Real-time dynamic monitoring method and system for stray current of buried pipeline

Abstract

This invention discloses a method and system for real-time dynamic monitoring of stray current in buried pipelines, relating to the field of buried pipeline monitoring and protection technology. By accurately identifying the interference characteristics of electrical interference signals in the pipeline environment, and based on the corrosion risk correlation model and interference characteristics, the method assesses the impact of different interference sources on pipeline corrosion, obtains risk assessment results for different interference sources on pipeline corrosion, and finally generates targeted protection parameters based on the risk assessment results and adjusts the cathodic protection system in real time. This enables real-time dynamic monitoring of external interference characteristics in complex environments, accurately identifies its source and its impact on pipeline corrosion, and achieves accurate assessment and dynamic protection of pipeline corrosion risk, effectively improving the safe operation level and service life of buried pipelines.

Description

Technical Field

[0001] This invention relates to the field of buried pipeline monitoring and protection technology, and in particular to a method and system for real-time dynamic monitoring of stray current in buried pipelines. Background Technology

[0002] As a crucial component of urban infrastructure, buried pipelines play a vital role in energy transmission and resource allocation, making their safe operation paramount. However, electrical interference from the external environment often poses a significant corrosion threat to pipelines, severely impacting their service life and reliability. Therefore, researching effective monitoring and protection methods for buried pipelines is particularly urgent.

[0003] Currently, although there are some monitoring methods for pipeline corrosion, most of them suffer from insufficient coverage and delayed response. Especially in complex urban environments, the sources of interference are diverse and dynamic. Existing methods are unable to fully capture the instantaneous characteristics of interference or accurately distinguish the impact of different sources, resulting in a lack of targeted and precise protective measures.

[0004] Against this backdrop, the core challenges facing this field are becoming increasingly apparent. First, the sources of external electrical interference are extremely complex, encompassing various scenarios such as subway operation and industrial equipment. These sources exhibit significant differences in characteristics, with varying interference intensity and duration, increasing the difficulty of identification. The inability to effectively distinguish and locate the characteristics of these sources leads to an inability to establish a clear correlation between interference and pipeline corrosion risk. Consequently, protection strategies often remain superficial and fail to address the root cause of the problem.

[0005] Therefore, how to monitor the characteristics of external interference in complex environments in real time and accurately identify its source and its impact on pipeline corrosion has become a key issue in ensuring the safe operation of buried pipelines. Summary of the Invention

[0006] This invention provides a method and system for real-time dynamic monitoring of stray current in buried pipelines, enabling real-time dynamic monitoring of external interference characteristics in complex environments, accurately identifying its source and its impact on pipeline corrosion, achieving precise assessment and dynamic protection of pipeline corrosion risks, and effectively improving the safe operation level and service life of buried pipelines.

[0007] This invention provides a method for real-time dynamic monitoring of stray current in buried pipelines, comprising:

[0008] Obtain electrical interference signals from the environment surrounding buried pipelines;

[0009] Based on the electrical interference signal, the interference characteristics of the electrical interference signal are determined, wherein the interference characteristics include the interference source, interference intensity, and duration of the electrical interference signal.

[0010] The interference features are input into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and the risk assessment results output by the corrosion risk association model are obtained.

[0011] Based on the risk assessment results, target protection parameters corresponding to the electrical interference signal are generated;

[0012] Based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system, the dynamic protection configuration is determined.

[0013] The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines, and the corrosion risk association model is trained using interference feature datasets corresponding to different interference sources.

[0014] According to the present invention, a real-time dynamic monitoring method for stray current in buried pipelines is provided, wherein the corrosion risk correlation model is obtained through the following method:

[0015] Electrical interference signals are obtained from the environment surrounding the buried pipeline, and voltage and current data are collected using multi-channel sensors to obtain the raw interference dataset.

[0016] The signal spectrum is decomposed using the Fast Fourier Transform algorithm, and the frequency and amplitude features of the original interference dataset are extracted to obtain the interference characteristic dataset.

[0017] Based on the aforementioned interference feature dataset, a subway interference subset and an industrial interference subset are determined;

[0018] Based on the aforementioned subway interference subset and the aforementioned industrial interference subset, the support vector machine algorithm is used to classify the interference intensity and duration to obtain the interference classification results;

[0019] Based on the interference classification results, the data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate, and a corrosion risk association model is obtained.

[0020] The interference classification results include the correlation between the interference source, interference intensity, and duration.

[0021] According to the present invention, a real-time dynamic monitoring method for stray current in buried pipelines is provided, wherein, based on the interference classification results, a data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate to obtain a corrosion risk association model, including:

[0022] Based on the interference classification results and pipeline corrosion rate data, feature values ​​of interference intensity and corrosion rate are extracted from a pre-established database to obtain an initial dataset.

[0023] Principal component analysis is used to extract features from the initial dataset, reducing the data dimensionality and obtaining a dimensionality-reduced feature set.

[0024] The Pearson correlation coefficient method was used to calculate the correlation coefficient between the interference intensity in the dimensionality-reduced feature set and the pipeline corrosion rate, and the correlation coefficient matrix was obtained.

[0025] Based on the correlation coefficient matrix and the interference classification results, the feature pairs of interference intensity and corrosion rate in the dimensionality reduction feature set are marked as highly correlated feature pairs, thus obtaining a highly correlated feature set;

[0026] Based on the highly correlated feature set, a linear regression algorithm is used to train a preset training model to obtain the corrosion risk correlation model.

[0027] According to the present invention, a method for real-time dynamic monitoring of stray current in buried pipelines, wherein determining a subway interference subset and an industrial interference subset based on the interference feature dataset includes:

[0028] If the frequency characteristics in the interference characteristic dataset match the preset subway operating frequency band, then the source of interference is determined to be subway operation through time domain analysis, and a subway interference subset is obtained.

[0029] If the frequency characteristics in the interference characteristic dataset match the preset industrial equipment frequency band, then the source of interference is determined to be industrial equipment through waveform analysis, thus obtaining an industrial interference subset.

[0030] According to the present invention, a method for real-time dynamic monitoring of stray current in buried pipelines, wherein determining the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system includes:

[0031] Based on the target protection parameters, a digital signal processor is used to adjust the current output of the pipeline cathodic protection system to obtain protection optimization parameters.

[0032] By optimizing the protection parameters, the operating status of the pipeline cathodic protection system is updated in real time, and the real-time operating status fed back by the pipeline cathodic protection system in response to the target protection parameters is determined.

[0033] Based on the real-time operating status, determine the dynamic protection configuration.

[0034] According to the present invention, a real-time dynamic monitoring method for stray current in buried pipelines is provided. The corrosion risk correlation model includes a corrosion rate prediction model, a corrosion risk level prediction model, and a corrosion risk outcome prediction model. The method involves inputting the interference features into the corrosion risk correlation model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtaining the risk assessment result output by the corrosion risk correlation model, including:

[0035] The interference features are input into the corrosion rate prediction model to obtain the corrosion rate result output by the corrosion rate prediction model, wherein the corrosion rate result reflects the influence of different interference sources on pipeline corrosion.

[0036] The corrosion rate results are input into the corrosion risk level prediction model to determine the degree of influence of different interference sources on pipeline corrosion risk, and the risk level list output by the corrosion risk level prediction model is obtained. The risk level list is determined by marking interference features whose influence weights exceed a preset threshold as high-risk factors.

[0037] Input the risk level list into the corrosion risk result prediction model to obtain the risk assessment result output by the corrosion risk result prediction model;

[0038] The risk assessment results include the pipeline's corrosion risk level, the time required for pipeline perforation, the pipeline risk trend, and the pipeline protection strategy.

[0039] According to the present invention, a method for real-time dynamic monitoring of stray current in buried pipelines is provided, wherein the corrosion risk prediction model is used for:

[0040] Based on the risk level list, in-depth data mining is conducted to determine the action path of high-risk factors on pipeline corrosion and the corrosion risk change trend corresponding to the action path.

[0041] Based on the corrosion risk change trend and historical data of pipeline status, conditional judgment logic is used to determine the key monitoring objects in the risk level list.

