Oil and gas pipeline intelligent security and protection integrated platform and multi-system linkage method

Abstract

The invention discloses an oil and gas pipeline intelligent security integrated platform and a multi-system linkage method, particularly relates to the technical field of pipeline safety monitoring, and is used for solving the problems of vibration characteristic drift and misjudgment caused by long-term stress accumulation of a pipeline in an existing security system. According to the method, a self-adaptive vibration feature library is constructed by dynamically sensing a pipeline stress state and a microcrack propagation signal, and the influence of stress interference on monitoring data is eliminated in combination with a time domain alignment and propagation path correction technology; based on real-time fusion of multi-node vibration data and stress distribution data, an energy judgment threshold weight distribution proportion is dynamically adjusted, a threat level is evaluated through an AI classification model in combination with stress distribution characteristics, and a collaborative response instruction matched with the state of a pipeline stress concentration area is generated. The environmental adaptability and the emergency response reliability of the security and protection system are improved, and dynamic collaborative optimization of safety monitoring and multi-system linkage in the whole life cycle of the pipeline is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline safety monitoring. More specifically, the present invention relates to an intelligent security integration platform for oil and gas pipelines and a multi-system linkage method. Background Art

[0002] In the field of intelligent security for oil and gas pipelines, the existing technology generally adopts a multi-system linkage mechanism, integrating subsystems such as vibration monitoring, video surveillance, and perimeter protection, and realizing comprehensive early warning and disposal of abnormal events through data fusion; relying on technologies such as fiber optic vibration sensing and infrared thermal imaging to collect pipeline status information in real time, and combining artificial intelligence algorithms to perform pattern recognition on physical quantities such as vibration spectra and image features to distinguish normal environmental interference from human sabotage behavior.

[0003] In the existing technology, the logic of multi-system linkage depends on a preset vibration feature library and threshold judgment rules. However, the stress history caused by the long-term service of oil and gas pipelines will implicitly change their dynamic characteristics (such as natural frequency shift, vibration wave propagation path distortion, etc.), making the same external excitation (such as mechanical excavation) present inconsistent characteristics in the monitoring data of different pipe segments or different service stages, resulting in misjudgment or missed detection of the AI model trained based on historical data in real-time monitoring, seriously weakening the reliability of multi-system linkage, and causing the security system to be unable to accurately distinguish real threats from false signals caused by stress effects. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the existing technology, the embodiments of the present invention provide an intelligent security integration platform for oil and gas pipelines and a multi-system linkage method to solve the problems proposed in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: An intelligent security integration multi-system linkage method for oil and gas pipelines, comprising the following steps: S1. Collect multi-node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline; S2. Generate a vibration feature library containing energy determination thresholds corresponding to stress-sensitive frequency bands according to the dynamic characteristic parameters; S3. Capture the micro-crack propagation signal generated by pipeline stress relaxation, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal; S4. Perform time-domain alignment processing on the multi-node vibration data of the same external excitation event based on the weight distribution ratio, and correct the vibration wave propagation path difference in combination with the stress distribution data, and output the corrected vibration features; S5. Input the corrected vibration features into the AI classification model, and output a threat level assessment in combination with the stress distribution data; S6. Generate a collaborative response instruction based on the threat level assessment and the status data of the pipeline stress concentration area.

[0006] In a preferred embodiment, collect multi-node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline, including: Collect multi-node vibration data of the pipeline, including obtaining pipeline axial vibration acceleration data and frequency spectrum distribution data through a distributed fiber optic vibration sensor; Collect stress distribution data of the pipeline, including measuring the circumferential stress value and longitudinal stress gradient data of the pipeline through embedded strain gauges; Based on the collected multi-node vibration data and stress distribution data, calculate the dynamic characteristic parameters of the pipeline through a stress accumulation model. The dynamic characteristic parameters include the pipeline natural frequency offset and the equivalent stiffness coefficient of the vibration wave propagation path.

[0007] In a preferred embodiment, generate a vibration feature library containing the energy determination threshold corresponding to the stress-sensitive frequency band according to the dynamic characteristic parameters, including: Divide the stress-sensitive frequency band based on the pipeline natural frequency offset in the dynamic characteristic parameters. The stress-sensitive frequency band is the frequency interval where the natural frequency offset exceeds the set threshold; Determine the energy determination threshold corresponding to the stress-sensitive frequency band according to the equivalent stiffness coefficient of the vibration wave propagation path in the dynamic characteristic parameters. The energy determination threshold is inversely proportional to the equivalent stiffness coefficient of the vibration wave propagation path; Establish a mapping relationship between the stress-sensitive frequency band and the energy determination threshold and store it as a vibration feature library; Dynamically correct the energy determination threshold in the vibration feature library based on the pipeline service life.

[0008] In a preferred embodiment, capture the micro-crack propagation signal caused by pipeline stress relaxation, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal, including: Capture the micro-crack propagation signal caused by pipeline stress relaxation in real time through an acoustic emission sensor, and distinguish between sudden crack signals and continuous crack signals; For sudden crack signals, calculate the local stress release factor based on the relationship between the signal energy peak and the position of the current pipeline stress concentration area. The local stress release factor is proportional to the signal energy peak and the energy density of the stress concentration area; For continuous crack signals, calculate the cumulative damage factor based on the product of the signal duration and the crack propagation rate; According to the weighted sum of the local stress release factor and the cumulative damage factor, dynamically adjust the weight distribution ratio of the energy determination threshold in the vibration feature library. The weight distribution ratio decays exponentially as the weighted sum increases; The adjusted weight allocation ratio is fed back to the vibration feature library in real time.

[0009] In a preferred embodiment, the sudden crack signal is a single pulse with an amplitude exceeding a preset amplitude threshold, and the continuous crack signal is a periodic signal with an amplitude lower than the preset amplitude threshold but a duration exceeding a set time length.

[0010] In a preferred embodiment, time-domain alignment processing is performed on the vibration data of multiple nodes for the same external excitation event based on the weight allocation ratio, and the vibration wave propagation path difference is corrected by combining stress distribution data, and the corrected vibration features are output, including: Group the vibration data of multiple nodes according to the weight allocation ratio and priority. The frequency band with a weight allocation ratio lower than the set ratio threshold is marked as the high-priority group, and the rest is the low-priority group; For the high-priority group, calculate the equivalent stiffness difference of the propagation path and the crack dynamic interference factor according to the stress distribution data and the micro-crack propagation rate; For the low-priority group, calculate only the equivalent stiffness difference of the propagation path based on the stress distribution data; According to the piecewise linear mapping relationship between the equivalent stiffness difference of the propagation path and the crack dynamic interference factor, dynamically correct the phase compensation time delay and the time-domain amplitude attenuation of the vibration waveforms of each node. The phase compensation time delay of the high-priority group adds the time delay increment corresponding to the crack dynamic interference factor; Perform joint time-frequency domain feature extraction on the corrected vibration waveforms of multiple nodes, fuse the time-domain peak interval variance and the frequency-domain harmonic distortion degree parameters, and generate the corrected vibration features matching the vibration feature library.

