hall current sensors, and magnetic field strength detection units are deployed at equal intervals along the axial direction of the oil and gas pipeline to form a distributed node;

[0011] A synchronous sampling module is set up to synchronously collect data on current intensity, potential difference, and magnetic field intensity of each distributed node at a fixed frequency.

[0012] All collected data are labeled with spatiotemporal coordinates to construct a three-dimensional data matrix, where the spatiotemporal coordinates are determined by the pipeline location, detection time, and burial depth parameters.

[0013] Preferably, the step of constructing corrosion hotspot regions based on dynamic current density characteristics and potential gradient distribution includes:

[0014] Based on the dynamic current density characteristics, a sliding window algorithm is used to identify the peak current density region and mark it as the initial hot spot region;

[0015] The potential offset of the initial hot spot area is calculated based on the potential gradient distribution, and areas that meet the potential offset threshold are selected as candidate corrosion areas.

[0016] The soil resistivity of the candidate corrosion zone is obtained, and the current density, potential gradient distribution and soil resistivity of the candidate corrosion zone are normalized by a weighted fusion algorithm to generate a corrosion intensity thermogram.

[0017] Regions in the corrosion intensity thermogram where the intensity value exceeds the critical value are marked as corrosion hotspots.

[0018] Preferably, the calculation of the stray current corrosion intensity index includes:

[0019] ;

[0020] In the formula, The stray current corrosion intensity index is used. The environmental corrosion sensitivity coefficient is dimensionless and ranges from 0.5 to 3. The maximum current density in the corrosion hotspot region, in units of ; It is a soil infiltration correction factor, dimensionless, with a value range of 0.8 to 2.5; The equivalent penetration depth of stray current, in units of ; Soil resistivity, unit: .

[0021] Preferably, the corrosion rate prediction model based on historical stray current corrosion intensity index and environmental variables includes:

[0022] Historical stray current corrosion intensity indices and corresponding environmental variable datasets were obtained as training samples. The environmental variables included humidity, temperature, soil pH, and alternating current frequency.

[0023] Based on the training samples, a multivariate regression model is trained using the random forest algorithm until the model converges, generating a corrosion rate prediction model.

[0024] Preferably, the currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the output includes the corrosion degree and remaining safe life of the pipeline:

[0025] The calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion rate prediction value is output. The corrosion degree of the pipeline is determined based on the corrosion rate prediction value.

[0026] Pipeline wall thickness degradation curves are constructed based on corrosion rate predictions, and current pipeline wall thickness detection data are obtained.

[0027] Based on the pipe wall thickness degradation curve and current pipe wall thickness detection data, the probability distribution of remaining safe life is generated through Monte Carlo simulation.

[0028] Preferably, the method further includes real-time monitoring of abrupt changes in stray current parameters, triggering a tiered early warning if the changes exceed a preset threshold, including:

[0029] The coefficient of variation of stray current parameters is calculated in real time. If the coefficient of variation exceeds the first threshold, a first-level early warning is triggered and a local detection command is generated.

[0030] If the current density exceeds the second threshold continuously within a preset time, a level two warning will be triggered and the emergency current diversion device will be activated.

[0031] When both the corrosion intensity index and the predicted remaining safe life exceed the safe range, a level 3 warning is triggered and a pipeline replacement recommendation is sent.

[0032] Secondly, the present invention also provides a multimodal quantitative assessment system for the degree of stray current corrosion in oil and gas pipelines, the system comprising:

[0033] The oil and gas pipeline data acquisition module is used to acquire stray current parameters, potential gradient distribution and magnetic field strength data of oil and gas pipelines in real time through a multi-modal sensor array.

[0034] The corrosion hotspot region construction module is used to extract dynamic current density characteristics from stray current parameters and construct corrosion hotspot regions based on dynamic current density characteristics and potential gradient distribution.

[0035] The corrosion intensity index calculation module is used to calculate the stray current corrosion intensity index based on corrosion hotspot areas, by integrating magnetic field strength data and soil resistivity parameters.

[0036] The remaining safe life assessment module is used to establish a corrosion rate prediction model based on historical stray current corrosion intensity index and environmental variables. The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion degree and remaining safe life of the pipeline are output.

[0037] Thirdly, the present invention also provides an electronic device including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs a method as described in the first aspect above and any possible implementation thereof.

[0038] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor of an electronic device, cause the processor to perform a method as described in the first aspect above and any possible implementation thereof.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] This invention proposes a multimodal quantitative assessment method for stray current corrosion in oil and gas pipelines, effectively solving the aforementioned technical challenges. By deploying potential gradient probes, Hall current sensors, and magnetic field strength detection units at equal intervals along the pipeline axis, and constructing a three-dimensional data matrix using a synchronous sampling module, comprehensive and real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data is achieved. During the construction of corrosion hotspot areas, multimodal information such as dynamic current density characteristics, potential gradient distribution, and soil resistivity is integrated. Sliding window algorithms, potential offset filtering, weighted fusion algorithms, and corrosion intensity heatmaps are used to accurately identify corrosion hotspot areas. When calculating the stray current corrosion intensity index, multiple factors such as environmental corrosion sensitivity coefficients and soil permeability correction factors are fully considered, significantly improving the accuracy of index calculation. By constructing a corrosion rate prediction model based on historical data and a random forest algorithm trained with multiple environmental variables, and inputting the current stray current corrosion intensity index into the model, the corrosion degree and remaining safe life of the pipeline can be quickly and accurately output. Furthermore, this invention also has the function of real-time monitoring of abrupt changes in stray current parameters and providing graded early warnings, greatly improving the safety and reliability of the pipeline.