[0042] Based on the key monitoring objects, in-depth data mining is conducted to determine the target action path of the key monitoring objects and the target corrosion risk change trend corresponding to the target action path;

[0043] The risk assessment result is determined based on the target action path and the target corrosion risk change trend;

[0044] If the trend of corrosion risk changes matches the historical high-risk pattern in the historical data, then key monitoring targets are identified.

[0045] This invention also provides a real-time dynamic monitoring system for stray current in buried pipelines, comprising:

[0046] The signal acquisition module is used to acquire electrical interference signals from the environment surrounding the buried pipeline;

[0047] An interference feature extraction module is used to determine the interference features of the electrical interference signal based on the electrical interference signal, wherein the interference features include the interference source, interference intensity, and duration corresponding to the electrical interference signal;

[0048] The result prediction module inputs the interference features into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtains the risk assessment result output by the corrosion risk association model.

[0049] The parameter generation module is used to generate target protection parameters corresponding to the electrical interference signal based on the risk assessment results.

[0050] The dynamic protection configuration module is used to determine the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system.

[0051] The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines, and the corrosion risk association model is trained using interference feature datasets corresponding to different interference sources.

[0052] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the real-time dynamic monitoring method for stray current in buried pipelines as described above.

[0053] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the real-time dynamic monitoring method for stray current in buried pipelines as described above.

[0054] The present invention provides a method and system for real-time dynamic monitoring of stray current in buried pipelines. By accurately identifying the interference characteristics of electrical interference signals in the pipeline environment, and based on the corrosion risk correlation model and interference characteristics, the method assesses the impact of different interference sources on pipeline corrosion, obtains risk assessment results for different interference sources on pipeline corrosion, and finally generates targeted protection parameters based on the risk assessment results and adjusts the cathodic protection system in real time. This enables real-time dynamic monitoring of external interference characteristics in complex environments, accurately identifies its source and its impact on pipeline corrosion, and achieves accurate assessment and dynamic protection of pipeline corrosion risk, effectively improving the safe operation level and service life of buried pipelines. Attached Figure Description

[0055] Figure 1 This is one of the flowcharts illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0056] Figure 2 This is the second flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0057] Figure 3 This is the third flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0058] Figure 4 This is the fourth flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0059] Figure 5 This is the fifth flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0060] Figure 6 This is the sixth flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0061] Figure 7 This is the seventh flowchart illustrating the real-time dynamic monitoring method for stray current in buried pipelines provided in this embodiment of the invention.

[0062] Figure 8 This is a schematic diagram of the structure of the real-time dynamic monitoring system for stray current in buried pipelines provided in an embodiment of the present invention;

[0063] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0064] 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.

[0065] Reference Figure 1 This invention provides a method for real-time dynamic monitoring of stray current in buried pipelines, comprising the following steps:

[0066] Step 100: Obtain electrical interference signals from the environment surrounding the buried pipeline;

[0067] Multi-channel sensors can be installed on buried pipelines to collect real-time electrical interference signals from the surrounding environment. These sensors can then be deployed on corrosion probes to acquire these electrical signals. The electrical interference signal is a physical magnetic field signal within the buried pipeline environment, reflecting the impact of electrical interference on the pipeline. Therefore, predicting the corrosion risk of buried pipelines using electrical interference signals allows for the analysis of the sources of interference, enabling targeted protection at the source.

[0068] Step 200: Based on the electrical interference signal, determine the interference characteristics of the electrical interference signal, wherein the interference characteristics include the interference source, interference intensity, and duration of the electrical interference signal.

[0069] Furthermore, interference features of electrical interference signals can be extracted using the following methods: First, the signal spectrum is decomposed using a Fast Fourier Transform (FFT) algorithm to determine the frequency and amplitude characteristics of the electrical interference signal. Second, based on the frequency and amplitude characteristics, the interference features of the electrical interference signal are determined. This method, which involves collecting electrical interference signals from the environment surrounding buried pipelines using multi-channel sensors, extracting interference features using a FFT algorithm to facilitate subsequent determination of the interference source by matching with a preset frequency band, and classifying the interference using a Support Vector Machine (SVM) algorithm, can improve the accuracy of interference feature identification, enhance the accuracy of interference source identification, and improve the accuracy of pipeline corrosion risk prediction.

[0070] Step 300: Input the interference features into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtain the risk assessment results output by the corrosion risk association model;

[0071] After collecting electrical interference signals from multiple sources, the influencing factors on pipeline corrosion can be identified by extracting the interference features from these signals. These interference features represent the environmental characteristics that contribute to pipeline corrosion. The interference features related to pipeline corrosion can be pre-classified and organized, and missing and outlier values ​​can be cleaned using data preprocessing techniques to obtain preliminary processed interference features. Based on these preliminary processed interference features, a corrosion risk association model can be constructed using linear regression, decision tree, or random forest algorithms. It should be noted that the algorithm used to construct the corrosion risk association model can be selected as needed and is not limited to the algorithm proposed in this embodiment. The corrosion risk association model is used to extract features related to the relationship between interference sources and pipeline corrosion, determine the weight distribution of key influencing factors, and thus quantify the impact of different interference sources through the constructed risk association model, obtaining risk assessment results.

[0072] The risk assessment results can include the pipeline's corrosion risk level, the time required for pipeline perforation, the pipeline risk trend, and pipeline protection strategies. Therefore, risk correlation models can predict the corrosion risk level of interference sources affecting pipeline corrosion, the time required for buried pipeline perforation under current interference factors, whether the pipeline corrosion risk trend is increasing or decreasing under current interference factors, and predict corresponding pipeline protection strategies. These pipeline protection strategies can include electrochemical methods, applying a protective layer to the pipeline exterior, repairing with special materials, or adding a mechanical protective layer.

[0073] Step 400: Based on the risk assessment results, generate the target protection parameters corresponding to the electrical interference signal;

[0074] The risk assessment results of buried pipelines are obtained by using a pre-established corrosion risk correlation model. Then, based on the risk assessment results, the distribution of protection requirements for buried pipelines is determined. According to the distribution of protection requirements, target protection parameters corresponding to electrical interference signals are generated for key areas of pipeline protection. The target protection parameters are the configuration parameters of protection measures or protection devices. Therefore, in this step, the target protection parameters can be input into the protection device so that the protection device can perform the target configuration operation corresponding to the target protection parameters, thereby realizing the corresponding protection operation for buried pipelines based on the risk assessment results.

[0075] For example, based on the risk assessment results, targeted protection parameters are generated to obtain the corresponding target protection parameters. First, the electrical interference signal to be predicted for the pipeline is acquired through a data acquisition system, and the interference characteristics are determined based on the electrical interference signal. These characteristics include a soil resistivity of 1000 ohm-cm, a pipeline potential of -0.85 volts, and a stray current density of 0.5 mA / m². Using a risk assessment algorithm, a weighted scoring model is employed to calculate the risk index. The risk assessment result includes the risk index, and the formula for calculating the risk index is as follows:

[0076]

[0077] Where R is the risk index. Soil resistivity factor For potential deviation, For stray current factor, The aging factor is for the pipeline material. It should be noted that the soil resistivity factor reflects the soil corrosivity. The higher the soil resistivity factor value, the lower the corrosivity. The potential deviation is the deviation between the pipeline potential and the protection potential, which is used to measure the cathodic protection effect. The stray current factor is the stray current intensity, which is commonly found near electrified railways or high-voltage lines. To determine the degree of aging of the pipeline, When the value is 0, the pipe is brand new. When the value is 1, the pipe is fully aged. Assuming the material aging factor is 0.2, the calculated value is R = 400 + 0.3 × |-0.85 - (-0.85)| + 0.2 × 0.5 + 0.1 × 0.2 = 400.12. When the risk index is higher than 350, the protection parameters need to be optimized. Next, a digital signal processor (DSP) is used to adjust the current output of the cathodic protection system. Based on the PI control algorithm, the target protection potential is set to -0.85 volts, the deviation between the actual potential and the target is the input, and the proportional coefficient is... =0.8, integration time =0.5 seconds, calculate the output current adjustment, where the formula for calculating the current adjustment is as follows:

[0078]

[0079] in, For error signals, This is the proportionality coefficient. For deviation, For the integral coefficients, by substituting the above variables into the above formula, we can know... =0.8 / 0.5=1.6. Assuming the current deviation is 0.05 volts, the calculation yields... =0.8×0.05+1.6×0.05×0.5=0.08 Amps. Based on this, the digital signal processor (DSP) adjusts the rectifier output current from 1.5 Amps to 1.58 Amps. The optimized parameters are verified through a real-time monitoring system, which collects the adjusted potential data, for example, -0.86 V, reducing the deviation to 0.01 V, meeting the protection standard. The system records the optimized parameters and stores them in the database, generating a log file containing a timestamp, the adjustment amount of 0.08 Amps, and the new potential of -0.86 V, for subsequent analysis. The entire process is automatically executed by an embedded control system, ensuring a closed-loop logic and closely linking parameter adjustments with risk assessment results.

