A method, system and device for monitoring the progress of a pipe cleaning detection device

Abstract

The present invention discloses a method, system and equipment for monitoring the progress of a pipe cleaning detection device, which relates to the field of oil and gas transportation. The present invention continuously collects vibration signal data in the pipeline through a Φ-OTDR sensor installed in the pipeline; uses data mining technology to analyze the vibration signal data of each defense zone in the pipeline, determines the time domain trigger threshold of each defense zone, and then locks the event defense zone suspected of having a vibration event through a time domain triggering method; performs correlation and refinement analysis on the adjacent defense zones of the event defense zone, and calculates the vibration characteristics of the adjacent defense zones; uses image recognition technology to convert the vibration characteristics into image characteristics, and combines 3D multiple convolutional neural network technology to perform deep learning on the image characteristics to determine the travel position of the pipe cleaning detection device; calculates the travel speed of the pipe cleaning detection device based on the travel position, and issues an early warning when the travel speed is abnormal. The present invention can greatly improve the pipe cleaning efficiency of the pipe cleaning detection equipment and the safety and reliability of oil and gas pipeline transportation.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas transportation, and in particular to a traveling monitoring method, system and equipment for pigging detection equipment. Background Art

[0002] Oil and gas pipelines play a vital role in energy production and transportation, requiring a high degree of safety and integrity. In-line inspection technology is a key method for diagnosing and maintaining pipelines. It can detect pipeline defects and problems, playing a vital role in the safe management and maintenance of long-distance oil and gas pipelines. Major oil and gas pipeline accidents in recent years have triggered higher demands for pipeline safety. While in-line inspection technology is widely used in oil and gas pipeline production and operation, existing technologies suffer from various issues, including inaccurate tracking and positioning of pigging and inspection equipment, reliance on manual intervention for monitoring, and low data reliability. Summary of the Invention

[0003] In response to the problems raised in the above background technology, the present invention provides a method, system and equipment for monitoring the progress of a pipeline cleaning detection device to achieve real-time monitoring of the progress status of the pipeline cleaning detection device in the pipeline and abnormal warning, thereby improving the pipeline cleaning efficiency and pipeline safety.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] In one aspect, the present invention provides a method for monitoring the progress of a pigging detection device, comprising:

[0006] The Φ-OTDR sensor installed in the pipeline continuously collects vibration signal data in the pipeline;

[0007] Use data mining technology to analyze the vibration signal data of each defense zone in the pipeline and determine the time domain trigger threshold of each defense zone;

[0008] Based on the time domain trigger threshold of each defense zone, the event defense zone suspected of vibration events is locked through time domain triggering;

[0009] Conduct correlation and refinement analysis on adjacent defense zones of the event defense zone and calculate the vibration characteristics of adjacent defense zones;

[0010] Image recognition technology is used to convert vibration features into image features. 3D multi-convolutional neural network technology is then used to perform deep learning on these features to determine the travel position of the pigging inspection equipment.

[0011] The travel speed of the pigging detection equipment is calculated based on the travel position, and an abnormal warning is issued when the travel speed is abnormal.

[0012] Optionally, the continuously collecting vibration signal data in the pipeline by a Φ-OTDR sensor installed in the pipeline specifically includes:

[0013] Multiple Φ-OTDR sensors installed in the pipeline are connected to the computer via optical cables laid in the same trench as the pipeline;

[0014] The Φ-OTDR sensor continuously collects vibration signal data in the pipeline for 7-15 days.

[0015] Optionally, the method of analyzing the vibration signal data of each defense zone in the pipeline by using data mining technology to determine the time domain trigger threshold of each defense zone specifically includes:

[0016] The variance of the vibration signal data for each zone in the pipeline Statistical analysis was performed to analyze the variance distribution of each defense zone for 7-15 consecutive days, and a rough time-domain trigger threshold for each defense zone was set using Gaussian normal distribution. Where X is the single vibration signal data of the defense zone; μ is the overall mean of all vibration signal data of the defense zone; and F is the total number of cases.

[0017] Fine-tune the rough time domain trigger threshold according to the actual vibration occurring during the movement of the pigging detection equipment, and precisely set the time domain trigger threshold for each defense zone.