[0041] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the accompanying drawings used in the embodiments of the present invention or the background art will be described below.

[0043] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the specification, serve to illustrate the technical solutions of this disclosure.

[0044] Figure 1 This is a flowchart illustrating a multimodal quantitative assessment method for stray current corrosion degree in oil and gas pipelines, provided in an embodiment of the present invention.

[0045] Figure 2 for Figure 1 A flowchart illustrating the sub-steps of step S10;

[0046] Figure 3 for Figure 1 A flowchart illustrating the sub-steps of step S20;

[0047] Figure 4 for Figure 1 A flowchart illustrating a sub-step of step S40;

[0048] Figure 5 for Figure 1 A flowchart illustrating another seed step in step S40;

[0049] Figure 6 A flowchart illustrating a sub-step of step S50 in another multi-modal quantitative assessment method for stray current corrosion degree in oil and gas pipelines provided in an embodiment of the present invention.

[0050] Figure 7 This is a schematic diagram of a multimodal quantitative assessment system for stray current corrosion degree in oil and gas pipelines, provided in an embodiment of the present invention. Detailed Implementation

[0051] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0053] Please see Figure 1 , Figure 1 This is a flowchart illustrating a multimodal quantitative assessment method for stray current corrosion levels in oil and gas pipelines, provided as an embodiment of the present invention. Figure 1 As shown, a multimodal quantitative assessment method for stray current corrosion degree in oil and gas pipelines includes:

[0054] S10. Real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data of oil and gas pipelines through a multi-modal sensor array;

[0055] In the operation of oil and gas pipelines, stray current parameters, potential gradient distribution, and magnetic field strength data are important indicators for assessing pipeline safety, corrosion risk, and electromagnetic interference.

[0056] Stray current parameters refer to currents that are not part of the design path, such as those from high-voltage transmission lines, rail transit systems, other buried metal facilities, or changes in the geomagnetic field, leaking through soil or pipelines. These currents can cause electrochemical corrosion of pipeline metals (stray current corrosion). By measuring stray currents, pipeline corrosion risk can be assessed, cathodic protection system design optimized, and drainage measures implemented. Key parameters may include:

[0057] Current intensity: The magnitude of stray current (unit: ampere, A) directly affects the corrosion rate.

[0058] Current direction: The direction in which current flows into or out of the pipe, used to locate areas at risk of corrosion.

[0059] Fluctuation characteristics: The time-varying nature of current (such as DC, AC, or pulsed current) affects the corrosion mechanism.

[0060] Frequency (e.g., AC stray current): High-frequency currents may cause stronger alternating electromagnetic interference.

[0061] Potential gradient distribution is the change in potential per unit distance (unit: mV / m) on the surface of a pipeline or in the surrounding soil, reflecting the distribution of current at the pipeline-soil interface or in the soil. Key scenarios include: Cathodic protection effectiveness: Measuring the pipeline's potential to ground determines whether cathodic protection meets standards. Corrosion hotspot location: An abnormally high potential gradient may indicate coating damage or current leakage points. Near-surface potential gradient: Dangerous step voltages may occur near high-voltage transmission lines. Measurement methods typically involve using reference electrodes placed along the pipeline or perpendicular to the pipeline direction, combined with numerical model analysis of the current path.

[0062] Magnetic field strength reflects the magnetic field intensity around a pipeline generated by stray currents or external electromagnetic sources (such as high-voltage lines or electrified railways). Magnetic field anomalies can indirectly reflect the presence and distribution of stray currents. By dynamically monitoring the time-varying characteristics of the magnetic field, the risk of AC corrosion or the impact of electromagnetic interference on pipelines can be assessed. This can often be done using magnetometers (such as Hall effect sensors) or electromagnetic field mapping systems, combined with geographic information systems to analyze the spatial distribution of the magnetic field.

[0063] Stray current parameters and potential gradient distribution are directly related to pipeline corrosion risk, while magnetic field strength data can help locate interference sources. Anomalies in the potential gradient may indicate coating damage or step voltage risk; therefore, combining these three factors allows for a comprehensive analysis of corrosion mechanisms (such as DC corrosion and AC corrosion) and the effectiveness of protective measures.

[0064] Thus, step S10 enables the comprehensive and real-time acquisition of multimodal data related to oil and gas pipelines, forming an overall perception of the pipeline's condition, avoiding the omission of key corrosion hazard areas, and laying the foundation for subsequent accurate assessment of corrosion levels.

[0065] See Figure 2 In one embodiment, the real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data of the oil and gas pipeline via a multimodal sensor array includes:

[0066] S101, along the axial direction of the oil and gas pipeline, potential gradient probes, Hall current sensors and magnetic field strength detection units are deployed at equal intervals to form a distributed node.