[0080] Step 500: Determine the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system;

[0081] The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines, and the corrosion risk association model is trained using interference feature datasets corresponding to different interference sources.

[0082] Electrochemical dynamic protection operation guides the real-time control of pipeline cathodic protection systems. It dynamically adjusts the protection of buried pipelines using electrochemical methods based on input dynamic protection configuration parameters or electrochemical parameters fed back from the buried pipeline, preventing pipeline rusting. Electrochemical dynamic protection operation includes sacrificial anode operation and impressed current operation. The principle of sacrificial anode operation is to release a corresponding amount of easily corroded metal around the pipeline according to configuration or electrochemical parameters, using the easily corroded metal as the anode to replace the pipeline in corrosion. Impressed current operation, on the other hand, applies a corresponding current to the pipeline through a power supply device according to configuration or electrochemical parameters, forcing the pipeline to become the cathode and preventing corrosion.

[0083] By collecting real-time data from the pipeline cathodic protection system, the real-time operating status is obtained, such as actual output current / voltage, pipeline potential, and system temperature. The current real-time operating status is analyzed to obtain preliminary status feedback information. Based on this preliminary status feedback, the target protection parameters are analyzed and compared against preset thresholds to determine if any deviations exist, resulting in deviation analysis results. If the deviation analysis results exceed a preset range, the target protection parameters are optimized and adjusted to determine the adjusted parameter values. Then, the operating status of the cathodic protection system is updated using the adjusted parameter values, obtaining updated status feedback information. Based on the updated status feedback information and a real-time monitoring mechanism, the protection effect of the adjusted parameter values ​​is analyzed to obtain effect evaluation information. If the protection effect does not meet expectations, a dynamic configuration mechanism is triggered to adjust the system update strategy and determine the final protection configuration scheme, resulting in a dynamic protection configuration. Therefore, this embodiment can achieve continuous monitoring of the pipeline system's operating status through the final protection configuration scheme, obtaining long-term status feedback data and realizing dynamic real-time protection for buried pipelines.

[0084] The present invention provides a method and system for real-time dynamic monitoring of stray current in buried pipelines. By accurately identifying the interference characteristics of electrical interference signals in the pipeline environment, and based on the corrosion risk correlation model and interference characteristics, the method assesses the impact of different interference sources on pipeline corrosion, obtains risk assessment results for different interference sources on pipeline corrosion, and finally generates targeted protection parameters based on the risk assessment results and adjusts the cathodic protection system in real time. This enables real-time dynamic monitoring of external interference characteristics in complex environments, accurately identifies its source and its impact on pipeline corrosion, and achieves accurate assessment and dynamic protection of pipeline corrosion risk, effectively improving the safe operation level and service life of buried pipelines.

[0085] In one embodiment, please refer to Figure 2 The corrosion risk association model was obtained in the following way:

[0086] Step 301: Obtain electrical interference signals from the environment surrounding the buried pipeline, and use multi-channel sensors to collect voltage and current data to obtain the original interference dataset;

[0087] Electrical interference signals are acquired from the surrounding environment of buried pipelines using multi-channel sensors, collecting voltage and current data to form an initial dataset. Based on this initial dataset, signal preprocessing methods are used to denoise the voltage and current data, resulting in a clean dataset. For this clean dataset, a Fast Fourier Transform (FFT) method is applied to perform frequency domain analysis on the electrical interference signals to determine the main frequency components. If the main frequency components exceed a preset threshold range, further time-domain feature extraction is performed on the clean dataset to obtain time-domain feature parameters. Based on these time-domain feature parameters, a Support Vector Machine (SVM) algorithm is used to classify the interference signals, determining their type and source. Using the classification results, an adaptive filtering method is applied to optimize the clean dataset for different types of interference signals, resulting in a final optimized dataset that serves as the original interference dataset. If the interference signal characteristics in the original interference dataset still exhibit anomalies, a second frequency domain analysis is performed to determine the specific distribution characteristics of the abnormal signals.

[0088] Specifically, to acquire and process electrical interference signals from the environment surrounding the buried pipeline, a multi-channel sensor system is first deployed to collect voltage and current data. Assuming a sensor node is placed every 100 meters along the pipeline, each node is equipped with four channels to collect voltage signals (range 0 to 50V) and current signals (range 0 to 10A), with a sampling frequency of 1000Hz to ensure the capture of high-frequency interference signals, the data is transmitted to a central server via a wireless network, generating approximately 1GB of raw data per hour. Subsequently, the collected raw interference dataset is analyzed in the frequency domain using a Fast Fourier Transform (FFT) algorithm. This decomposes the time-domain signal into frequency components, extracting the main interference frequencies. For example, 50Hz and its harmonics (100Hz, 150Hz) are identified as the main interference sources. The power spectral density shows a peak value of 3.2W at 50Hz, indicating that it may be power frequency interference. Next, wavelet transform was used to denoise the signal. The Daubechies wavelet basis (db4) was selected, with a decomposition level of 5 layers and a soft threshold of 0.05. After filtering out high-frequency noise, the signal-to-noise ratio improved from the initial 12.5 dB to 18.7 dB, verifying the denoising effect. Finally, a support vector machine-based classification model was constructed to identify the interference type, extracting the signal's time-domain features (e.g., mean 0.8V, variance 0.2) and frequency-domain features (e.g., dominant frequency 50Hz). The training dataset contained 5000 samples, and the test set accuracy reached 92.3%, thus determining whether the interference source was a natural factor such as power lines or lightning.

[0089] Step 302: Decompose the signal spectrum using the Fast Fourier Transform algorithm, extract the frequency and amplitude features of the original interference dataset, and obtain the interference characteristic dataset;

[0090] The original interference signal in the original interference dataset is decomposed using Fast Fourier Transform (FFT) to obtain spectral data. Spectral data analysis separates the frequency and amplitude information from the original interference dataset, generating spectral feature vectors. If periodic variations exist in the frequency information, the frequency information in the original interference dataset is normalized to obtain standardized frequency features; otherwise, the original frequency features are retained. The original interference dataset after this processing becomes a standardized dataset. Based on the standardized frequency features in the standardized dataset, principal component analysis (PCA) is used to extract the dominant frequency components and determine key interference features. For the key interference features and their corresponding amplitude information, K-Means clustering is used to classify interference patterns and obtain interference categories. Pattern distribution features are extracted from the interference categories, and the probability distribution between each category is calculated to generate an interference characteristic dataset.

[0091] Specifically, for processing the original interference dataset, a series of information technology methods can be used to decompose the signal spectrum and extract features. The specific implementation method is as follows: Assuming there is an interference signal dataset containing 1000 sampling points at a sampling rate of 10 kHz, the signal is first decomposed into a spectrum using the Fast Fourier Transform (FFT) algorithm. The algorithm used is based on the Cooley-Tukey FFT method, which converts the time-domain signal into a frequency-domain signal and calculates the amplitude and phase information of each frequency component. For example, the calculation reveals a significant peak at 500 Hz with an amplitude of 2.5 V, while the amplitude at 1000 Hz is 1.8 V, indicating that the signal has strong interference components in the low-frequency band. Next, when extracting frequency and amplitude features, an amplitude threshold of 0.5 V is set, and only frequency components with amplitudes higher than this threshold are retained. Finally, 10 main frequency points are selected, among which 500 Hz and 1000 Hz are identified as key interference frequencies. Subsequently, through statistical analysis of these frequency and amplitude data, the energy percentage of each frequency component is calculated, for example, the energy percentage of 500 Hz is 35% and that of 1000 Hz is 20%, thereby constructing an interference characteristic dataset containing a three-dimensional feature matrix of frequency, amplitude, and energy percentage.