[0018] Optionally, performing correlation and refinement analysis on adjacent defense zones of the event defense zone and calculating vibration characteristics of the adjacent defense zones specifically includes:

[0019] The adjacent defense zones of the event defense zone are subjected to correlation and refinement analysis, and four vibration characteristics of the adjacent defense zones, namely, square difference, short-time Fourier transform, short-time level crossing rate and disturbance duration, are calculated. The calculation formula of the average level crossing rate L in each time period is: N is the number of vibration signal data points in each time period; I(n) is the amplitude of the nth vibration signal data point; α is the set level threshold; ψ is the indicator function, which is 1 when the condition in the brackets is met and 0 otherwise.

[0020] Optionally, the method of converting the vibration features into image features using image recognition technology and performing deep learning on the image features in combination with 3D multi-convolutional neural network technology to determine the travel position of the pigging detection equipment specifically includes:

[0021] The four vibration eigenvalues of each defense zone are normalized to represent the vibration intensity of each defense zone;

[0022] Use points of different colors to represent the vibration intensity of each defense zone, and then plot the vibration intensity of each defense zone on the entire optical cable into a waterfall chart based on time and space;

[0023] The vibration intensity of different color depths on the waterfall chart is used as image features, and deep learning is performed using a 3D multi-convolutional neural network to determine the travel position of the pipe cleaning inspection equipment.

[0024] On the other hand, the present invention also provides a travel monitoring system for a pigging detection device, comprising:

[0025] A data acquisition module is used to continuously collect vibration signal data in the pipeline through a Φ-OTDR sensor installed in the pipeline;

[0026] A data mining module is used to analyze the vibration signal data of each defense zone in the pipeline using data mining technology to determine the time domain trigger threshold of each defense zone;

[0027] The time domain trigger module is used to lock the event zone suspected of vibration events through time domain triggering based on the time domain trigger threshold of each zone;

[0028] Adjacent defense zone association and vibration feature extraction module, used to perform association and refinement analysis on the adjacent defense zones of the event defense zone and calculate the vibration features of the adjacent defense zones;

[0029] The path tracking module uses image recognition technology to convert vibration features into image features, and combines 3D multi-convolutional neural network technology to perform deep learning on image features to determine the travel position of the pigging inspection equipment;

[0030] The travel status monitoring and early warning module is used to calculate the travel speed of the pipe cleaning detection equipment according to the travel position, and issue an abnormal early warning when the travel speed is abnormal.

[0031] On the other hand, 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 implements the travel monitoring method of the pipe cleaning detection device when executing the computer program.

[0032] Optionally, the memory is a non-transitory computer-readable storage medium.

[0033] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0034] The present invention provides a travel monitoring method, system and equipment for pipe cleaning detection equipment, which continuously collect vibration signal data in the pipeline through a Φ-OTDR sensor installed in the pipeline; use data mining technology to analyze the vibration signal data of each defense zone in the pipeline to determine the time domain trigger threshold of each defense zone; based on the time domain trigger threshold of each defense zone, lock the event defense zone where a vibration event is suspected to exist through a time domain triggering method; perform correlation and refinement analysis on adjacent defense zones of the event defense zone, and calculate the vibration characteristics of the adjacent defense zones; use image recognition technology to convert the vibration characteristics into image characteristics, and combine 3D multi-convolutional neural network technology to perform deep learning on the image characteristics to determine the travel position of the pipe cleaning detection equipment; calculate the travel speed of the pipe cleaning detection equipment according to the travel position, and issue an abnormality warning when the travel speed is abnormal, which can greatly improve the pipe cleaning efficiency of the pipe cleaning detection equipment and the safety and reliability of oil and gas pipeline transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 This is a flow chart of a method for monitoring the progress of a pigging detection device according to the present invention;

[0037] Figure 2 Schematic diagram of the monitoring process of the pigging detection device of the present invention;

[0038] Figure 3 Schematic diagram of the process of extracting signal frequency features using short-time level transition rate. DETAILED DESCRIPTION

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0040] To ensure the safe operation of pipelines, more intelligent, high-tech means are needed to achieve real-time online monitoring and feedback of pigging detection equipment. This will help reduce the risk of potential pipeline accidents and protect people's lives and property. Therefore, the purpose of the present invention is to provide a method, system, and device for monitoring the progress of pigging detection equipment in pipelines to achieve real-time monitoring of the progress of pigging detection equipment in pipelines and provide early warning of abnormalities, thereby improving pigging efficiency and pipeline safety and reliability, addressing the challenges faced by existing in-pipeline detection technologies, and ensuring the continued safety of oil and gas pipeline transportation.