[0067] First, plan the deployment spacing. Based on factors such as the length and diameter of the oil and gas pipeline, and the complexity of the surrounding environment, determine a reasonable deployment spacing. For example, for long-distance pipelines in relatively stable environments, the spacing can be appropriately increased; while for pipelines passing through complex geological areas or near electrical facilities, the spacing should be appropriately reduced, generally ranging from tens to hundreds of meters. Then, when installing the sensors, install the potential gradient probe, Hall current sensor, and magnetic field strength detection unit sequentially along the axial direction of the oil and gas pipeline according to the planned spacing. During installation, ensure good contact between the sensors and the pipeline to avoid inaccurate data acquisition due to improper installation. Simultaneously, number and record the location of each sensor for subsequent data management and analysis. To facilitate detection, further construct a distributed node: treat the potential gradient probe, Hall current sensor, and magnetic field strength detection unit installed in the same location as a distributed node. These nodes are connected via wired or wireless communication to form a complete sensor network.

[0068] In this way, by deploying multiple sensors at equal intervals, comprehensive monitoring along the oil and gas pipeline can be achieved, acquiring stray current parameters, potential gradient distribution, and magnetic field strength data at different locations, thus avoiding monitoring blind spots. The distributed node setup ensures that the data from each sensor is associated with a specific location, facilitating accurate assessment of the condition of different parts of the pipeline and providing precise location information for subsequently identifying potential corrosion areas.

[0069] S102. Set up a synchronous sampling module to synchronously collect current intensity, potential difference and magnetic field intensity data of each distributed node at a fixed frequency.

[0070] Based on the scale of the sensor network and data acquisition requirements, select a suitable synchronous sampling module. This module should have high-precision clock synchronization and multi-channel data acquisition capabilities to ensure consistent data acquisition times across all distributed nodes. Set a fixed sampling frequency based on the stray current's changing characteristics and monitoring needs. For example, for rapidly changing stray currents, the sampling frequency can be appropriately increased to capture their dynamic changes; while for relatively stable conditions, the sampling frequency can be appropriately decreased. Generally, the sampling frequency can be set to a few times per second to a few times per minute.

[0071] After the synchronous sampling module is started, it simultaneously collects the current intensity, potential difference and magnetic field intensity data of each distributed node according to the set sampling frequency, and temporarily stores the collected data in the local memory.

[0072] Synchronous sampling ensures the temporal consistency of data across distributed nodes, making data from different locations comparable and facilitating subsequent comprehensive analysis and processing. Fixed-frequency sampling enables real-time tracking of stray current parameters, potential differences, and magnetic field strength changes, allowing for the timely detection of abnormal fluctuations and providing a basis for timely warnings and interventions.

[0073] S103. Mark all collected data according to spatiotemporal coordinates to construct a three-dimensional data matrix, wherein the spatiotemporal coordinates are determined by the pipeline location, detection time and burial depth parameters.

[0074] Specifically, when marking spatiotemporal coordinates, after acquiring the collected data from the synchronous sampling module, pipeline location information is added to each data point based on the sensor's number and location record. Simultaneously, the data acquisition time is recorded, and combined with the sensor's burial depth parameters, complete spatiotemporal coordinates are assigned to each data point. The data marked with spatiotemporal coordinates is then organized and arranged according to certain rules to construct a three-dimensional data matrix. The three dimensions of the matrix correspond to the pipeline location, detection time, and burial depth parameters, respectively. Each matrix element stores stray current parameters, potential differences, or magnetic field strength data at the corresponding spatiotemporal coordinates. Constructing a three-dimensional data matrix makes the collected data clearer, allowing for convenient querying and extraction of data from different locations, times, and burial depths. This provides a more comprehensive understanding of the oil and gas pipeline's condition and a reliable data foundation for accurately assessing the degree of corrosion.

[0075] S20. Extract dynamic current density characteristics from stray current parameters, and construct corrosion hotspot regions based on dynamic current density characteristics and potential gradient distribution;

[0076] Existing technologies typically rely on single or a few parameters for assessment, often using static data, such as focusing only on the positive shift of the pipe-to-soil potential from the natural potential or the value of the soil potential gradient at a fixed moment. Therefore, when identifying corrosion hotspots, they may lack systematicity and precision, easily leading to misjudgments or omissions. Stray current is a significant factor contributing to corrosion in oil and gas pipelines, and current density directly reflects the strength of its corrosive effect. Dynamic current density characteristics can demonstrate the change of current over time, providing a more accurate reflection of the actual process of stray current action compared to static current values. Potential gradient distribution reflects the distribution of the electric field around the pipeline and is closely related to the flow of stray current. By analyzing dynamic current density characteristics and potential gradient distribution, it is possible to more accurately pinpoint which areas of the pipeline are most affected by stray current and prone to corrosion.

[0077] Therefore, by comprehensively considering the dynamic current density characteristics and potential gradient distribution, the corrosive effect of stray currents on pipelines can be reflected more comprehensively and accurately, thereby improving the accuracy of pipeline corrosion assessment. Subsequent calculations and predictions based on accurate corrosion hotspot areas also yield more reliable results.

[0078] See Figure 3 Furthermore, in one embodiment, constructing the corrosion hotspot region based on dynamic current density characteristics and potential gradient distribution includes:

[0079] S201. Based on the dynamic current density characteristics, a sliding window algorithm is used to identify the peak current density region and mark it as the initial hot spot region.