[0011] In a preferred embodiment, the equivalent stiffness difference is characterized by the product of the energy density in the stress concentration area and the vibration wave propagation speed, and the crack dynamic interference factor is the ratio of the micro-crack propagation rate to the fracture toughness of the pipeline material.

[0012] In a preferred embodiment, the corrected vibration features are input into the AI classification model, and the threat level assessment is output in combination with the stress distribution data, including: The corrected vibration features and the stress distribution data are spliced into a comprehensive feature vector according to the pipeline segment number. The comprehensive feature vector includes the time-domain peak interval variance, the frequency-domain harmonic distortion degree, and the energy density of the stress concentration area corresponding to the segment; Normalize the comprehensive feature vector, and input the normalized comprehensive feature vector into the pre-trained AI classification model; Dynamically set the confidence threshold of the AI classification model based on the energy density distribution of the stress concentration area in the stress distribution data; Output the threat level assessment through the AI classification model. The threat level includes three levels: normal, warning, and danger.

[0013] In a preferred embodiment, a collaborative response instruction is generated according to the threat level assessment and the status data of the pipeline stress concentration area, including: Generating a collaborative response instruction based on the threat level assessment to match a preset response strategy table, where the response strategy table defines the device control parameters and inspection path parameters corresponding to different threat levels; Dynamically correcting the device control parameters according to the status data of the pipeline stress concentration area, and the correction logic is: if the energy density of the stress concentration area exceeds the set density threshold, the pressure adjustment rate in the device control parameters is reduced proportionally; Performing dynamic priority sorting on the collaborative response instructions, where the priority is determined according to the linear combination value of the threat level and the energy density of the stress concentration area; Issuing the collaborative response instruction to the execution terminal through an industrial communication protocol.

[0014] On the other hand, the present invention provides an intelligent security integration platform for oil and gas pipelines, including: Dynamic characteristic module: Collecting multi-node vibration data and stress distribution data of the pipeline to determine the current dynamic characteristic parameters of the pipeline; Frequency band threshold module: Generating a vibration feature library containing the energy determination threshold corresponding to the stress-sensitive frequency band according to the dynamic characteristic parameters; Crack weight module: Capturing the micro-crack propagation signal generated by the pipeline stress relaxation, and dynamically adjusting the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal; Time domain correction module: Performing time domain alignment processing on the multi-node vibration data of the same external excitation event based on the weight distribution ratio, and correcting the vibration wave propagation path difference in combination with the stress distribution data, and outputting the corrected vibration characteristics; Intelligent evaluation module: Inputting the corrected vibration characteristics into the AI classification model, and outputting the threat level assessment in combination with the stress distribution data; Collaborative response module: Generating a collaborative response instruction according to the threat level assessment and the status data of the pipeline stress concentration area.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By performing time domain alignment and propagation path correction based on multi-node data, the distortion error of vibration signals in different pipe sections is significantly reduced, enabling the AI model to accurately identify the difference between real threats and stress interference; through the multi-dimensional analysis of fusing stress distribution data and vibration characteristics, the threat levels of different regions can be dynamically evaluated, and a hierarchical collaborative response strategy can be generated according to the real-time stress state, greatly improving the timeliness and accuracy of abnormal event handling.

[0016] 2. Compared with static threshold judgment and fixed linkage logic, it realizes the full - process dynamic optimization from data acquisition, feature extraction to response decision - making. Through the weight allocation and propagation path correction driven by micro - crack signals, it can adapt to implicit state changes such as pipeline material aging and stress relaxation, avoiding the model failure problem caused by traditional methods relying on historical data. At the same time, the deep binding of the collaborative instructions of multi - system linkage and the pipeline stress state ensures the accurate matching of disposal measures with the current pipeline risk level, improving the security reliability while reducing the secondary risks caused by misoperations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flowchart of the intelligent security integration multi - system linkage method for oil and gas pipelines of the present invention; Figure 2 is a schematic structural diagram of the intelligent security integration platform for oil and gas pipelines of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] Embodiment 1: Figure 1 The intelligent security integration multi - system linkage method for oil and gas pipelines of the present invention is given, including the following steps: S1. Collect multi - node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline; S2. Generate a vibration feature library containing energy determination thresholds corresponding to stress - sensitive frequency bands according to the dynamic characteristic parameters; S3. Capture the micro - crack propagation signals generated due to pipeline stress relaxation, and dynamically adjust the weight allocation ratio of the energy determination threshold based on the micro - crack propagation signals; S4. Perform time - domain alignment processing on the multi - node vibration data of the same external excitation event based on the weight allocation ratio, and correct the vibration wave propagation path difference in combination with the stress distribution data, and output the corrected vibration features; S5. Input the corrected vibration features into the AI classification model, and output the threat level assessment in combination with the stress distribution data; S6. Generate collaborative response instructions according to the threat level assessment and the state data of the pipeline stress concentration area.

[0020] S1. Collect multi - node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline, including: Collect multi-node vibration data of the pipeline, including obtaining pipeline axial vibration acceleration data and spectrum distribution data through a distributed fiber optic vibration sensor; Collect stress distribution data of the pipeline, including measuring the circumferential stress value and longitudinal stress gradient data of the pipeline through embedded strain gauges; Based on the collected multi-node vibration data and stress distribution data, calculate the dynamic characteristic parameters of the pipeline through a stress accumulation model. The stress accumulation model correlates the mapping relationship between the service life of the pipeline, the material fatigue coefficient, and the vibration transfer attenuation factor. The dynamic characteristic parameters include the pipeline natural frequency offset and the equivalent stiffness coefficient of the vibration wave propagation path.