[0092] Step 303: Based on the interference feature dataset, determine the subway interference subset and the industrial interference subset;

[0093] After determining the interference feature dataset, spectral analysis is performed on it to divide it into a subway interference subset and an industrial interference subset. The principle behind this step is that the frequency bands corresponding to the data in the subway and industrial interference subsets are different. Therefore, by analyzing the differences in their spectra, the frequency bands to which each data point in the interference feature dataset belongs are analyzed, thus distinguishing between the subway and industrial interference subsets.

[0094] The step of determining the subway interference subset and the industrial interference subset based on the interference feature dataset includes:

[0095] a) Frequency domain discrimination

[0096] For each set of frequency-amplitude features in the interference characteristic dataset, perform the following judgment:

[0097] If its main frequency Satisfying 10Hz≤ ≤80Hz, and second harmonic 2 amplitude With the amplitude of the main frequency The ratio 0.2 ≤ If the value is ≤0.6, it is marked as a "subway candidate";

[0098] If its main frequency Satisfying 40Hz≤ ≤120 Hz, and 50 Hz energy percentage The proportion of total energy E If the value is ≥0.7, it is marked as an "industrial candidate";

[0099] b) Time-domain discrimination

[0100] For time-domain signal segments labeled "Metro Candidate" or "Industrial Candidate", calculate their short-time average amplitude M and short-time variance σ²:

[0101] If M / σ²≥3 and the pulse width τ satisfies 20ms≤τ≤200ms, then it is finally determined to be a subset of subway interference;

[0102] If M / σ²≤1.5 and the waveform has a continuous sinusoidal envelope, then it is finally determined to be a subset of industrial interference;

[0103] c) Joint discrimination

[0104] When the same signal segment satisfies both conditions a) and b), it is assigned to the corresponding subset; if neither condition is met, it is marked as another interference source and removed.

[0105] Step 304: Based on the subway interference subset and the industrial interference subset, the support vector machine algorithm is used to classify the interference intensity and duration to obtain the interference classification result;

[0106] After determining the interference feature dataset, spectral analysis is performed to divide it into subway interference and industrial interference subsets. Original data records are obtained from these subsets to determine the initial interference dataset. Based on this initial dataset, feature extraction methods are used to separate the feature values ​​of interference intensity and duration, resulting in a characteristic dataset. A support vector machine (SVM) algorithm is then used to train a classification model for this characteristic dataset. This model is used to classify the subway and industrial interference subsets based on interference intensity and duration. The trained classification model is then used to classify the characteristic dataset, determining the distribution of interference intensity and duration categories to obtain preliminary classification results. If the preliminary classification results show an imbalance in category distribution, the data subsets are resampled to obtain adjusted classification results. Based on the adjusted classification results and interference source information, the final interference classification mapping is determined. This final mapping generates a correlation between interference sources, interference intensity, and duration, assessing the completeness of the classification effect.

[0107] Specifically, in classifying the interference subsets from the subway and the industry, the original interference data was first cleaned and its features extracted through data preprocessing. The subway interference subset contained 1000 data points, each recording the interference intensity (in decibels, ranging from 30.5 to 80.5) and duration (in seconds, ranging from 10 to 300). The industrial interference subset contained 800 data points, with intensity ranging from 40.5 to 90.5 and duration ranging from 15 to 400. Standardization was used to normalize the data to a range of 0 to 1, ensuring that features of different dimensions had a consistent impact on the model. Next, a support vector machine model was constructed using a radial basis function (RBF) kernel. The penalty parameter C was set to 1.0, and the kernel function parameter gamma was set to 0.1. The model performance was evaluated using five-fold cross-validation, yielding an average accuracy of 85.3%, indicating that the model has a strong ability to distinguish between the two types of interference data. Then, the normalized data was input into the model for training. The training and test sets were divided in an 8:2 ratio, with 1440 data points in the training set and 360 data points in the test set. The model output classification results, showing that 87.2% of the subway interference data and 83.5% of the industrial interference data were correctly classified. Finally, analysis of the classification results revealed that samples with an intensity higher than 70.5 dB and an impact time exceeding 200 seconds were more likely to be classified as industrial interference, while samples with an intensity lower than 50.5 dB and an impact time less than 100 seconds were mostly classified as subway interference. Through feature importance analysis, the influence weight of intensity on the classification results was 0.65, and the impact time was 0.35. When further optimizing the model, the threshold setting of the intensity feature can be adjusted first.

[0108] To establish business connections, the classification results can be combined with the interference source location system. Assuming the location system adjusts the detection frequency based on the classification results, industrial interference has a higher priority and the detection frequency is increased to 5 times per minute, while subway interference is detected 3 times per minute, thereby optimizing resource allocation. The entire process is completed through automated algorithms and system interfaces, ensuring rigorous logic and data-driven operation.

[0109] Step 305: Based on the interference classification results, the data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate to obtain a corrosion risk association model.

[0110] The interference classification results include the correlation between the interference source, interference intensity, and duration.

[0111] By acquiring interference classification results and pipeline corrosion rate data, a linear regression algorithm is used for data association modeling to train a corrosion risk association model. The correlation coefficient between interference intensity and pipeline corrosion rate is calculated to obtain a trained corrosion risk model. This model can then be used to predict pipeline corrosion rate by inputting real-time interference characteristics, yielding a predicted corrosion rate value. Based on the predicted corrosion rate value and pre-defined risk assessment rules, the corrosion risk level of the pipeline is determined, resulting in a corrosion risk assessment result.

[0112] This implementation collects electrical interference signals from the environment surrounding buried pipelines using multi-channel sensors, extracts interference features using a fast Fourier transform algorithm, determines the source of interference by matching preset frequency bands, classifies the interference using a support vector machine algorithm, and then establishes a corrosion risk correlation model. This allows the corrosion rate of pipelines to be predicted by inputting real-time interference features into the corrosion risk model, thereby improving the accuracy of pipeline corrosion risk prediction and achieving precise assessment of pipeline corrosion risk.

[0113] In one embodiment, please refer to Figure 3 Based on the interference classification results, a data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate, resulting in a corrosion risk association model, including:

[0114] Step 3051: Based on the interference classification results and pipeline corrosion rate data, extract the feature values ​​of interference intensity and corrosion rate from the pre-established database to obtain the initial dataset;

[0115] Step 3052: Use principal component analysis to extract features from the initial dataset, reduce the data dimensionality, and obtain a dimensionality-reduced feature set;

[0116] Step 3053: Using the Pearson correlation coefficient method, calculate the correlation coefficient between the interference intensity in the dimensionality-reduced feature set and the pipeline corrosion rate to obtain the correlation coefficient matrix;

[0117] Step 3054: Based on the correlation coefficient matrix and the interference classification result, the feature pairs of interference intensity and corrosion rate in the dimensionality reduction feature set are marked as highly correlated feature pairs to obtain a highly correlated feature set.

[0118] Step 3055: Based on the highly correlated feature set, a linear regression algorithm is used to train a preset training model to obtain the corrosion risk association model.