[0041] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] Figure 1 This is a flow chart of a method for monitoring the progress of a pigging detection device according to the present invention. Figure 1 The present invention provides a method for monitoring the progress of a pigging detection device, comprising:

[0043] Step 1: Continuously collect vibration signal data in the pipeline through the Φ-OTDR sensor installed in the pipeline.

[0044] The method of the present invention is based on Φ-OTDR (Optical Time Domain Reflectometer, phase-sensitive optical time domain reflection technology) distributed optical fiber sensing technology. The optical cable is laid in the same trench as the oil and gas pipeline, and the Φ-OTDR distributed optical fiber sensor is installed in the pipeline to collect the vibration signal data in the pipeline in real time. During installation, it is necessary to ensure that the optical fiber covers the entire pipeline. Usually, a Φ-OTDR sensor is installed every 40km. The Φ-OTDR sensor is connected to an external computer via an optical cable and transmits the collected vibration signal data to the computer. The computer is connected to the Φ-OTDR sensor through a data transmission interface, receives the vibration signal data collected by the sensor, analyzes and processes the sensor data, and decides whether to trigger an alarm.

[0045] In some specific embodiments, the computer continuously collects vibration signal data output by the Φ-OTDR sensor for a duration of 7-15 days, and ensures that the sampling integrity reaches more than 90%, typically 100%.

[0046] Step 2: Use data mining technology to analyze the vibration signal data of each defense zone in the pipeline and determine the time domain trigger threshold of each defense zone.

[0047] Data acquisition software collects at least seven consecutive days of time-domain signals, or vibration signal data, and uses data mining techniques to accurately select time-domain trigger thresholds. Global data collection lasts for 7-15 days. This long-term data accumulation allows for a comprehensive sampling of actual pipeline vibrations during the operation of pigging equipment (typically pigs equipped with internal detectors), achieving sampling integrity exceeding 90%.

[0048] Then, using sufficient vibration signal data samples, we perform data analysis, statistically analyze the data variance for each defense zone, and roughly set the time domain trigger threshold. The defense zones are divided according to the resolution of the optical cable positioning and are set according to the width of the laser pulse, usually 50 meters per zone.

[0049] The variance (sample variance) in statistics is the average of the squares of the differences between each sample value and the mean of all sample values. The formula for calculating the population variance is:

[0050]

[0051] Where X is each vibration signal data sample in the defense zone; μ is the overall mean of all vibration signal data samples in the defense zone; F is the total number of cases; σ 2 is the overall variance of the vibration signal data of each zone in the pipeline.

[0052] When the data distribution is relatively dispersed (i.e., the vibration signal data fluctuates significantly around the mean), the sum of the squares of the differences between the individual data points and the mean is large, and the variance is large, indicating that vibration has occurred in the defense zone. When the data distribution is relatively concentrated, the sum of the squares of the differences between the individual data points and the mean is small, indicating that the defense zone is relatively quiet. Therefore, the larger the variance, the greater the data fluctuation; the smaller the variance, the smaller the data fluctuation.

[0053] In the actual calculation process, the number of samples for variance calculation in each defense zone is one frame (i.e., F = 512 sample data). By analyzing the variance distribution of each defense zone for 7-15 consecutive days, a Gaussian normal distribution is used to roughly estimate the time domain trigger threshold for each defense zone. Specifically, a mean is calculated using Gaussian statistics, and this mean is then used as the rough time domain trigger threshold for triggering pig positioning calculations.

[0054] Then, the time domain trigger threshold is finely set. The data of the initially set rough time domain trigger threshold is compared with the actual vibrations that occur during the movement of the pipe cleaning detection equipment. The comparison time is 2-4 weeks. During this process, the set rough time domain trigger threshold is fine-tuned according to the actual vibrations to increase sensitivity and achieve accurate tracking of the pipe cleaning device. Specifically, after 2-4 weeks of observation, the rough time domain trigger threshold is fine-tuned according to the mean of the observed samples. It is raised in places with large external interference and lowered when the external interference is small. In this way, the time domain trigger threshold of each defense zone is finely set, which can reduce the misjudgment of the pipe cleaning device and improve sensitivity.

[0055] Step 3: Based on the time domain trigger threshold of each defense zone, lock the event defense zone suspected of vibration event through time domain triggering.