[0021] Deploy distributed fiber optic vibration sensors on the outer wall of the pipeline at a set spacing. For example, the spacing between adjacent sensors can be set in the range of 1 / 100 to 1 / 50 of the pipeline length. The distributed fiber optic vibration sensor is based on the optical time domain reflectometry principle and obtains pipeline axial vibration acceleration data and spectrum distribution data by detecting the phase change of the backward Rayleigh scattered light in the optical fiber; the data acquisition terminal collects signals at a set sampling frequency. For example, the sampling frequency is not less than 2000 Hz. The pipeline axial vibration acceleration data includes the vibration acceleration amplitude in the pipeline axis direction and its waveform characteristics changing with time. The spectrum distribution data converts the time domain signal into a frequency domain energy distribution through fast Fourier transform. For example, the frequency domain range covers 0 Hz to 5000 Hz.

[0022] Embed resistive strain gauges on the inner wall of the pipeline. The strain gauges are evenly distributed along the circumferential direction of the pipeline and the spacing is a set proportion of the pipeline circumference. For example, the circumferential spacing is 1 / 4 of the circumference. The longitudinal spacing distance of the measurement section can be set as a multiple of the pipeline diameter. For example, the spacing distance is 5 to 10 times the diameter; measure the circumferential stress value and longitudinal stress gradient data of the pipeline in real time through the strain gauges. The circumferential stress value is the circumferential stress component under the action of the internal pressure of the pipeline. The longitudinal stress gradient data is obtained by calculating the ratio of the stress value difference between adjacent measurement sections to the spacing; the strain gauge leads are connected to the data acquisition system, and the data acquisition system performs amplification, filtering, and analog-to-digital conversion processing on the original signal. For example, the filter bandwidth is set to 0 Hz to 2000 Hz to eliminate high-frequency noise interference.

[0023] Based on the collected vibration data of multiple nodes and stress distribution data, calculate the dynamic characteristic parameters of the pipeline through a stress accumulation model. The construction method of the stress accumulation model is as follows: establish the mapping relationship between the service life of the pipeline, the material fatigue coefficient, and the vibration transfer attenuation factor; the service life is the actual operation time of the pipeline, and the material fatigue coefficient is calculated by combining the stress-life curve of the pipeline material with the cumulative damage theory. For example, the rain flow counting method is used to count the stress cycle times of the pipeline historical load data, and the cumulative damage degree is calculated; the vibration transfer attenuation factor is calculated by the spectral energy attenuation rate of the multi-node vibration data. For example, the attenuation rate is the percentage of energy loss of the vibration signal from the starting node to the end node.

[0024] The calculation process of the dynamic characteristic parameters includes: determining the material aging correction coefficient according to the service life. For example, the aging correction coefficient increases exponentially with the increase of the service life, and the exponential base refers to the standard aging rate of the pipeline material; determining the stress relaxation correction factor according to the material fatigue coefficient. For example, the stress relaxation correction factor is proportional to the cumulative damage degree; calculating the natural frequency offset of the pipeline and the equivalent stiffness coefficient of the vibration wave propagation path in combination with the vibration transfer attenuation factor; the natural frequency offset is the deviation percentage between the resonance peak frequency of the measured vibration spectrum and the design value. For example, the deviation is determined by comparing the peak positions of the measured spectrum and the design spectrum; the equivalent stiffness coefficient of the vibration wave propagation path is obtained through the inversion calculation of the stress distribution data and the vibration wave propagation speed. For example, the equivalent elastic modulus of the pipeline is calculated according to the stress distribution data, and the stiffness coefficient is deduced in combination with the elastic wave propagation speed formula.

[0025] During the data acquisition process, the installation positions of the distributed fiber optic vibration sensors avoid the pipeline welds and elbow areas, and the sensors are bonded and fixed to the pipeline surface with epoxy resin glue; the installation positions of the embedded strain gauges are preset during the pipeline manufacturing stage, and the strain gauge leads are connected to the data acquisition system through a waterproof junction box; the parameter calibration of the stress accumulation model is completed through a laboratory accelerated aging test. For example, specimens of the same material as the pipeline are used in the test, and the specimens are subjected to accelerated aging treatment in a high-temperature and high-pressure environment to simulate different service lives, and a standard sine sweep signal is applied through a vibration table to measure the vibration transfer characteristics of the specimens at different aging stages to calibrate the numerical relationship of the attenuation factor; the calculation results of the dynamic characteristic parameters are stored in matrix form, the rows of the matrix correspond to the pipeline section numbers, and the columns correspond to the natural frequency offset and the equivalent stiffness coefficient. The matrix update frequency is dynamically adjusted according to the pipeline operation status. For example, the minimum update frequency is once per hour.

[0026] S2. Generate a vibration feature library containing the energy determination threshold corresponding to the stress-sensitive frequency band according to the dynamic characteristic parameters, including: Divide the stress-sensitive frequency band based on the natural frequency offset of the pipeline in the dynamic characteristic parameters. The stress-sensitive frequency band is the frequency interval where the natural frequency offset exceeds the set threshold; Determine the energy determination threshold corresponding to the stress-sensitive frequency band according to the equivalent stiffness coefficient of the vibration wave propagation path in the dynamic characteristic parameters. The energy determination threshold is inversely proportional to the equivalent stiffness coefficient of the vibration wave propagation path; Establish a mapping relationship between the stress-sensitive frequency band and the energy determination threshold and store it as a vibration characteristic library; Dynamically correct the energy determination threshold in the vibration characteristic library based on the service life of the pipeline. The correction coefficient corresponding to the dynamic correction is calibrated according to the laboratory aging test data, and the correction coefficient increases linearly with the increase of the pipeline service life.

[0027] Divide the stress-sensitive frequency band based on the pipeline natural frequency offset in the dynamic characteristic parameters. The division method is as follows: Take the percentage deviation of the current natural frequency of the pipeline from the design value as the division basis. When the natural frequency offset exceeds the set threshold, for example, the set threshold is 5% of the design value, mark the corresponding frequency range as the stress-sensitive frequency band, and the set threshold is determined according to historical data statistics; The frequency range division is realized through spectrum analysis after fast Fourier transform. The spectrum analysis identifies the resonance peak positions in the vibration signal, and expands a certain bandwidth to both sides of the resonance peak center frequency as the stress-sensitive frequency band. For example, expand 50 Hz to the low-frequency side and 50 Hz to the high-frequency side based on the resonance peak center frequency to form a frequency band.