[0080] First, a suitable window size is selected based on the characteristics of the dynamic current density data and the actual pipeline conditions. For example, if the data fluctuates frequently, the window can be set smaller; if the data is relatively stable, the window can be appropriately increased. Then, a sliding window traversal is performed: the sliding window slides across the dynamic current density data sequence, calculating the average or maximum current density within the window each time. When the average or maximum current density within the window exceeds a preset threshold, the area corresponding to that window is marked as a current density peak area, i.e., the initial hot spot area. The sliding window algorithm can promptly capture the peak current density in dynamically changing data, marking these peak areas as initial hot spot areas, which can initially locate areas with potential corrosion risks. Considering the dynamic characteristics of stray currents, this algorithm can adapt to changes in current density over time, improving the accuracy of identifying potential corrosion areas.

[0081] S202. Calculate the potential offset of the initial hot spot area based on the potential gradient distribution, and select areas that meet the potential offset threshold as candidate corrosion areas.

[0082] For each initial hot spot region, its potential offset relative to a reference potential is calculated based on the potential gradient distribution of that region. The reference potential can be the pipeline's natural potential or a stable potential value. A potential offset threshold is set based on experience or historical data. Initial hot spot regions with potential offsets exceeding the threshold are selected as candidate corrosion zones. By combining the potential offset calculation with the potential gradient distribution, the initial hot spot regions are further filtered, eliminating some regions with only high current density but no significant potential offset, thus improving the accuracy of candidate corrosion zones.

[0083] S203. Obtain the soil resistivity of the candidate corrosion zone, and use a weighted fusion algorithm to normalize the current density, potential gradient distribution and soil resistivity of the candidate corrosion zone to generate a corrosion intensity heat map.

[0084] When acquiring soil resistivity, the soil resistivity of candidate corrosion zones was measured using a soil resistivity measuring instrument. The current density, potential gradient distribution, and soil resistivity of the candidate corrosion zones were normalized to convert data with different dimensions into a unified scale, facilitating subsequent fusion calculations. Furthermore, during weighted fusion, different weights were assigned to current density, potential gradient distribution, and soil resistivity, and a weighted fusion algorithm was used to calculate the comprehensive corrosion intensity value for each candidate corrosion zone. Finally, based on the comprehensive corrosion intensity value, a corrosion intensity heatmap was generated using a visualization tool, with different colors representing different corrosion intensities.

[0085] S204. Mark the areas in the corrosion intensity thermogram where the intensity value exceeds the critical value as corrosion hot spots.

[0086] A critical corrosion intensity value is set based on factors such as the pipe material and operating environment. In the corrosion intensity heat map, areas with intensity values ​​exceeding the critical value are marked as the final corrosion hotspots.

[0087] This embodiment uses a weighted fusion of three factors—current density, potential gradient distribution, and soil resistivity—to more comprehensively and accurately assess the corrosion degree of candidate corrosion zones. The corrosion intensity heatmap visually displays the corrosion intensity distribution of each candidate corrosion zone. By setting intensity thresholds, it is possible to precisely locate the truly critical corrosion hotspots from the candidate corrosion zones, providing clear targets for subsequent protection and maintenance work.

[0088] S30. Based on corrosion hotspots, the stray current corrosion intensity index is calculated by integrating magnetic field strength data and soil resistivity parameters.

[0089] Preferably, the calculation of the stray current corrosion intensity index includes:

[0090] ;

[0091] In the formula, The stray current corrosion intensity index is used. The environmental corrosion sensitivity coefficient is dimensionless and ranges from 0.5 to 3. The maximum current density in the corrosion hotspot region, in units of ; It is a soil infiltration correction factor, dimensionless, with a value range of 0.8 to 2.5; The equivalent penetration depth of stray current, in units of ; Soil resistivity, unit: .

[0092] Stray current corrosion is a complex process influenced by a combination of factors. In corrosion hotspots, the distribution of stray currents, soil environment, and magnetic fields work together to cause pipeline corrosion.

[0093] In the above formula, current is the direct factor initiating corrosion, and the maximum current density in the corrosion hotspot region is... The higher the value, the stronger the stray current in the area, and the greater the driving force for pipeline corrosion. Soil resistivity This reflects the soil's electrical conductivity. Lower soil resistivity allows stray currents to flow more easily through the soil, increasing the likelihood of pipe corrosion. The equivalent penetration depth of stray current is also relevant. This indicates the effective depth to which stray currents can affect the pipeline. A greater penetration depth means a wider range of stray current influence on the pipeline, and a correspondingly increased risk of corrosion. Environmental corrosion sensitivity coefficient. The influence of surrounding environmental factors on corrosion, such as humidity, temperature, and soil pH, was considered. The sensitivity of pipelines to stray current corrosion varies under different environmental conditions, and this coefficient can correct for corrosion patterns in different environments. Soil permeability correction factor. The influence of soil physical properties on stray current infiltration, such as soil porosity and particle size, was considered. Different types of soil have varying abilities to absorb stray currents, and this factor can be used to adjust for soil characteristics. By integrating these factors in calculations, the corrosion intensity of pipelines caused by stray currents can be assessed more comprehensively and accurately.