[0056] The pipe cleaner will generate large vibrations during its movement. The present invention uses optical fibers to collect the vibration signals. When the vibration signal of a certain defense zone frequently exceeds the time domain trigger threshold set for that defense zone, it can be considered that there is a suspected pipe cleaner movement vibration event in that defense zone, thereby locating the defense zone as the event defense zone where the pipe cleaning detection equipment is moving. The adjacent defense zones of the event defense zone refer to the defense zones before and after the defense zone that has been determined to be suspected of having a vibration event. The subsequent correlation analysis mainly looks at whether the defense zones before and after have similar signals in the time domain. It is determined that there is a suspected vibration event. The pipe cleaner is always moving. For example, if the vibration signal of the suspected pipe cleaner is detected in defense zone No. 5, the signals before and after defense zones No. 4 and No. 6 will be analyzed to determine whether it is a real pipe cleaner movement signal.

[0057] Step 4: Perform correlation and refinement analysis on the adjacent defense zones of the event defense zone and calculate the vibration characteristics of the adjacent defense zones.

[0058] An association and refinement analysis is performed on the adjacent defense zones of the event zone, and four vibration characteristic quantities, namely the square difference, short-time Fourier transform, short-time level crossing rate and disturbance duration of the adjacent defense zones, are calculated. Combined with deep learning 3D multi-convolutional neural network technology and image recognition technology, the vibration characteristics of the pipe cleaner vibration event are converted into image features for online real-time recognition, thereby enabling the tracking of the pipe cleaner's movement.

[0059] Specifically, the following four vibration characteristic parameters are extracted from the vibration signal data output by the Φ-OTDR sensor: squared error, short-time level crossing rate, short-time Fourier transform, and disturbance duration. These characteristic parameters will be used for subsequent event classification and analysis.

[0060] ①Square difference:

[0061] Square Difference is the most commonly used statistical distribution indicator in probability statistics, reflecting the degree of dispersion between individuals within a group. In a Φ-OTDR distributed fiber sensor, the fluctuations in the backscattered Rayleigh scattering signal can effectively reflect the instantaneous intensity of the disturbance, thereby identifying sudden high-intensity disturbances (such as stress failure). Assuming the sampling length during the disturbance is P, divided equally into M time periods, the number of sampling points in each time period is m = P / M. The square difference of each sampling point in each time period is concatenated to form a square difference curve for the entire disturbance period. The square difference in each time period represents the degree of data fluctuation within that time period. Different disturbance patterns result in different degrees of fluctuation in the backscattered Rayleigh scattering curve, which serves as a basis for identification.

[0062] ②Short-time over-voltage rate:

[0063] The level crossing rate (LCR) refers to the number of times a signal crosses a certain level while fluctuating near that level. It is a basic method for extracting a signal's frequency-time relationship, reflecting the frequency of signal fluctuations around a certain level per unit time. The short-term level crossing rate varies with the short-term average frequency and short-term phase fluctuations of the sensor's output signal. Therefore, analyzing the short-term level crossing rate can provide insights into the characteristics of the disturbance signal.

[0064] The process of extracting signal frequency features using short-time level-crossing rate is as follows: Figure 3 As shown. The vibration signal is divided into S segments according to the time point. The number of data points in each time segment is N. The average level crossing rate L is calculated for each time segment, and finally the short-time average level crossing curve L(s) of the signal is obtained. The formula for calculating the average level crossing rate L in each time segment is:

[0065]

[0066] Where N is the number of vibration signal data points in each time period; I(n) is the amplitude of the nth vibration signal data point; α is the set level threshold; ψ is the indicator function, which is 1 when the condition in the brackets is met and 0 otherwise.

[0067] ③Short-time Fourier transform:

[0068] The short-time Fourier transform (STFT) is a commonly used method for signal time-frequency analysis. Its fundamental principle is to localize the integration interval of the signal's Fourier transform. It is an effective tool for preserving both the time and frequency domains of a signal. By performing a short-time Fourier transform on a vibration signal, the frequency information of the vibration signal within different disturbance time regions can be effectively analyzed, thereby distinguishing high-frequency disturbances from low-frequency disturbances.

[0069] ④Disturbance duration:

[0070] The disturbance duration is the duration of the disturbance on the sensing fiber. It is typically the time difference between the first and last threshold trigger times, reflecting the time domain characteristics of the disturbance signal. Analysis of the disturbance duration can effectively distinguish between instantaneous and long-term effects.