[0028] The implementation method for determining the energy determination threshold corresponding to the stress-sensitive frequency band according to the equivalent stiffness coefficient of the vibration wave propagation path in the dynamic characteristic parameters, where the energy determination threshold is inversely proportional to the equivalent stiffness coefficient of the vibration wave propagation path, is as follows: Simulate the vibration energy propagation characteristics under different equivalent stiffness coefficients in the laboratory. For example, simulate the equivalent stiffness change by adjusting the specimen support stiffness, measure the energy critical values of the corresponding frequency bands under each stiffness coefficient, and establish an inverse relationship lookup table between the equivalent stiffness coefficient and the energy determination threshold; The lookup table is stored in the form of a two-dimensional array. The row index of the array corresponds to the discretized values of the equivalent stiffness coefficient, the column index corresponds to the stress-sensitive frequency band number, and the array element value is the corresponding energy determination threshold.

[0029] Establish a mapping relationship between the stress-sensitive frequency band and the energy determination threshold and store it as a vibration characteristic library. The establishment method of the mapping relationship is as follows: For each stress-sensitive frequency band, associate the corresponding energy determination threshold to form a key-value pair structure; The vibration characteristic library is stored in the form of a structured query language database table. The database table contains four fields: frequency band number, start frequency, stop frequency, and energy determination threshold. The frequency band number corresponds one-to-one with the pipeline section number in step S1.

[0030] Dynamically correct the energy determination threshold in the vibration feature library based on the service life of the pipeline. The corresponding correction coefficient for dynamic correction is calibrated according to the laboratory aging test data. The calibration method is as follows: Accelerate the aging of specimens made of the same material as the pipeline in the laboratory. For example, place the specimens in an environment with a temperature of 80 °C and a humidity of 90% for 30 days to simulate 1 year of service life. Measure the decay rate of the energy determination threshold before and after the aging of the specimens, and fit the relationship between the correction coefficient and the service life through linear regression. For example, it is measured that for every 1-year increase in service life, the correction coefficient increases by 0.5%. The correction coefficient is applied to the energy determination threshold through linear interpolation. The adjustment formula is: Corrected threshold = Original threshold × (1 + Correction coefficient × Service life).

[0031] During the implementation process, the dynamic correction of the energy determination threshold is synchronized with the update of the dynamic characteristic parameters in step S1. For example, read the latest service life data and update the correction coefficient every hour.

[0032] S3. Capture the microcrack propagation signal generated by the stress relaxation of the pipeline, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the microcrack propagation signal, including: Capture the microcrack propagation signal generated by the stress relaxation of the pipeline in real time through an acoustic emission sensor, and distinguish between sudden crack signals and continuous crack signals; Among them, the sudden crack signal is a single pulse with an amplitude exceeding the preset amplitude threshold, and the continuous crack signal is a periodic signal with an amplitude lower than the preset amplitude threshold but a duration exceeding the set time; For the sudden crack signal, calculate the local stress release factor based on the relationship between the signal energy peak and the position of the current stress concentration area of the pipeline. The local stress release factor is proportional to the signal energy peak and the energy density of the stress concentration area; For the continuous crack signal, calculate the cumulative damage factor based on the product of the signal duration and the crack propagation rate; According to the weighted sum of the local stress release factor and the cumulative damage factor, dynamically adjust the weight distribution ratio of the energy determination threshold in the vibration feature library. The weight distribution ratio decays exponentially as the weighted sum increases; Feed back the adjusted weight distribution ratio to the vibration feature library in real time.

[0033] The micro-crack propagation signals generated by the stress relaxation of the pipeline are captured in real time by acoustic emission sensors. The acoustic emission sensors are deployed along the outer wall of the pipeline at a preset spacing. For example, the spacing between adjacent sensors ranges from 3 to 5 times the pipeline diameter. The sensor signals are transmitted to the data acquisition system through armored cables. The micro-crack propagation signals include sudden crack signals and continuous crack signals. The determination condition for sudden crack signals is that the signal amplitude exceeds the preset amplitude threshold and the duration is less than the set duration. For example, the preset amplitude threshold is 100 mV and the set duration is 0.1 s. The determination condition for continuous crack signals is that the signal amplitude is lower than the preset amplitude threshold but the duration exceeds the set duration. For example, the duration exceeds 10 s and the signal waveform shows periodic fluctuations.

[0034] For sudden crack signals, the local stress release factor is calculated based on the relationship between the signal energy peak and the position of the current stress concentration area of the pipeline. The signal energy peak is the product of the square of the maximum amplitude of the sudden crack signal in the time domain and the duration. The energy density of the current stress concentration area of the pipeline is calculated from the stress distribution data collected in step S1. For example, the energy density of the stress concentration area is the product of the square of the stress value in this area and the area. The local stress release factor is the product of the signal energy peak and the energy density.

[0035] For continuous crack signals, the cumulative damage factor is calculated based on the product of the signal duration and the crack propagation rate. The crack propagation rate is calculated from the rising edge slope of the micro-crack propagation signal. For example, the rising edge slope is the reciprocal of the time required for the signal amplitude to rise from 10% of the peak to 90% of the peak. If the time is 0.02 s, the slope is 1 / 0.02 = 50. The cumulative damage factor is the product of the signal duration and the crack propagation rate divided by the reference fatigue life value of the pipeline material. For example, the reference fatigue life value is determined by the number of cycles in the material S-N curve at the set stress amplitude. If the S-N curve shows that the stress amplitude of 100 MPa corresponds to the number of cycles of 6 10 6 cycles, then the reference fatigue life value is 10

[0036] Dynamically adjust the weight allocation ratio of the energy determination threshold in the vibration feature library according to the weighted sum of the local stress release factor and the cumulative damage factor. The calculation method of the weighted sum is the local stress release factor multiplied by the first weight coefficient plus the cumulative damage factor multiplied by the second weight coefficient. For example, the first weight coefficient is 0.6 and the second weight coefficient is 0.4. The weight allocation ratio decays exponentially as the weighted sum increases. For example, when the weighted sum increases by 10%, the weight allocation ratio decreases by 15%. The exponential decay relationship is realized through a pre-generated look-up table that stores the mapping relationship between the weighted sum and the weight allocation ratio. The construction of the look-up table is based on laboratory simulation test data. For example, in the laboratory, by changing the crack state of the specimen and measuring the corresponding change in weight sensitivity, a mapping relationship table of the weighted sum and the weight decay ratio is generated.

[0037] Real-time feedback the adjusted weight allocation ratio to the vibration feature library to update the mapping relationship between the frequency band and the energy determination threshold. The update method of the vibration feature library is: adjust the energy determination threshold of the corresponding frequency band according to the weight allocation ratio. For example, when the weight allocation ratio decreases by 15%, the energy determination threshold of the corresponding frequency band floats up by 15%. The updated vibration feature library is linked with the time-domain alignment process in step S4. In the time-domain alignment process, the data of the frequency band with a weight allocation ratio lower than the set threshold is preferentially processed. For example, the set threshold is 70% of the original weight, and the frequency band lower than this threshold is allocated higher computing resources during time-domain alignment.