[0094] Specifically, corrosion hotspot regions are constructed based on the dynamic current density characteristics and potential gradient distribution, as described in the preceding steps. Maximum current density. Soil resistivity can be extracted from data collected by a multimodal sensor array. The equivalent penetration depth of stray current can then be obtained by measuring the soil in corrosion hotspot areas using a soil resistivity measuring instrument. The equivalent penetration depth can be determined through theoretical calculations, numerical simulations, or field tests. For example, based on the distribution patterns of stray currents and the physical properties of the soil, numerical simulation software can be used to calculate the equivalent penetration depth. The environmental corrosion sensitivity coefficient... and soil permeability correction factor Each has its own range of values, and an appropriate value can be selected within the given range based on the actual environmental conditions and soil characteristics. and Value. For example,

[0095] Regarding the environmental corrosion sensitivity coefficient The possible values ​​of:

[0096] Low-corrosion environments (such as dry, neutral soil): ;

[0097] Moderately corrosive environments (such as moist, weakly acidic / alkaline soils): ;

[0098] Highly corrosive environments (such as high-salt, strong acid / alkali, or industrially polluted areas): ;

[0099] Regarding soil permeability correction factors The possible values ​​of:

[0100] Low-permeability soils (such as clay, high water content): ;

[0101] Medium-permeability soils (such as loam and sandy loam): ;

[0102] Highly permeable soils (such as sandy soils and gravel layers): ;

[0103] Therefore, this embodiment comprehensively considers multiple factors affecting stray current corrosion, avoiding the limitations of single-factor assessment, and can more comprehensively and accurately reflect the corrosion intensity of pipelines in actual environments. The calculated corrosion intensity index allows for a quantitative comparison of corrosion risks in different corrosion hotspot areas. Based on the index value, targeted protective measures can be taken, such as strengthening the anti-corrosion coating in areas with high corrosion intensity indices and installing drainage devices, thus improving the effectiveness and economy of the protective measures.

[0104] S40. Establish a corrosion rate prediction model based on historical stray current corrosion intensity index and environmental variables. Input the currently calculated stray current corrosion intensity index into the corrosion rate prediction model and output the corrosion degree and remaining safe life of the pipeline.

[0105] The stray current corrosion intensity index is influenced by factors such as the environmental corrosion sensitivity coefficient, maximum current density, soil permeability correction factor, stray current equivalent penetration depth, and soil resistivity. Simultaneously, environmental variables such as humidity, temperature, soil pH, and alternating current frequency also affect the corrosion process. These factors are intertwined, making it difficult to accurately predict the corrosion rate through simple observation or single-factor analysis. Therefore, a model that comprehensively considers multiple factors is needed to predict the corrosion rate.

[0106] See Figure 4 In one embodiment, the establishment of a corrosion rate prediction model based on historical stray current corrosion intensity indices and environmental variables includes:

[0107] S401 acquires historical stray current corrosion intensity index and corresponding environmental variable dataset as training samples, the environmental variables including humidity, temperature, soil pH value and alternating current frequency;

[0108] Historical stray current corrosion intensity index data are extracted from the long-term monitoring system of oil and gas pipelines. Simultaneously, corresponding environmental variable data, including humidity, temperature, soil pH, and alternating current frequency, are collected. This data can come from various sensors distributed along the pipeline, such as temperature and humidity sensors, soil pH meters, and current monitors. The collected data undergoes preliminary processing to remove obviously erroneous or anomalous data points, such as extreme values ​​caused by sensor malfunctions or data that does not conform to actual physical laws. The processed stray current corrosion intensity index and environmental variable data are paired to form a complete dataset. Each dataset contains the stray current corrosion intensity index along with the corresponding humidity, temperature, soil pH, and alternating current frequency. This dataset is divided into training and testing samples, typically in an 8:2 or 7:3 ratio. The training samples are used for model training, and the testing samples are used to evaluate model performance.

[0109] S402. Based on the training samples, a multivariate regression model is trained using the random forest algorithm until the model converges, and then a corrosion rate prediction model is generated.

[0110] The random forest algorithm was chosen to construct the multivariate regression model. Random forest is an ensemble learning method consisting of multiple decision trees. Environmental variables (humidity, temperature, soil pH, alternating current frequency) from the training samples were used as input features, and stray current corrosion intensity index was used as the target variable. These were fed into the random forest regressor for training. During training, model parameters were continuously adjusted, and test samples were used to evaluate model performance. Commonly used evaluation metrics include mean squared error (MSE) and root mean square error (RMSE). The model was considered converged when the evaluation metrics no longer showed significant improvement after multiple iterations. For example, a small error threshold was set, and training stopped when the change in mean squared error over several consecutive iterations was less than this threshold. Once the model converged, the resulting random forest regression model became the corrosion rate prediction model.

[0111] The corrosion rate of oil and gas pipelines is influenced by a combination of stray current corrosion intensity index and multiple environmental variables, which exhibit complex nonlinear relationships. The Random Forest algorithm can automatically capture these complex nonlinear relationships without prior assumptions about the specific functional forms between variables, thus fitting the data more accurately and improving the model's prediction accuracy. By integrating multiple decision trees, the Random Forest algorithm reduces the risk of overfitting from individual decision trees and possesses strong generalization ability. Furthermore, in practical applications, sensor-collected data may contain noise and missing data; the Random Forest algorithm maintains relatively stable prediction performance even under these conditions, improving the model's reliability.