[0071] Step 5: Use image recognition technology to convert vibration features into image features, and combine 3D multi-convolutional neural network technology to perform deep learning on image features to determine the travel position of the pipe cleaning detection equipment.

[0072] The path tracking of various pigging detection devices can be achieved by analyzing the characteristics of vibration signals. This invention utilizes optical fibers laid in the same trench as a natural gas pipeline or oil pipeline as sensors to monitor the movement of pigs within the pipeline. The pig's movement primarily includes its position and speed; excessive speed, excessive slowness, or even pauses are all signs of abnormal pig operation. By collecting vibration signals generated by the pig and using statistics and Fourier transform techniques to study the distribution characteristics of the vibration signals in the time and frequency domains, the characteristic quantities of each vibration condition are understood, providing a theoretical basis for pig path tracking.

[0073] Based on the rules and results of the above-mentioned technical analysis, the present invention combines the 3D multiple convolutional neural network technology and image recognition technology of deep learning to convert the vibration characteristics of the pipe cleaner vibration event into image features for online real-time recognition, thereby being able to track the path of the pipe cleaner. Specifically, the four vibration characteristic values of each defense zone are normalized to characterize the vibration intensity of each defense zone; points of different colors are used to represent the vibration intensity of each defense zone, and then the vibration intensity of each defense zone on the entire optical cable is plotted into a waterfall chart based on time and space. The waterfall chart can display the vibration intensity of each defense zone over a period of time. The vibration intensity of different color depths on the waterfall chart is used as image features, and deep learning is performed using a 3D multiple convolutional neural network. In this way, the actual vibration of the pipe cleaner on the entire optical cable can be identified, and the defense zone to which the pipe cleaner has traveled can be located, that is, the travel position of the pipe cleaning detection equipment can be determined.

[0074] Step 6: Calculate the travel speed of the pigging detection equipment based on the travel position, and issue an abnormal warning when the travel speed is abnormal.

[0075] Through the GIS geographic system, the geographical characteristics of dynamic vibration events are grasped. Then, the complete vibration signal of the pig's movement is collected to study the dynamic change patterns of the pig's movement vibration signal in the time domain and geographic space. The collected vibration signal data is used to calculate the pig's travel speed, thereby judging the safety of the current pig's movement. Specifically, the pig's travel position at each moment is first calculated through previous images and deep learning. Then, the travel speed is calculated based on the relationship between the travel position and time. If the pig's travel speed is too fast or too slow, it is judged as abnormal, requiring an abnormal warning. The staff is notified to go to the site to check the cause of the pig's abnormality, and then measures are taken to ensure normal movement.

[0076] Furthermore, by calculating the pig's position and speed, it's possible to determine whether the pipeline is flat or tilted, uphill or downhill. Combined with geographic information, this can also help determine whether the pipeline has undergone significant deformation. For example, if the original speed on flat ground should be 1, but the measured speed is 2, then this section of the pipeline is likely tilted, with a downward slope from near to far, causing the speed to increase.

[0077] Furthermore, intelligent algorithms such as genetic algorithms and neural networks can be combined with pattern recognition to distinguish normal background vibration from pig movement, improving alarm effectiveness and reducing false alarm rates. Further refinement can be achieved by defining an indicator system for pig movement status within the pipeline. Using genetic fuzzy neural networks, a control model for the pipeline pigging and monitoring device based on pig speed and other gas transmission parameters can be established. System software developed using programming languages and platforms integrates pig movement status monitoring and control, as well as movement anomaly warnings, into a unified system platform based on the movement status control model. The platform provides an operational interface for monitoring the pig's movement status and provides early warning information, data viewing, and alarm functions.

[0078] Further development is possible, based on the oil and gas pipeline pigging and progress status control system platform, to develop a mobile prototype that implements voice and data viewing, SMS alarms, and other functions. Through the alarm platform and mobile prototype, early warning information can be pushed to the responsible person's mobile phone, allowing timely handling of pigging anomalies.

[0079] Compared with existing pipeline detection technologies, the method of the present invention has at least the following advantages:

[0080] (1) Improved data collection efficiency. By using the Φ-OTDR sensor to continuously collect vibration signal data for 7-15 days and ensuring that the sampling integrity reaches more than 90%, more data can be obtained for subsequent analysis.