[0119] The process begins by acquiring interference classification results and pipeline corrosion rate data. Feature values ​​of interference intensity and corrosion rate are extracted from a pre-established database to obtain an initial dataset. Principal component analysis (PCA) is then used to extract features from the initial dataset, reducing its dimensionality to obtain a dimensionality-reduced feature set. The Pearson correlation coefficient method is then used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate within the dimensionality-reduced feature set, resulting in a correlation coefficient matrix. If the absolute value of a correlation coefficient in the matrix exceeds a preset threshold, this feature pair is marked as a highly correlated feature pair. The feature data corresponding to these highly correlated feature pairs in the dimensionality-reduced feature set are then used as the highly correlated feature set. Based on this highly correlated feature set, a pre-defined linear regression model is trained using a linear regression algorithm to obtain a trained corrosion risk association model. Finally, real-time interference intensity data is input into the corrosion risk association model to predict the pipeline corrosion rate, yielding the predicted corrosion rate value. The advantage of using a linear regression model for training is that it can train a linear model to determine which features of electrical interference sources have the greatest impact on pipeline corrosion. For example, by inputting different frequency band values, temperature values, or humidity values ​​into the trained linear regression model, i.e., inputting different interference features, the linear regression model outputs the pipeline corrosion rate, which is used to predict the pipeline corrosion rate under different interference conditions, thereby achieving the prediction of which interference source has the greatest impact on pipeline corrosion risk.

[0120] Specifically, based on the interference classification results, a data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate, resulting in the following implementation method for the corrosion risk association model:

[0121] First, historical data on the pipeline system was collected, including interference intensity (e.g., electromagnetic interference intensity, in μT) and corrosion rate (in mm / year). The dataset is assumed to contain 1000 records, with interference intensity ranging from 0.5 to 50 μT and corrosion rate ranging from 0.01 to 2.5 mm / year. A Support Vector Machine (SVM) algorithm was used to classify the interference intensity into three levels: low (0.5-10 μT), medium (10-30 μT), and high (30-50 μT). The SVM used a radial basis function (RBF) kernel with parameters C=1.0 and γ=0.1, achieving a classification accuracy of 85%. Next, the Pearson correlation coefficient was used to calculate the relationship between interference intensity and corrosion rate. The formula for calculating the correlation coefficient is as follows:

[0122]

[0123] Where r is the correlation coefficient. For interference intensity, For corrosion rate, and These are the mean values. The calculation results show that the correlation coefficient r between interference intensity and corrosion rate is 0.78, indicating a strong positive correlation. Subsequently, a multiple linear regression model was constructed, where the model formula is:

[0124]

[0125] in, For corrosion rate, For interference intensity, The hardness of the pipe material (assuming a hardness range of 100-300 HB), , and These are the regression coefficients. The least squares method is used for fitting to obtain... =0.02, =0.045, =-0.001, the explained coefficient of the linear regression model is R²=0.82, indicating that the model has strong explanatory power. Finally, based on the regression model, the formula for the corrosion risk index is:

[0126]

[0127] in, For corrosion rate, The interference intensity is represented by a risk index ranging from 0 to 1. When R > 0.7, it is marked as high risk. Assuming the interference intensity of a pipeline is 40 μT and the hardness is 200 HB, substituting these values ​​into the model formula yields y = 0.02 + 0.045 × 40 - 0.001 × 200 = 1.62 mm / year, R = 0.6 × 1.62 + 0.4 × 40 / 50 = 1.292. After normalization, R = 0.81, thus classifying it as high risk.

[0128] In one embodiment, please refer to Figure 4 The step of determining the subway interference subset and the industrial interference subset based on the interference feature dataset includes:

[0129] Step 3031: If the frequency characteristics in the interference characteristic dataset match the preset subway operating frequency band, then the source of interference is determined to be subway operation through time domain analysis, and a subway interference subset is obtained.

[0130] Signal data is obtained from an interference characteristic dataset, and frequency features are extracted using frequency domain analysis to determine if they match the preset operating frequency band range, thus obtaining preliminary matching results. If the preliminary matching results show that the frequency features match the operating frequency band, time domain analysis is performed on the extracted frequency features to determine the temporal variation pattern of the signal. Based on the temporal variation pattern obtained from the time domain analysis, classification is performed to determine whether the interference is caused by subway operation, resulting in a classified interference subset. For the classified interference subset, its corresponding time-frequency feature distribution is obtained, and a support vector machine algorithm is used to perform pattern recognition on the feature distribution to determine the feature pattern of the interference subset. If the feature pattern matches the preset subway operation interference pattern, key time period features are further extracted from the interference subset through comparative analysis, resulting in a refined interference feature set. Based on the refined interference feature set and preset standards, signal screening is performed to determine whether there are abnormal interference signals, thus obtaining the subway interference subset.

[0131] Specifically, firstly, frequency features were extracted from the interference characteristic dataset through frequency domain analysis. The dataset was assumed to contain electromagnetic signals lasting one hour, with a sampling rate of 1000Hz and a total of 3,600,000 sampling points. The Fast Fourier Transform (FFT) algorithm was used to convert the time-domain signal to the frequency domain, obtaining a spectrum with a resolution of 0.001Hz. Feature peaks in the frequency range of 10Hz to 100Hz were extracted. Significant peaks were found at 15Hz, 30Hz, and 60Hz, which highly matched the preset subway operating frequency band (10Hz to 80Hz, with a main frequency of 30Hz). The matching degree was calculated using the Pearson correlation coefficient, yielding a coefficient of 0.92, higher than the threshold of 0.85, indicating that the frequency characteristics conformed to subway operating characteristics. Next, time-domain analysis was performed. For the subset matching in the frequency domain, the original signal for the corresponding time period was extracted, assuming the signal amplitude fluctuated between 0.1V and 0.5V. The autocorrelation function was used to analyze the signal periodicity. The calculation results showed a period of approximately 33.3ms, consistent with the 30Hz main frequency, further confirming that the interference source was subway operation. To generate a subset of subway interference, a bandpass filter (10Hz to 80Hz) is used to filter out irrelevant frequencies, retaining the matched signal. The filtered signal is then segmented into interference segments using threshold detection (amplitude > 0.2V), assuming 100 interference subsets are obtained, each approximately 0.5s in duration. The final subsets are stored as a matrix of timestamp and amplitude pairs for subsequent interference suppression processing. This process, through a logical progression from the frequency domain to the time domain, ensures accurate identification of the interference source and that the generated subsets meet operational requirements.

[0132] Step 3032: If the frequency characteristics in the interference characteristic dataset match the preset industrial equipment frequency band, then the source of interference is determined to be industrial equipment through waveform analysis, and an industrial interference subset is obtained.

[0133] Frequency features are obtained from the interference characteristic dataset through signal processing, and spectral features are calculated using the Fast Fourier Transform (FFT) algorithm to obtain a frequency feature set. If the feature values ​​in the frequency feature set match the preset industrial equipment frequency band range, the degree of matching is determined through spectral analysis to obtain a matching feature subset. Waveform analysis is used to perform joint time-domain and frequency-domain processing on the matching feature subset, and interference waveform features are extracted through Short-Time Fourier Transform (SFT) to obtain an interference waveform set. Similarity calculation is performed between the interference waveform set and the preset industrial equipment waveform template, and the interference source is determined using the Dynamic Time Warping (RTD) algorithm to obtain a preliminary interference source set. Interference features related to industrial equipment are extracted from the preliminary interference source set, and interference subsets are divided through cluster analysis to obtain the industrial interference subset. Furthermore, spectral denoising processing can be performed on the industrial interference subset, and adaptive filtering technology can be used to separate the interference signals to obtain a clean interference signal set. Pattern matching is performed between the clean interference signal set and the preset industrial equipment signal library to determine the equipment type of the interference source, resulting in the final industrial interference subset.