[0112] See Figure 5 In one embodiment, the currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, which outputs the corrosion degree and remaining safe life of the pipeline, including:

[0113] S403. Input the currently calculated stray current corrosion intensity index into the corrosion rate prediction model, output the corrosion rate prediction value, and determine the current corrosion degree of the pipeline based on the corrosion rate prediction value.

[0114] The calculated stray current corrosion intensity index, along with corresponding environmental variable data (humidity, temperature, soil pH, and alternating current frequency), are used as inputs to the previously trained corrosion rate prediction model. The corrosion rate prediction model calculates based on the input data and outputs the predicted corrosion rate for the pipeline.

[0115] When determining the degree of corrosion, it is common practice to pre-define the corrosion severity levels corresponding to different corrosion rate ranges, such as mild corrosion, moderate corrosion, and severe corrosion. The predicted corrosion rate values ​​are then compared with these pre-defined ranges to determine the current corrosion severity level of the pipeline.

[0116] S404. Construct a pipe wall thickness degradation curve based on the corrosion rate prediction value and obtain the current pipe wall thickness detection data;

[0117] Pipeline wall thickness degradation curves visually demonstrate the trend of pipeline wall thickness changes over time, helping managers understand the aging status of pipelines in advance. Assuming the pipeline corrosion process is continuous, the curve is plotted with time on the horizontal axis and pipeline wall thickness on the vertical axis. Based on the predicted corrosion rate and the initial wall thickness, the pipeline wall thickness at future time points is calculated at regular intervals, such as annually or semi-annually, thus creating the pipeline wall thickness degradation curve. Then, non-destructive testing techniques, such as ultrasonic thickness gauges, are used to actually measure the current pipeline wall thickness, obtaining accurate data on the current pipe wall thickness.

[0118] S405. Based on the pipe wall thickness degradation curve and the current pipe wall thickness detection data, generate the remaining safe life probability distribution through Monte Carlo simulation.

[0119] A critical wall thickness value related to pipeline safety is determined; that is, when the pipeline wall thickness is below this value, the pipeline cannot guarantee safe operation. Meanwhile, considering the uncertainty of the corrosion rate, a probability distribution function for the corrosion rate is set, such as a normal distribution or a log-normal distribution, whose parameters can be estimated based on historical data or experience.

[0120] Furthermore, the Monte Carlo simulation process is as follows:

[0121] Random sampling: Randomly select a large number of corrosion rate samples from the set corrosion rate probability distribution function.

[0122] Simulation calculation: For each extracted corrosion rate sample, combined with the current pipeline wall thickness detection data and pipeline wall thickness degradation curve, the time required for the pipeline wall thickness to drop to the critical wall thickness is calculated, i.e., the remaining safe life.

[0123] Statistical analysis: Repeat the above sampling and calculation process multiple times (e.g., thousands or even tens of thousands of times) to obtain a large number of remaining safe lifetime samples. Perform statistical analysis on these samples, plot a probability distribution histogram of the remaining safe lifetime, and thus obtain the probability distribution of the remaining safe lifetime.

[0124] Therefore, this embodiment obtains a predicted corrosion rate by inputting the current stray current corrosion intensity index into the model, and determines the degree of corrosion accordingly. This allows for timely and accurate understanding of the pipeline's current corrosion status, providing a basis for taking appropriate protective measures. The pipeline wall thickness degradation curve visually demonstrates the trend of pipeline wall thickness change over time, helping managers understand the pipeline's aging status in advance. Monte Carlo simulation considers the uncertainty of the corrosion rate and, by generating a remaining safe life probability distribution, provides a more comprehensive and objective assessment of the remaining safe life. Combining the information on corrosion degree and the remaining safe life probability distribution provides a scientific basis for pipeline management and maintenance. This information can be used to optimize resource allocation, avoid over-maintenance or under-maintenance, and improve the economic efficiency and safety of pipeline operation.

[0125] See Figure 6 In one embodiment, the method further includes:

[0126] S50. Real-time monitoring of abrupt changes in stray current parameters; if the changes exceed a preset threshold, a tiered warning system is triggered, including:

[0127] S501. Calculate the coefficient of variation of stray current parameters in real time. If the coefficient of variation exceeds the first threshold, trigger a first-level early warning and generate a local detection command.

[0128] S502. If the current density exceeds the second threshold continuously within a preset time, a level 2 warning will be triggered and the emergency current diversion device will be activated.

[0129] S503. When both the corrosion intensity index and the predicted remaining safe life exceed the safe range, a level 3 warning is triggered and a pipeline replacement recommendation is sent.

[0130] Stray current corrosion intensity index can serve as an important indicator for predicting pipeline corrosion trends and providing early warnings. When the index exceeds a certain threshold, an alarm is issued promptly, reminding staff to take appropriate measures to ensure the safe operation of the pipeline.

[0131] In this embodiment, stray current parameter data is continuously collected, and the coefficient of variation (COP) is calculated in real time according to the formula (COP = standard deviation / mean). A first threshold is preset. When the calculated COP exceeds this threshold, a level one warning is immediately triggered, and the system automatically generates instructions for detailed inspection of a specific area of ​​the pipeline. Current density is monitored in real time, with preset time and a second threshold set. If the current density remains above the second threshold for a preset time, a level two warning is triggered, and an emergency current diversion device is automatically activated to guide the stray current to a safe area. The corrosion intensity index and remaining safe life prediction values ​​are continuously acquired, and their safe ranges are set. When both exceed their safe ranges simultaneously, a level three warning is triggered, and the system automatically pushes a pipeline replacement recommendation to relevant management personnel.