[0038] During the implementation process, the signal acquisition frequency of the acoustic emission sensor is consistent with the vibration data sampling rate in step S1. For example, the sampling rate is 2000 Hz. The determination threshold of the micro-crack propagation signal is dynamically adjusted according to the pipeline material type. For example, the preset amplitude threshold for carbon steel pipelines is 100 mV, and for stainless steel pipelines is 80 mV. The calibration of the weight coefficient and the fatigue life reference value is completed through laboratory accelerated aging tests. For example, in the test, specimens of the same material as the pipeline are used to simulate the crack propagation state of different service years under the conditions of a temperature of 80 °C and a pressure of 10 MPa, measure the actual influence weights of the local stress release factor and the cumulative damage factor, and fit the best ratio of the first weight coefficient of 0.6 and the second weight coefficient of 0.4.

[0039] The update period of the weight allocation ratio in the vibration feature library is synchronized with the update frequency of the dynamic characteristic parameters in step S1. For example, the weight adjustment is performed once per hour. The adjusted weight allocation ratio ensures data consistency through the database transaction mechanism. For example, the ACID transaction characteristics are used to ensure the atomic operation of the time-domain alignment process and the feature library update. The determination logic of the micro-crack propagation signal is compatible with the input features of the AI classification model in step S5. For example, the AI model classifies the modified vibration features according to the weight allocation ratio, and the feature of the frequency band with a weight allocation ratio lower than the set threshold is given a higher confidence weight during classification.

[0040] S4. Align the vibration data of multiple nodes for the same external excitation event in the time domain based on the weight distribution ratio, correct the difference in the vibration wave propagation path by combining the stress distribution data, and output the corrected vibration characteristics, including: Group the vibration data of multiple nodes according to the priority based on the weight distribution ratio. The frequency band with a weight distribution ratio lower than the set ratio threshold is marked as the high-priority group, and the rest is the low-priority group; For the high-priority group, calculate the equivalent stiffness difference of the propagation path and the crack dynamic interference factor according to the stress distribution data and the microcrack propagation rate. The equivalent stiffness difference is characterized by the product of the energy density in the stress concentration area and the vibration wave propagation speed. The crack dynamic interference factor is the ratio of the microcrack propagation rate to the fracture toughness of the pipeline material; For the low-priority group, calculate only the equivalent stiffness difference of the propagation path based on the stress distribution data; According to the piecewise linear mapping relationship between the equivalent stiffness difference of the propagation path and the crack dynamic interference factor, dynamically correct the phase compensation time delay and the time-domain amplitude attenuation amount of the vibration waveforms of each node. The phase compensation time delay of the high-priority group adds the time delay increment corresponding to the crack dynamic interference factor; Perform joint time-frequency domain feature extraction on the corrected vibration waveforms of multiple nodes, fuse the time-domain peak interval variance and the frequency-domain harmonic distortion degree parameters, and generate the corrected vibration characteristics that match the vibration feature library.

[0041] Group the vibration data of multiple nodes according to the priority based on the weight distribution ratio. The frequency band with a weight distribution ratio lower than the set ratio threshold is marked as the high-priority group, and the rest is the low-priority group; The set ratio threshold is determined by statistical analysis of historical alarm data. For example, statistically analyze the distribution law of the weight distribution ratio in false alarm events in the past year, and select the weight value corresponding to the sudden increase in the false alarm rate as the threshold; The grouping determination is implemented by a comparator circuit. The comparator circuit receives the weight distribution ratio signal output in step S3 and triggers the high-priority group marking signal when the input signal is lower than the set ratio threshold.

[0042] For the high-priority group, calculate the equivalent stiffness difference of the propagation path and the crack dynamic interference factor according to the stress distribution data collected in step S1 and the microcrack propagation rate; The equivalent stiffness difference is characterized by the product of the energy density in the stress concentration area and the vibration wave propagation speed. The energy density is the product of the square of the stress value in this area and the acting area; The vibration wave propagation speed is calculated by the time difference method of the vibration data of multiple nodes in step S1. For example, calculate the propagation speed according to the time difference of the vibration wave arrival time between node A and node B and the distance; The crack dynamic interference factor is the ratio of the microcrack propagation rate to the fracture toughness of the pipeline material. The fracture toughness value is extracted from the material property database, and the microcrack propagation rate is the real-time value calculated by the rising edge slope of the acoustic emission signal in step S3.

[0043] For the low-priority group, the equivalent stiffness difference of the propagation path is calculated only based on the stress distribution data in step S1. The calculation method is the same as that of the high-priority group, but there is no need to introduce the crack dynamic interference factor. The calculation resource allocation ratio of the low-priority group is 30% of that of the high-priority group. For example, if the high-priority group allocates 70% of the computing resources for real-time correction, the low-priority group allocates 21%.

[0044] According to the piecewise linear mapping relationship between the equivalent stiffness difference of the propagation path and the crack dynamic interference factor, the phase compensation time delay and the time-domain amplitude attenuation of the vibration waveforms of each node are dynamically corrected. The piecewise linear mapping relationship is established through laboratory calibration tests. In the tests, specimens of the same material as the pipeline are used, different stress levels are applied, and micro-cracks are artificially prefabricated. The waveform distortion under the combination of the equivalent stiffness difference and the crack dynamic interference factor is measured, and three interference intervals of low, medium, and high are divided and the corresponding compensation parameters are determined. For example, the low interference interval corresponds to an equivalent stiffness difference < 100 kPa·m / s and a crack dynamic interference factor < 0.1, and the compensation parameters are a time delay increment of 5 ms and an amplitude attenuation of 5%. The medium interference interval corresponds to an equivalent stiffness difference of 100 - 300 kPa·m / s or a crack dynamic interference factor of 0.1 - 0.3, and the compensation parameters are a time delay increment of 10 ms and an amplitude attenuation of 10%. The high interference interval corresponds to an equivalent stiffness difference > 300 kPa·m / s or a crack dynamic interference factor > 0.3, and the compensation parameters are a time delay increment of 20 ms and an amplitude attenuation of 20%.