[0134] Specifically, when processing interference characteristic datasets, frequency features are first extracted through spectrum analysis. Assuming the dataset contains multiple signal samples, each with a sampling rate of 10 kHz and a duration of 1 second, the time-domain signal is converted to a frequency-domain signal using a Fast Fourier Transform (FFT) algorithm to obtain the frequency components of each sample. The main frequency peaks are then calculated; for example, if a sample's main frequency is found to be 50 Hz and 100 Hz, these frequency features are then matched against preset industrial equipment frequency bands. Assuming the typical frequency band for industrial equipment is 40 Hz to 120 Hz, a matching algorithm is written, setting a frequency error tolerance of ±5 Hz. If a sample's main frequency falls within this range, it is marked as a potential industrial interference signal. For example, if both 50 Hz and 100 Hz meet the criteria, the sample is marked as suspected industrial interference. Next, waveform analysis was used to further identify the source of interference. Wavelet transform was employed to perform multi-resolution decomposition of the signal, extracting its time-frequency features. Assuming a five-level decomposition, the energy proportion of low-frequency components (e.g., 20 Hz to 60 Hz) relevant to industrial equipment operation was analyzed in each level. If the proportion exceeded 60%, such as a sample with a low-frequency energy proportion of 75%, the interference source was confirmed to be industrial equipment. Finally, all samples meeting the criteria were categorized into an industrial interference subset. Assuming a total of 1000 samples, 200 samples were ultimately selected as the industrial interference subset with a main frequency between 40 Hz and 120 Hz and a low-frequency energy proportion exceeding 60%. This subset was saved to a database using a data storage module, generating a spectral characteristic report containing the main frequency value and energy proportion data for each sample, forming a complete data processing chain to ensure traceability for subsequent analyses.

[0135] In this embodiment, the Fast Fourier Transform algorithm is used to extract interference features, and the interference source is determined by matching the preset frequency band. The Support Vector Machine algorithm is used to classify the interference, so as to accurately classify the feature data in the interference feature dataset into a subway interference subset and an industrial interference subset, thereby improving the prediction effect of corrosion risk of buried pipelines from different interference sources.

[0136] In one embodiment, please refer to Figure 5 The step of determining the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system includes:

[0137] Step 501: Based on the target protection parameters, adjust the current output of the pipeline cathodic protection system using a digital signal processor to obtain protection optimization parameters;

[0138] Step 502: Update the operating status of the pipeline cathodic protection system in real time through the protection optimization parameters, and determine the real-time operating status fed back by the pipeline cathodic protection system when the target protection parameters are applied.

[0139] Step 503: Determine the dynamic protection configuration based on the real-time operating status.

[0140] When the deviation between the real-time operating status and the target protection parameters is detected to exceed a preset threshold, the closed-loop adjustment of the current output is completed within 30 seconds, and the adjustment result is written into the real-time control register of the cathodic protection system.

[0141] Based on the target protection parameters and the real-time operating status of the pipeline cathodic protection system, the target protection parameters are adjusted to obtain a new current output scheme and determine the dynamic protection configuration. Specifically, the target protection parameters are input into the pipeline cathodic protection system to obtain the current output value. The current output value is dynamically corrected to determine whether it meets the preset threshold range. If it exceeds the threshold range, the current output value is recalculated until the current output value meets the condition. Based on the corrected current output value, the protection optimization parameters are obtained. The protection optimization parameters are input into the pipeline cathodic protection system to obtain the real-time operating status feedback from the pipeline cathodic protection system. Based on the protection optimization parameters and the real-time operating status feedback from the pipeline protection system, the changing trend of the protection effect is analyzed to obtain the effect evaluation result. If the effect evaluation result shows that the protection effect does not meet expectations, the generation logic of the target protection parameters is adjusted to determine the final protection parameter configuration and obtain the dynamic protection configuration. Through the system adjustment function, the final protection parameter configuration, i.e., the dynamic protection configuration, is applied to the pipeline cathodic protection system to obtain continuous protection effect monitoring data, completing the closed-loop processing of parameter optimization.

[0142] In another embodiment, protection optimization parameters are determined based on the target protection parameters. Then, the protection optimization parameters are run, and the current operating status is analyzed by collecting real-time data from the pipeline cathodic protection system to obtain preliminary status feedback information. This preliminary status feedback information is compared with a preset threshold to determine if there is any deviation, and deviation analysis results are obtained. If the deviation analysis results exceed a preset range, the protection optimization parameters are optimized and adjusted to determine the adjusted parameter values. The operating status of the cathodic protection system is updated using the protection optimization parameters, and updated system operating data is obtained. Based on the updated system operating data, the protection effect is analyzed using a real-time monitoring mechanism to obtain effect evaluation information. If the protection effect does not meet expectations based on the effect evaluation information, a dynamic configuration mechanism is triggered to adjust the system update strategy and determine the final protection configuration scheme, resulting in a dynamic protection configuration. Using the final protection configuration scheme, the operating status of the pipeline system is continuously monitored to obtain long-term status feedback data.

[0143] In this embodiment, by combining the target protection parameters with the real-time operating status fed back by the pipeline cathodic protection system, the protection optimization parameters are continuously optimized. Through a dynamic configuration mechanism, the system update strategy is adjusted to obtain the final protection configuration scheme, namely dynamic protection configuration. This realizes the dynamic protection configuration of the pipeline cathodic protection system and achieves precise protection for buried pipelines, thereby improving the protection effect of the pipeline.

[0144] In one embodiment, please refer to Figure 6 The corrosion risk correlation model includes a corrosion rate prediction model, a corrosion risk level prediction model, and a corrosion risk outcome prediction model. The interference features are input into the corrosion risk correlation model to determine the impact of different interference sources on pipeline corrosion risk, and the risk assessment results output by the corrosion risk correlation model are obtained, including:

[0145] Step 311: Input the interference features into the corrosion rate prediction model to obtain the corrosion rate result output by the corrosion rate prediction model, wherein the corrosion rate result reflects the influence of different interference sources on pipeline corrosion.

[0146] Step 312: Input the corrosion rate result into the corrosion risk level prediction model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtain the risk level list output by the corrosion risk level prediction model. The risk level list is determined by marking interference features whose influence weight exceeds a preset threshold as high-risk factors.

[0147] Step 313: Input the risk level list into the corrosion risk result prediction model to obtain the risk assessment result output by the corrosion risk result prediction model;

[0148] The risk assessment results include the pipeline's corrosion risk level, the time required for pipeline perforation, the pipeline risk trend, and the pipeline protection strategy.

[0149] The corrosion risk correlation model includes a corrosion rate prediction model, a corrosion risk level prediction model, and a corrosion risk outcome prediction model. The output of the corrosion rate prediction model is connected to the input of the corrosion risk level prediction model, and vice versa. Specifically, interference features are input into the corrosion rate prediction model to obtain the corrosion rate result output by the model. For example, it can predict the pipeline corrosion rate under subway interference signals or industrial interference signals, and also predict the pipeline corrosion rate under different temperature and humidity conditions. Then, the corrosion rate results corresponding to different interference sources are input into the corrosion risk level prediction model, which outputs the risk level results corresponding to different interference sources. If the influence weight of a certain interference source exceeds a preset threshold, it is marked as a high-risk factor, resulting in a classified risk level list. The risk level list is input into the corrosion risk outcome prediction model. Based on the classified risk level list, in-depth data mining is performed on the high-risk factors to obtain their specific action paths on pipeline corrosion, determine the potential corrosion risk change trend, predict the time required for pipeline perforation, and obtain the risk assessment results for different interference sources. In one embodiment, the corrosion rate prediction model can be a linear regression model, the corrosion risk level prediction model can be a decision tree model, and the corrosion risk outcome prediction model can be a random forest model.

[0150] In this embodiment, a corrosion rate prediction model is used to predict the corrosion rate of the pipeline from different sources of interference. A corrosion risk level prediction model is used to predict the level of corrosion risk of different sources of interference under their corresponding corrosion rate conditions. A corrosion risk result prediction model is used to predict the risk assessment results of pipeline corrosion under the corresponding corrosion rate and corrosion risk level conditions for different sources of interference. These results include the pipeline corrosion risk trend, the time required for pipeline perforation, and pipeline protection strategies. This achieves accurate assessment of pipeline corrosion risk, greatly reduces the risk of pipeline corrosion, and improves the protection level of buried pipelines.

[0151] In one embodiment, please refer to Figure 7 The corrosion risk prediction model is used for:

[0152] Step 3131: Based on the risk level list, conduct in-depth data mining to determine the action path of high-risk factors on pipeline corrosion and the corrosion risk change trend corresponding to the action path.