[0132] In this way, changes in stray current parameters can be quickly detected through real-time monitoring, potential dangers can be identified immediately, and early warnings can be issued promptly. A tiered early warning mechanism can take different countermeasures based on the degree of danger, rationally allocate resources, and improve response efficiency. Timely warnings and corresponding measures can prevent pipeline corrosion from worsening, reduce the probability of pipeline damage and safety accidents, and ensure the safety of oil and gas transportation.

[0133] In a preferred embodiment, the method further includes a corrosion protection optimization step:

[0134] The output current and potential distribution of the cathodic protection system are dynamically adjusted according to the corrosion hotspots.

[0135] Based on the spatial distribution of corrosion intensity index, differentiated coating repair schemes are generated, prioritizing the treatment of high-index areas.

[0136] In the aforementioned embodiments, corrosion hotspots have been identified and connected to the intelligent control system of the cathodic protection system. The intelligent control system dynamically adjusts the output current of the cathodic protection system in real time based on the corrosion status of different areas as reflected in the model. For corrosion hotspots, the output current is increased to enhance the cathodic protection effect; for areas with lower corrosion levels, the output current is appropriately reduced to avoid over-protection and energy waste. Simultaneously, by optimizing the anode layout and parameter settings, the potential distribution is adjusted to ensure that the entire protected area receives reasonable and effective cathodic protection. Furthermore, based on the spatial distribution map of the corrosion intensity index, differentiated coating repair schemes can be generated using CAD and optimization algorithms. Areas with high corrosion intensity indices are marked as priority repair areas, and detailed repair plans are developed, including coating material selection, coating thickness requirements, and construction techniques. For areas with low corrosion intensity indices, the coating repair cycle can be appropriately extended or simple maintenance measures can be adopted. Then, a construction team is organized to carry out coating repair work according to the repair plan, and quality monitoring is conducted during the repair process to ensure that the repair effect meets expectations.

[0137] In summary, the multimodal quantitative assessment method for stray current corrosion intensity in oil and gas pipelines proposed in this invention effectively solves the aforementioned technical challenges. By deploying potential gradient probes, Hall current sensors, and magnetic field strength detection units at equal intervals along the axial direction of the oil and gas pipeline, and constructing a three-dimensional data matrix using a synchronous sampling module, comprehensive and real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data is achieved. During the construction of corrosion hotspot areas, multimodal information such as dynamic current density characteristics, potential gradient distribution, and soil resistivity is integrated, and methods such as sliding window algorithm, potential offset screening, weighted fusion algorithm, and corrosion intensity heatmap are used to accurately identify corrosion hotspot areas. When calculating the stray current corrosion intensity index, multiple factors such as environmental corrosion sensitivity coefficient and soil permeability correction factor are fully considered, significantly improving the accuracy of index calculation. By constructing a corrosion rate prediction model based on historical data and a random forest algorithm trained with multiple environmental variables, and inputting the current stray current corrosion intensity index into the model, the corrosion intensity and remaining safe life of the pipeline can be output quickly and accurately. In addition, this invention also has the function of real-time monitoring of abrupt changes in stray current parameters and graded early warning, greatly improving the safety and reliability of the pipeline.

[0138] See Figure 7 In one embodiment, the present invention also provides a multimodal quantitative assessment system for the degree of stray current corrosion in oil and gas pipelines, the system comprising:

[0139] The oil and gas pipeline data acquisition module 100 is used to acquire stray current parameters, potential gradient distribution and magnetic field strength data of oil and gas pipelines in real time through a multi-modal sensor array.

[0140] The corrosion hotspot region construction module 200 is used to extract dynamic current density characteristics from stray current parameters and construct corrosion hotspot regions based on dynamic current density characteristics and potential gradient distribution.

[0141] The corrosion intensity index calculation module 300 is used to calculate the stray current corrosion intensity index based on corrosion hotspot areas, by integrating magnetic field strength data and soil resistivity parameters.

[0142] The Remaining Safe Life Assessment Module 400 is used to establish a corrosion rate prediction model based on historical stray current corrosion intensity indices and environmental variables. The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion degree and remaining safe life of the pipeline are output.

[0143] It is understood that the system provided in this embodiment has functions or includes modules that can be used to execute the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0144] The present invention also provides an electronic device including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs a method as described in any of the above possible implementations.

[0145] The present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor of an electronic device, cause the processor to perform a method as described in any of the above possible implementations.

[0146] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

Claims

1. A multimodal quantitative assessment method for the degree of stray current corrosion in oil and gas pipelines, characterized in that, The method includes: Stray current parameters, potential gradient distribution, and magnetic field strength data of oil and gas pipelines are collected in real time using a multimodal sensor array. Dynamic current density features are extracted from stray current parameters, and corrosion hotspot regions are constructed by fusing magnetic field strength data, dynamic current density features, and potential gradient distribution. Based on corrosion hotspots and soil resistivity parameters, the stray current corrosion intensity index is calculated, including: ； In the formula, The stray current corrosion intensity index is used. The environmental corrosion sensitivity coefficient is dimensionless and ranges from 0.5 to 3. The maximum current density in the corrosion hotspot region, in units of ; It is a soil infiltration correction factor, dimensionless, with a value range of 0.8 to 2.5; The equivalent penetration depth of stray current, in units of ; Soil resistivity, unit: ; A corrosion rate prediction model is established based on historical stray current corrosion intensity indices and environmental variables. The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion degree and remaining safe life of the pipeline are output.