[0045] Perform joint time-frequency domain feature extraction on the corrected multi-node vibration waveforms. The time-domain peak interval variance is obtained by statistically analyzing the fluctuation degree of the time intervals between adjacent vibration peaks. For example, peak points exceeding the set amplitude threshold in the waveform are selected, and the standard deviation of their time intervals is calculated. The frequency-domain harmonic distortion is determined by analyzing the energy ratio of the fundamental wave and the third harmonic through fast Fourier transform. For example, when the fundamental wave energy is 100 units and the third harmonic energy is 15 units, the harmonic distortion is 15%. The time-domain peak interval variance and the frequency-domain harmonic distortion are fused according to the preset weights to generate a comprehensive feature parameter. For example, if the time-domain weight is 60% and the frequency-domain weight is 40%, then the comprehensive feature = time-domain peak interval variance × 0.6 + harmonic distortion × 0.4.

[0046] During the implementation process, the phase compensation time delay increment is realized through the programmable delay module of the digital signal processor, and the time delay accuracy is controlled within 0.1 ms. The time-domain amplitude attenuation is adjusted by a digital controlled attenuator, and the attenuation coefficient error does not exceed 1%. The interval boundaries and compensation parameters of the piecewise linear mapping relationship are stored in an electrically erasable memory, and support on-site update through the calibration interface. The time-frequency domain feature fusion weights are dynamically adjusted according to the pipeline operation environment. For example, in a low-temperature environment (< -20°C), the time-domain weight is increased to 70% to enhance the detection sensitivity to brittle fracture of the material. The corrected vibration characteristics are transmitted to the AI classification model in step S5 through a standard communication protocol, and the characteristic format is fully compatible with the field definitions of the vibration characteristic library, for example, including the frequency band number, the variance of the time-domain peak interval, the frequency-domain harmonic distortion degree, and the fusion characteristic value.

[0047] S5. Input the corrected vibration characteristics into the AI classification model, and output a threat level assessment in combination with the stress distribution data, including: Concatenate the corrected vibration characteristics and the stress distribution data according to the pipeline section number to form a comprehensive feature vector, which includes the variance of the time-domain peak interval, the frequency-domain harmonic distortion degree, and the energy density of the stress concentration area corresponding to the section; Perform normalization processing on the comprehensive feature vector, and input the normalized comprehensive feature vector into the pre-trained AI classification model; Based on the energy density distribution of the stress concentration area in the stress distribution data, dynamically set the confidence threshold of the AI classification model; Output a threat level assessment through the AI classification model, and the threat levels include three levels: normal, warning, and danger.

[0048] Concatenate the corrected vibration characteristics and the stress distribution data collected in step S1 according to the pipeline section number to form a comprehensive feature vector, which includes the variance of the time-domain peak interval, the frequency-domain harmonic distortion degree, and the energy density of the stress concentration area corresponding to the section; the unit of the variance of the time-domain peak interval is square seconds (s²), which is obtained by calculating the variance of the time intervals between adjacent vibration peaks. For example, if the time interval sequence is [1.0 s, 1.2 s, 0.9 s], the variance is calculated as the average of the squared differences between each data and the mean; the frequency-domain harmonic distortion degree is expressed as a percentage and is calculated by the ratio of the fundamental wave energy to the third harmonic energy. For example, when the fundamental wave energy is 100 units and the third harmonic is 15 units, the distortion degree is 15%; the unit of the energy density of the stress concentration area is kilojoules per cubic meter (kJ / m³), which is calculated by multiplying the square of the stress value by the acting area and dividing by the volume.

[0049] Perform normalization processing on the comprehensive feature vector, and the normalization parameters are dynamically adjusted according to the energy determination threshold of the stress-sensitive frequency band in the vibration characteristic library generated in step S2; for example, if the energy determination threshold of a certain stress-sensitive frequency band is 1000 millivolt square seconds (mV²·s), the variance of the time-domain peak interval corresponding to the frequency band is scaled according to the threshold ratio, specifically, the variance value is divided by the threshold to obtain the normalization result; the energy density of the stress concentration area is normalized according to the material yield strength. For example, if the material yield strength is 500 megapascals (MPa), the energy density of 500 kJ / m³ is normalized to 500 / 500 = 1; the normalized feature vector is input into the pre-trained AI classification model, and the AI classification model adopts a support vector machine architecture, and the training data is a combination of vibration characteristics and stress data labeled under historical working conditions.

[0050] Dynamically set the confidence threshold of the AI classification model based on the energy density distribution in the stress concentration area of the stress distribution data; the implementation method for the confidence threshold to be inversely proportional to the energy density is as follows: when the energy density in the stress concentration area exceeds the set critical value, the confidence threshold is proportionally lowered to reduce false negatives. For example, when the energy density critical value is 800 kJ / m³ and the measured energy density is 1000 kJ / m³, the confidence threshold is lowered from the default 90% to 80%; the adjustment step size of the confidence threshold is set according to the pipeline safety level. For example, the adjustment step size for a first-level safety pipeline is 5%, and for a second-level pipeline is 10%.

[0051] Output the threat level assessment through the AI classification model. The threat level is divided into three levels: normal, warning, and danger; the judgment logic of the assessment result is as follows: if the output probability value of the model is lower than the confidence threshold and the energy density is lower than the critical value, it is judged as normal; if the output probability value is lower than the threshold but the energy density exceeds the critical value, it is judged as a warning; if the output probability value is higher than the threshold, it is directly judged as dangerous; the assessment results are stored in the distributed database according to the pipeline segment number. The database table structure includes fields such as timestamp, segment number, threat level, and original feature vector.

[0052] During the implementation process, the training data of the AI classification model comes from the historical pipeline operation database, and the training data covers different stress states, crack propagation stages, and external interference event scenarios; when training the model, higher weights are assigned to the data in the high-stress area. For example, the weight of data samples with an energy density exceeding 500 kJ / m³ is increased by 50%; the dynamic adjustment logic of the confidence threshold is implemented through a programmable logic controller, and the adjustment parameters are stored in a non-volatile memory, supporting on-site modification through a calibration tool; the storage format of the threat level assessment results is compatible with the corrected vibration feature format in step S4, facilitating historical data traceability and model optimization.

[0053] S6. Generate collaborative response instructions based on the threat level assessment and the status data of the pipeline stress concentration area, including: Generate collaborative response instructions by matching the preset response strategy table based on the threat level assessment. The response strategy table defines the equipment control parameters and inspection path parameters corresponding to different threat levels; Dynamically correct the equipment control parameters according to the status data of the pipeline stress concentration area. The correction logic is as follows: if the energy density in the stress concentration area exceeds the set density threshold, the pressure adjustment rate in the equipment control parameters is proportionally reduced; Perform dynamic priority sorting on the collaborative response instructions. The priority is determined according to the linear combination value of the threat level and the energy density in the stress concentration area; Send the collaborative response instructions to the execution terminal through the industrial communication protocol.