[0153] Step 3132: Based on the corrosion risk change trend and historical data of pipeline status, use conditional judgment logic to determine the key monitoring objects in the risk level list;

[0154] Step 3133: Based on the key monitoring objects, conduct in-depth data mining to determine the target action path of the key monitoring objects and the target corrosion risk change trend corresponding to the target action path;

[0155] Step 3134: Based on the target action path and the target corrosion risk change trend, determine the risk assessment result;

[0156] If the trend of corrosion risk changes matches the historical high-risk pattern in the historical data, then key monitoring targets are identified.

[0157] Specifically, the corrosion risk prediction model performs in-depth data mining on high-risk factors in the risk level list to predict their specific action paths on pipeline corrosion. It also determines the potential corrosion risk trends under these action paths. For example, under a high-risk level, if the predicted action path is high humidity and coating aging, the corrosion risk trend of the corresponding pipeline is predicted to become increasingly severe. Therefore, under high-risk conditions with high humidity and coating aging as the action path, the corresponding interference source and pipeline are identified as key monitoring targets. Conversely, under a low-risk level, if the predicted action path is low humidity, the corrosion of the corresponding pipeline is predicted to be relatively minor, and the risk trend is expected to decrease. Therefore, under low-risk conditions with low humidity as the action path, the corresponding interference source and pipeline are identified as ordinary monitoring targets. By analyzing the corrosion risk trends and combining them with historical pipeline safety data, a conditional judgment logic is used. If a trend matches a historical high-risk pattern, it is prioritized to determine key monitoring targets. Subsequently, based on the key monitoring targets, in-depth data mining was conducted on the high-risk factors exhibited by these targets to predict the target action path of the key monitoring targets on pipeline corrosion and the trend of target corrosion risk changes under this target action path. This yielded the target action path and target corrosion risk change trend corresponding to the key monitoring targets. Finally, based on the target action path and target corrosion risk change trend corresponding to the key monitoring targets, a targeted risk assessment strategy was generated. A comparative analysis of real-time monitoring data and model prediction results was performed to determine if there were any discrepancies and to adjust the model parameters. Using the adjusted model parameters, the decision tree algorithm was re-run to continuously analyze the dynamic changes in pipeline corrosion risk, resulting in updated assessment results.

[0158] In this embodiment, a corrosion risk prediction model is used to predict the sources of interference and the objects in the pipeline that need to be controlled, as well as the corrosion risk trend of the key monitoring objects, the time required for pipeline perforation, and pipeline protection strategies under the corresponding corrosion rate and corrosion risk level. This results in a risk assessment of pipeline corrosion, thereby achieving an accurate assessment of pipeline corrosion risk, greatly reducing the risk of pipeline corrosion, and improving the protection level of buried pipelines.

[0159] The following describes the real-time dynamic monitoring system for stray current in buried pipelines provided by the present invention. The real-time dynamic monitoring system for stray current in buried pipelines described below can be referred to in correspondence with the real-time dynamic monitoring method for stray current in buried pipelines described above.

[0160] Please refer to Figure 8 This invention provides a real-time dynamic monitoring system for stray current in buried pipelines, comprising:

[0161] The signal acquisition module 810 is used to acquire electrical interference signals from the environment surrounding the buried pipeline.

[0162] The interference feature extraction module 820 is used to determine the interference features of the electrical interference signal based on the electrical interference signal, wherein the interference features include the interference source, interference intensity and duration corresponding to the electrical interference signal;

[0163] The result prediction module 830 inputs the interference features into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk and obtains the risk assessment result output by the corrosion risk association model.

[0164] The parameter generation module 840 is used to generate target protection parameters corresponding to the electrical interference signal based on the risk assessment results.

[0165] The dynamic protection configuration module 850 is used to determine the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system.

[0166] The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines, and the corrosion risk association model is trained using interference feature datasets corresponding to different interference sources.

[0167] Figure 9 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 9As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930, and a communication bus 940, wherein the processor 910, the communication interface 920, and the memory 930 communicate with each other through the communication bus 940. The processor 910 can call logic instructions in the memory 930 to execute a method for real-time dynamic monitoring of stray current in buried pipelines. This method includes: acquiring electrical interference signals from the environment surrounding the buried pipeline; determining interference characteristics of the electrical interference signals based on the signals, wherein the interference characteristics include the interference source, interference intensity, and duration of the interference; inputting the interference characteristics into a corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtaining a risk assessment result output by the corrosion risk association model; generating target protection parameters corresponding to the electrical interference signals based on the risk assessment result; and determining a dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the pipeline cathodic protection system. The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on the buried pipeline, and the corrosion risk association model is trained using a dataset of interference characteristics corresponding to different interference sources.

[0168] Furthermore, the logical instructions in the aforementioned memory 930 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0169] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for real-time dynamic monitoring of stray current in buried pipelines provided by the methods described above. This method includes: acquiring electrical interference signals from the environment surrounding the buried pipeline; determining interference characteristics of the electrical interference signals based on the electrical interference signals, wherein the interference characteristics include the interference source, interference intensity, and duration corresponding to the electrical interference signals; inputting the interference characteristics into a corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtaining a risk assessment result output by the corrosion risk association model; generating target protection parameters corresponding to the electrical interference signals based on the risk assessment result; and determining a dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the pipeline cathodic protection system. The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on the buried pipeline, and the corrosion risk association model is trained using a dataset of interference characteristics corresponding to different interference sources.

[0170] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units 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. Those skilled in the art can understand and implement this without any creative effort.

[0171] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0173] Terminology Definition

[0174] To enable those skilled in the art to accurately understand and implement this invention, the key terms used in this specification are defined as follows. Unless otherwise stated, the following terms have the following meanings:

[0175] 1. Buried Pipeline

[0176] It refers to metal or non-metal pipelines and their ancillary facilities laid below the ground surface for transporting gas, oil, water or other media, and usually includes main pipelines, valves, fittings, insulating joints and cathodic protection systems.

[0177] 2. Stray Current

[0178] Also known as stray current, it refers to the unexpected direct current or alternating current flowing through buried metal structures (such as pipelines), originating from urban rail transit, DC power transmission systems, industrial electrolysis equipment, high-voltage AC line induction, lightning or grounding faults, etc.

[0179] 3. Electrical Interference Signal

[0180] In the context of this invention, it specifically refers to the instantaneous values ​​and spectral components of voltage, current, or magnetic field acquired by sensors that reflect stray currents and their changes. This signal serves as the raw data for subsequent interference feature extraction.

[0181] 4. Interference Feature

[0182] Key quantitative indicators extracted from electrical interference signals to characterize the source of interference and corrosion risk, including but not limited to:

[0183] • Sources of interference (subway, industrial rectifier equipment, high-voltage AC lines, lightning, etc.);

[0184] • Interference intensity (current density mA / m², magnetic induction intensity μT, potential deviation mV);

[0185] • Duration of action (interference duration, periodicity, pulse width);

[0186] • Frequency domain characteristics (dominant frequency, harmonic distribution, power spectral density).

[0187] 5. Corrosion Risk Association Model

[0188] This refers to a mathematical model built using machine learning or statistical methods (such as support vector machines, random forests, linear regression, neural networks, or combinations thereof). The model takes disturbance features as input and outputs corrosion rate or corrosion risk level to quantify the impact of different disturbance sources on the corrosion of buried pipelines. The model is trained using historical disturbance feature datasets and corresponding corrosion rate datasets.

[0189] 6. Corrosion Rate

[0190] In this invention, the term refers to the amount of pipe metal thickness loss caused by stray current per unit time, commonly expressed in millimeters per year (mm / a), which can be measured in real time or periodically using a hanging test, electrochemical probe, or ultrasonic thickness gauge.

[0191] 7. Risk Assessment Result

[0192] The comprehensive assessment information output by the corrosion risk correlation model should include at least:

[0193] • Current corrosion risk level (high, medium, low);

[0194] • Predicting perforation time (Time-to-Perforation);

[0195] • Corrosion risk trends (increasing, stable, decreasing);

[0196] • Recommended protection strategies (such as current increment, coating repair, drain grounding, polarity drain switching).