2. The multi-modal quantitative assessment method for stray current corrosion degree in oil and gas pipelines according to claim 1, characterized in that, The method of real-time acquisition of stray current parameters, potential gradient distribution, and magnetic field strength data of oil and gas pipelines through a multi-modal sensor array includes: Potential gradient probes, Hall current sensors, and magnetic field strength detection units are deployed at equal intervals along the axial direction of the oil and gas pipeline to form a distributed node; A synchronous sampling module is set up to synchronously collect data on current intensity, potential difference, and magnetic field intensity of each distributed node at a fixed frequency. All collected data are labeled with spatiotemporal coordinates to construct a three-dimensional data matrix, where the spatiotemporal coordinates are determined by the pipeline location, detection time, and burial depth parameters.

3. The multimodal quantitative assessment method for stray current corrosion degree in oil and gas pipelines according to claim 1, characterized in that, The fusion of magnetic field strength data, dynamic current density characteristics, and potential gradient distribution constructs corrosion hotspot regions, including: Based on the dynamic current density characteristics, a sliding window algorithm is used to identify the peak current density region and mark it as the initial hot spot region; The potential offset of the initial hot spot area is calculated based on the potential gradient distribution, and areas that meet the potential offset threshold are selected as candidate corrosion areas. The soil resistivity of the candidate corrosion zone is obtained, and the current density, potential gradient distribution and soil resistivity of the candidate corrosion zone are normalized by a weighted fusion algorithm to generate a corrosion intensity thermogram. Regions in the corrosion intensity thermogram where the intensity value exceeds the critical value are marked as corrosion hotspots.

4. The multimodal quantitative assessment method for stray current corrosion degree in oil and gas pipelines according to claim 1, characterized in that, The corrosion rate prediction model established based on historical stray current corrosion intensity indices and environmental variables includes: Historical stray current corrosion intensity indices and corresponding environmental variable datasets were obtained as training samples. The environmental variables included humidity, temperature, soil pH, and alternating current frequency. Based on the training samples, a multivariate regression model is trained using the random forest algorithm until the model converges, generating a corrosion rate prediction model.

5. The multimodal quantitative assessment method for stray current corrosion degree of oil and gas pipelines according to claim 1, characterized in that, The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, which outputs the corrosion degree and remaining safe life of the pipeline, including: The calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion rate prediction value is output. The corrosion degree of the pipeline is determined based on the corrosion rate prediction value. Pipeline wall thickness degradation curves are constructed based on corrosion rate predictions, and current pipeline wall thickness detection data are obtained. Based on the pipe wall thickness degradation curve and current pipe wall thickness detection data, the probability distribution of remaining safe life is generated through Monte Carlo simulation.

6. The multimodal quantitative assessment method for stray current corrosion degree in oil and gas pipelines according to claim 1, characterized in that, The method also includes real-time monitoring of abrupt changes in stray current parameters, triggering tiered early warnings if the changes exceed a preset threshold, including: The coefficient of variation of stray current parameters is calculated in real time. If the coefficient of variation exceeds the first threshold, a first-level early warning is triggered and a local detection command is generated. If the current density exceeds the second threshold continuously within a preset time, a level two warning will be triggered and the emergency current diversion device will be activated. When both the corrosion intensity index and the predicted remaining safe life exceed the safe range, a level 3 warning is triggered and a pipeline replacement recommendation is sent.

7. A multimodal quantitative assessment system for stray current corrosion degree in oil and gas pipelines, characterized in that, The system includes: The oil and gas pipeline data acquisition module is used to acquire stray current parameters, potential gradient distribution and magnetic field strength data of oil and gas pipelines in real time through a multi-modal sensor array. The corrosion hotspot region construction module is used to extract dynamic current density features from stray current parameters and fuse magnetic field strength data, dynamic current density features and potential gradient distribution to construct corrosion hotspot regions. The corrosion intensity index calculation module is used to calculate the stray current corrosion intensity index based on corrosion hotspot areas and soil resistivity parameters, including: ； In the formula, The stray current corrosion intensity index is used. The environmental corrosion sensitivity coefficient is dimensionless and ranges from 0.5 to 3. The maximum current density in the corrosion hotspot region, in units of ; It is a soil infiltration correction factor, dimensionless, with a value range of 0.8 to 2.5; The equivalent penetration depth of stray current, in units of ; Soil resistivity, unit: ; The remaining safe life assessment module is used to establish a corrosion rate prediction model based on historical stray current corrosion intensity index and environmental variables. The currently calculated stray current corrosion intensity index is input into the corrosion rate prediction model, and the corrosion degree and remaining safe life of the pipeline are output.

8. An electronic device, characterized in that, include: A processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs the multimodal quantitative assessment method for stray current corrosion degree of oil and gas pipelines as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which includes program instructions that, when executed by a processor of an electronic device, cause the processor to perform the multimodal quantitative assessment method for stray current corrosion degree of oil and gas pipelines as described in any one of claims 1 to 6.

Images