[0054] Based on the threat level assessment result output in step S5, a collaborative response instruction is generated by matching a preset response strategy table. The response strategy table is stored in the form of a database table. The fields of the database table include threat level, device control parameters, and inspection path parameters. The device control parameters include the pressure adjustment rate and the valve opening gradient. The inspection path parameters include inspection trajectory coordinate points and residence time. For example, when the threat level is "dangerous", the pressure adjustment rate in the device control parameters is set to decrease by 5% per second, and a detour trajectory point 50 meters away from the stress concentration area is added to the inspection path parameters.

[0055] Dynamically correct the device control parameters according to the status data of the pipeline stress concentration area collected in step S1. The correction logic is as follows: If the energy density in the stress concentration area exceeds the set density threshold, the pressure adjustment rate in the device control parameters is reduced proportionally. The set density threshold is calibrated through laboratory blasting tests. For example, when the energy density exceeds 800 kJ / m³, the pressure adjustment rate is reduced to 70% of the original value. During the correction process, the reduction ratio of the pressure adjustment rate is proportional to the amplitude of the energy density exceeding the threshold. For example, for every 100 kJ / m³ increase in the energy density exceeding the threshold, the rate is reduced by 10%.

[0056] Perform dynamic priority sorting on the collaborative response instructions. The priority is determined according to the linear combination value of the threat level and the energy density of the stress concentration area. The calculation method of the linear combination value is the weighted sum of the threat level coefficient and the energy density coefficient. For example, the threat level "dangerous" is assigned a value of 3, "warning" is 2, and "normal" is 1. The energy density coefficient is the percentage of the actual value divided by the set density threshold. The linear combination value = threat level coefficient × 0.6 + energy density coefficient × 0.4. The priority queue is sorted from high to low according to the linear combination value, and high-priority instructions are sent first.

[0057] Send the collaborative response instruction to the execution terminal through an industrial communication protocol. The industrial communication protocol adopts the Modbus RTU or OPC UA standard. The instruction data packet format includes the target device address, control parameter value, and check code. For example, the pressure adjustment instruction format is [device address: V001, parameter type: pressure adjustment rate, parameter value: 3% / second, check code: CRC16]. The instruction execution status is fed back to step S1 in real time. The feedback data includes the instruction execution progress and the real-time status of the device, such as the current valve opening and the current position coordinates of the drone.

[0058] During implementation, the construction of the response strategy table is based on the historical emergency response case library. Each case in the case library is labeled with a threat level, the effect of control parameters, and the effectiveness score of path inspection; the dynamic correction period of the device control parameters is synchronized with the data acquisition frequency in step S1. For example, the latest energy density data is read every minute and the pressure adjustment rate is adjusted; the calculation results of the priority sorting are cached through a memory queue to ensure the real-time insertion of high-priority instructions and the delayed execution of low-priority instructions.

[0059] The execution terminals of the collaborative response instructions include an intelligent valve controller and an autonomous inspection drone. The intelligent valve controller realizes precise control of the pressure adjustment rate through a proportional-integral-derivative algorithm. The drone navigation generates an obstacle avoidance trajectory according to the inspection path parameters; before the instruction is issued, the data verification module verifies the parameter legality. For example, it checks whether the pressure adjustment rate exceeds the safe range of the device and whether the inspection path coordinates are within the pipeline geographical fence; the execution status feedback data is associated and stored with the threat level evaluation result in step S5 for subsequent accident tracing and model optimization.

[0060] It should be noted that the pipeline mentioned in this embodiment is an oil and gas pipeline.

[0061] Embodiment 2: Figure 2 The structural schematic diagram of the intelligent security integration platform for oil and gas pipelines of the present invention is given. The intelligent security integration platform for oil and gas pipelines includes: Dynamic characteristic module: Collect multi-node vibration data and stress distribution data of the pipeline to determine the current dynamic characteristic parameters of the pipeline; Frequency band threshold module: Generate a vibration feature library containing the energy determination threshold corresponding to the stress-sensitive frequency band according to the dynamic characteristic parameters; Crack weight module: Capture the micro-crack propagation signal generated by the stress relaxation of the pipeline, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal; Time domain correction module: Perform time domain alignment processing on the multi-node vibration data of the same external excitation event based on the weight distribution ratio, and correct the vibration wave propagation path difference in combination with the stress distribution data, and output the corrected vibration characteristics; Intelligent evaluation module: Input the corrected vibration characteristics into the AI classification model, and output the threat level evaluation in combination with the stress distribution data; Collaborative response module: Generate collaborative response instructions according to the threat level evaluation and the status data of the stress concentration area of the pipeline.

[0062] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by software simulation of a large amount of collected data to obtain a formula closest to the real situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0063] It should be noted that the present invention can be deployed on the device itself to implement embedded applications, or can also run on a PC or other terminal with a user interface, so as to meet various hardware environments and usage requirements.

[0064] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wire (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a set of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0065] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0066] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some interfaces, and the indirect couplings or communication connections of the devices or modules can be in electrical, mechanical or other forms.

[0067] The module described as a separation component may or may not be physically separated. The component shown as a module may or may not be a physical module. It may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0068] In addition, in each embodiment of this application, each functional module can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0069] If the described function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this 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 for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0070] The above description is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0071] Finally: The above description is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An integrated multi-system linkage method for intelligent security of oil and gas pipelines, characterized in that, It includes the following steps: S1. Collect multi-node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline; S2. Generate a vibration feature library containing energy determination thresholds corresponding to stress-sensitive frequency bands based on the dynamic characteristic parameters; S3. Capture the micro-crack propagation signal caused by pipeline stress relaxation, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal; S4. Perform time-domain alignment processing on the multi-node vibration data of the same external excitation event based on the weight distribution ratio, and correct the vibration wave propagation path difference by combining the stress distribution data, and output the corrected vibration characteristics; S5. Input the corrected vibration characteristics into the AI classification model, and output the threat level assessment by combining the stress distribution data; S6. Generate a collaborative response instruction according to the threat level assessment and the state data of the pipeline stress concentration area.

2. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, wherein Collect multi-node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline, including: Collect multi-node vibration data of the pipeline, including obtaining pipeline axial vibration acceleration data and spectrum distribution data through a distributed fiber optic vibration sensor; Collect stress distribution data of the pipeline, including measuring the circumferential stress value and longitudinal stress gradient data of the pipeline through embedded strain gauges; Based on the collected multi-node vibration data and stress distribution data, calculate the dynamic characteristic parameters of the pipeline through a stress accumulation model. The dynamic characteristic parameters include the pipeline natural frequency offset and the equivalent stiffness coefficient of the vibration wave propagation path.

3. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, wherein Generate a vibration feature library containing energy determination thresholds corresponding to stress-sensitive frequency bands based on the dynamic characteristic parameters, including: Divide the stress-sensitive frequency band based on the pipeline natural frequency offset in the dynamic characteristic parameters. The stress-sensitive frequency band is the frequency range where the natural frequency offset exceeds the set threshold; Determine the energy determination threshold corresponding to the stress-sensitive frequency band according to the equivalent stiffness coefficient of the vibration wave propagation path in the dynamic characteristic parameters. The energy determination threshold is inversely proportional to the equivalent stiffness coefficient of the vibration wave propagation path; Establish a mapping relationship between the stress-sensitive frequency band and the energy determination threshold and store it as a vibration feature library; Dynamically correct the energy determination threshold in the vibration feature library based on the pipeline service life.

4. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, characterized in that, Capture the micro-crack propagation signal caused by pipeline stress relaxation, and dynamically adjust the weight distribution ratio of the energy determination threshold based on the micro-crack propagation signal, including: Capture the micro-crack propagation signal caused by pipeline stress relaxation in real time through an acoustic emission sensor, and distinguish between sudden crack signals and continuous crack signals; For sudden crack signals, calculate the local stress release factor based on the relationship between the signal energy peak and the position of the current pipeline stress concentration area. The local stress release factor is proportional to the signal energy peak and the energy density of the stress concentration area; For continuous crack signals, calculate the cumulative damage factor based on the product of the signal duration and the crack propagation rate; According to the weighted sum of the local stress release factor and the cumulative damage factor, dynamically adjust the weight distribution ratio of the energy determination threshold in the vibration feature library. The weight distribution ratio decays exponentially as the weighted sum increases; Real-time feedback the adjusted weight distribution ratio to the vibration feature library.

5. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 4, wherein, The sudden crack signal is a single pulse with an amplitude exceeding the preset amplitude threshold, and the continuous crack signal is a periodic signal with an amplitude lower than the preset amplitude threshold but a duration exceeding the set time length.

6. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, characterized in that Perform time-domain alignment processing on the multi-node vibration data of the same external excitation event based on the weight distribution ratio, and correct the vibration wave propagation path difference by combining the stress distribution data, and output the corrected vibration characteristics, including: Group the multi-node vibration data according to the priority based on the weight distribution ratio. The frequency band with a weight distribution ratio lower than the set ratio threshold is marked as the high-priority group, and the rest is the low-priority group; For the high-priority group, calculate the equivalent stiffness difference of the propagation path and the crack dynamic interference factor according to the stress distribution data and the micro-crack propagation rate; For the low-priority group, calculate the equivalent stiffness difference of the propagation path only based on the stress distribution data; According to the piecewise linear mapping relationship between the equivalent stiffness difference of the propagation path and the crack dynamic interference factor, dynamically correct the phase compensation time delay and the time-domain amplitude attenuation amount of each node's vibration waveform. The phase compensation time delay of the high-priority group adds the time delay increment corresponding to the crack dynamic interference factor; Perform time-frequency domain joint feature extraction on the corrected multi-node vibration waveforms, fuse the time-domain peak interval variance and the frequency-domain harmonic distortion degree parameters, and generate the corrected vibration characteristics matching the vibration feature library.

7. The integrated multi-system linkage method for intelligent safety protection of oil and gas pipelines according to claim 6, characterized in that The equivalent stiffness difference is characterized by the product of the energy density in the stress concentration area and the vibration wave propagation speed, and the crack dynamic interference factor is the ratio of the micro-crack propagation rate to the fracture toughness of the pipeline material.

8. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, wherein Input the corrected vibration characteristics into the AI classification model, and output the threat level assessment by combining the stress distribution data, including: Splice the corrected vibration characteristics and the stress distribution data according to the pipeline segment number to form a comprehensive feature vector, which includes the time-domain peak interval variance, the frequency-domain harmonic distortion degree, and the energy density of the stress concentration area corresponding to the segment; Perform normalization processing on the comprehensive feature vector, and input the normalized comprehensive feature vector into the pre-trained AI classification model; Dynamically set the confidence threshold of the AI classification model based on the energy density distribution in the stress concentration area of the stress distribution data; Output the threat level assessment through the AI classification model. The threat levels include three levels: normal, warning, and danger.

9. The intelligent security integration multi-system linkage method for oil and gas pipelines according to claim 1, wherein Generate a collaborative response instruction according to the threat level assessment and the status data of the pipeline stress concentration area, including: Generate a collaborative response instruction based on the threat level assessment by matching the preset response strategy table. The response strategy table defines the device control parameters and inspection path parameters corresponding to different threat levels; Dynamically correct the device control parameters according to the status data of the pipeline stress concentration area. The correction logic is: if the energy density in the stress concentration area exceeds the set density threshold, the pressure adjustment rate in the device control parameters is reduced proportionally; Perform dynamic priority sorting on the collaborative response instructions, and the priority is determined according to the linear combination value of the threat level and the energy density in the stress concentration area; Send the collaborative response instruction to the execution terminal through the industrial communication protocol.

10. An intelligent security integration platform for oil and gas pipelines, which is used to implement the intelligent security integration multi-system linkage method for oil and gas pipelines described in any one of claims 1-9, and is characterized in that, Including: Dynamic characteristic module: Collect multi-node vibration data and stress distribution data of the pipeline to determine the dynamic characteristic parameters of the current pipeline; Frequency band threshold module: Generate a vibration feature library containing energy determination thresholds corresponding to stress-sensitive frequency bands according to dynamic characteristic parameters; Crack weight module: Capture the micro-crack propagation signals generated by pipeline stress relaxation, and dynamically adjust the weight allocation ratio of the energy determination threshold based on the micro-crack propagation signals; Time-domain correction module: Perform time-domain alignment processing on multi-node vibration data of the same external excitation event based on the weight allocation ratio, correct the vibration wave propagation path difference in combination with stress distribution data, and output the corrected vibration characteristics; Intelligent evaluation module: Input the corrected vibration characteristics into the AI classification model, and output the threat level evaluation in combination with stress distribution data; Collaborative response module: Generate collaborative response instructions according to the threat level evaluation and the status data of the pipeline stress concentration area.

Images