[0197] 8. Target Protection Parameter

[0198] Based on the risk assessment results, control commands or setpoints for real-time adjustment of the cathodic protection system are generated by algorithms, such as output current value (A), protection potential setpoint (V vs. Cu / CuSO4), duty cycle (%), or pulse frequency (Hz).

[0199] 9. Dynamic Protection Configuration

[0200] This refers to a closed-loop control strategy that is calculated and issued in real time by a digital signal processor (DSP) or microcontroller (MCU) based on the target protection parameters and real-time feedback from the cathodic protection system (actual output current, pipeline potential, system power consumption, etc.) to ensure that the cathodic protection system maintains the best protection level in stray current changing scenarios.

[0201] 10. Multi-Channel Sensor

[0202] A sensor array with ≥2 simultaneous acquisition capabilities, deployed along buried pipelines, may include any one or a combination of Hall current sensors, fluxgate sensors, reference electrodes, resistance probes, or electromagnetic induction coils, used to acquire electrical interference signals.

[0203] 11. Real-Time Operating Status

[0204] The current operating points of the cathodic protection system include, but are not limited to:

[0205] • Actual output current / voltage;

[0206] • The instantaneous potential of the pipe relative to the reference electrode;

[0207] • System temperature, power consumption, and fault codes;

[0208] • Deviation from the target protection parameters.

[0209] 12. Non-transitory computer-readable storage medium

[0210] This refers to a medium that can tangibly store computer program instructions and is not transient, such as flash memory, solid-state drive, read-only memory (ROM), random access memory (RAM), magnetic disk, optical disk, USB flash drive, cloud storage server, etc., used to implement the software deployment and distribution of the method described in this invention.

[0211] Unless otherwise specified, the terms used above apply to the entire contents of this specification, claims, and drawings.

Claims

1. A method for real-time dynamic monitoring of stray current in buried pipelines, characterized in that, Executed by a computer, including: Obtain electrical interference signals from the environment surrounding buried pipelines; Based on the electrical interference signal, the interference characteristics of the electrical interference signal are determined, wherein the interference characteristics include the interference source, interference intensity, and duration of the electrical interference signal. The interference features are input into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and the risk assessment results output by the corrosion risk association model are obtained. Based on the risk assessment results, target protection parameters corresponding to the electrical interference signal are generated; Based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system, the dynamic protection configuration is determined. The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines. The corrosion risk association model is trained using interference feature datasets corresponding to different interference sources, including: Electrical interference signals are obtained from the environment surrounding the buried pipeline, and voltage and current data are collected using multi-channel sensors to obtain the raw interference dataset. The signal spectrum is decomposed using the Fast Fourier Transform algorithm, and the frequency and amplitude features of the original interference dataset are extracted to obtain the interference characteristic dataset. Based on the aforementioned interference feature dataset, a subway interference subset and an industrial interference subset are determined; Based on the subway interference subset and the industrial interference subset, the support vector machine algorithm is used to classify the interference intensity and duration to obtain the interference classification results. The interference classification results include the correlation between the interference source and the interference intensity and duration. Based on the interference classification results, the data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate, and a corrosion risk association model is obtained. The step of determining the subway interference subset and the industrial interference subset based on the interference feature dataset includes: If the frequency characteristics in the interference characteristic dataset match the preset subway operating frequency band, then the source of interference is determined to be subway operation through time domain analysis, and a subway interference subset is obtained. If the frequency characteristics in the interference characteristic dataset match the preset industrial equipment frequency band, then the source of interference is determined to be industrial equipment through waveform analysis, and an industrial interference subset is obtained. The step of determining the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system includes: Based on the target protection parameters, a digital signal processor is used to adjust the current output of the pipeline cathodic protection system to obtain protection optimization parameters. By optimizing the protection parameters, the operating status of the pipeline cathodic protection system is updated in real time, and the real-time operating status fed back by the pipeline cathodic protection system in response to the target protection parameters is determined. Based on the real-time operating status, determine the dynamic protection configuration.

2. The method for real-time dynamic monitoring of stray current in buried pipelines according to claim 1, characterized in that, Based on the interference classification results, a data association modeling method is used to calculate the correlation coefficient between interference intensity and pipeline corrosion rate, resulting in a corrosion risk association model, including: Based on the interference classification results and pipeline corrosion rate data, feature values ​​of interference intensity and corrosion rate are extracted from a pre-established database to obtain an initial dataset. Principal component analysis is used to extract features from the initial dataset, reducing the data dimensionality and obtaining a dimensionality-reduced feature set. The Pearson correlation coefficient method was used to calculate the correlation coefficient between the interference intensity in the dimensionality-reduced feature set and the pipeline corrosion rate, and the correlation coefficient matrix was obtained. Based on the correlation coefficient matrix and the interference classification results, the feature pairs of interference intensity and corrosion rate in the dimensionality reduction feature set are marked as highly correlated feature pairs, thus obtaining a highly correlated feature set; Based on the highly correlated feature set, a linear regression algorithm is used to train a preset training model to obtain the corrosion risk correlation model.

3. The method for real-time dynamic monitoring of stray current in buried pipelines according to claim 1, characterized in that, The corrosion risk correlation model includes a corrosion rate prediction model, a corrosion risk level prediction model, and a corrosion risk outcome prediction model. The interference features are input into the corrosion risk correlation model to determine the degree of influence of different interference sources on pipeline corrosion risk, obtaining the risk assessment results output by the corrosion risk correlation model, including: The interference features are input into the corrosion rate prediction model to obtain the corrosion rate result output by the corrosion rate prediction model, wherein the corrosion rate result reflects the influence of different interference sources on pipeline corrosion. The corrosion rate results are input into the corrosion risk level prediction model to determine the degree of influence of different interference sources on pipeline corrosion risk, and the risk level list output by the corrosion risk level prediction model is obtained. The risk level list is determined by marking interference features whose influence weights exceed a preset threshold as high-risk factors. Input the risk level list into the corrosion risk result prediction model to obtain the risk assessment result output by the corrosion risk result prediction model; The risk assessment results include the pipeline's corrosion risk level, the time required for pipeline perforation, the pipeline risk trend, and the pipeline protection strategy.

4. The method for real-time dynamic monitoring of stray current in buried pipelines according to claim 3, characterized in that, The corrosion risk prediction model is used for: Based on the risk level list, in-depth data mining is conducted to determine the action path of high-risk factors on pipeline corrosion and the corrosion risk change trend corresponding to the action path. Based on the corrosion risk change trend and historical data of pipeline status, conditional judgment logic is used to determine the key monitoring objects in the risk level list. Based on the key monitoring objects, in-depth data mining is conducted to determine the target action path of the key monitoring objects and the target corrosion risk change trend corresponding to the target action path; The risk assessment result is determined based on the target action path and the target corrosion risk change trend; If the trend of corrosion risk changes matches the historical high-risk pattern in the historical data, then key monitoring targets are identified.

5. A real-time dynamic monitoring system for stray current in buried pipelines, used to implement the real-time dynamic monitoring method for stray current in buried pipelines as described in any one of claims 1 to 4, characterized in that, include: The signal acquisition module is used to acquire electrical interference signals from the environment surrounding the buried pipeline; An interference feature extraction module is used to determine the interference features of the electrical interference signal based on the electrical interference signal, wherein the interference features include the interference source, interference intensity, and duration corresponding to the electrical interference signal; The result prediction module inputs the interference features into the corrosion risk association model to determine the degree of influence of different interference sources on pipeline corrosion risk, and obtains the risk assessment result output by the corrosion risk association model. The parameter generation module is used to generate target protection parameters corresponding to the electrical interference signal based on the risk assessment results. The dynamic protection configuration module is used to determine the dynamic protection configuration based on the target protection parameters and the real-time operating status fed back by the target protection parameters acting on the pipeline cathodic protection system. The dynamic protection configuration is used by the pipeline cathodic protection system to perform electrochemical dynamic protection operations on buried pipelines, and the corrosion risk association model is trained using interference feature datasets corresponding to different interference sources.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the real-time dynamic monitoring method for stray current in buried pipelines as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the real-time dynamic monitoring method for stray current in buried pipelines as described in any one of claims 1 to 4.

Images