[0081] (2) More accurate identification of pig path vibration events. Using data mining technology and variance distribution analysis, the time domain trigger threshold of each defense zone can be automatically determined, making the system more adaptable and accurate.

[0082] (3) More accurate description of vibration events and improved detection accuracy. Multiple vibration characteristic parameters, such as square error, short-time level-crossing rate, short-time Fourier transform, and disturbance duration, are extracted from the vibration signal data output by the Φ-OTDR sensor. These vibration characteristic parameters can more accurately describe vibration events, thereby improving the detection accuracy of pig path tracking.

[0083] (4) Improved understanding and handling of potential problems. By calculating the vibration characteristic parameters of adjacent defense zones and converting them into image feature matrices, the dynamic relationship of events in space can be captured, which helps to identify the correlation between multiple defense zones.

[0084] (5) Improved accurate monitoring of pipe cleaner behavior and anomaly identification. Using a 3D multi-convolutional neural network (CNN) algorithm and image recognition technology to analyze the image feature matrix enables accurate tracking of pipe cleaner behavior and identification of abnormal events.

[0085] (6) Improved overall system efficiency and made operation more convenient. A natural gas pipeline pigging and detection status control system platform was developed, integrating multiple functions into a unified system, including pig movement status monitoring and abnormality warning functions.

[0086] Based on the method provided by the present invention, the present invention also provides a travel monitoring system for a pigging detection device, comprising:

[0087] A data acquisition module is used to continuously collect vibration signal data in the pipeline through a Φ-OTDR sensor installed in the pipeline;

[0088] A data mining module is used to analyze the vibration signal data of each defense zone in the pipeline using data mining technology to determine the time domain trigger threshold of each defense zone;

[0089] The time domain trigger module is used to lock the event zone suspected of vibration events through time domain triggering based on the time domain trigger threshold of each zone;

[0090] Adjacent defense zone association and vibration feature extraction module, used to perform association and refinement analysis on the adjacent defense zones of the event defense zone and calculate the vibration features of the adjacent defense zones;

[0091] The path tracking module uses image recognition technology to convert vibration features into image features, and combines 3D multi-convolutional neural network technology to perform deep learning on image features to determine the travel position of the pigging inspection equipment;

[0092] The travel status monitoring and early warning module is used to calculate the travel speed of the pipe cleaning detection equipment according to the travel position, and issue an abnormal early warning when the travel speed is abnormal.

[0093] Furthermore, the present invention provides an electronic device comprising: a processor, a communication interface, a memory, and a communication bus. The processor, the communication interface, and the memory communicate with each other via the communication bus. The processor can invoke a computer program stored in the memory to execute a method for monitoring the progress of a pigging detection device.

[0094] The Φ-OTDR sensor is connected to an electronic device (typically a computer) via optical fiber, transmitting the vibration signal data to the electronic device for analysis and processing. The electronic device internally includes a data acquisition module, a data mining module, a time-domain triggering module, a module for correlating adjacent zones and extracting vibration signatures, a path tracking module, and a travel status monitoring and early warning module. These modules are interconnected through an internal data processing unit and work together to perform data analysis and pig path tracking. The status control system platform serves as an external interface to the electronic device, receiving processed data and providing a user interface and alarm functions.

[0095] In addition, when the computer program in the above-mentioned memory is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-transitory computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk.

[0096] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0097] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for monitoring the progress of a pigging detection device, characterized in that: include: The Φ-OTDR sensor installed in the pipeline continuously collects vibration signal data in the pipeline; Use data mining technology to analyze the vibration signal data of each defense zone in the pipeline and determine the time domain trigger threshold of each defense zone; The method of analyzing the vibration signal data of each defense zone in the pipeline using data mining technology to determine the time domain trigger threshold of each defense zone specifically includes: The variance of the vibration signal data for each zone in the pipeline Perform statistical analysis. By analyzing the variance distribution of each defense zone for 7-15 consecutive days, a rough time-domain trigger threshold for each defense zone is set using Gaussian normal distribution. Specifically, a mean is calculated using Gaussian statistics, and this mean is then used as the rough time-domain trigger threshold for triggering the positioning calculation of the pigging detection equipment. Where X is the single vibration signal data of the defense zone; μ is the overall mean of all vibration signal data in the defense zone; and F is the total number of cases. Fine-tune the coarse time-domain trigger threshold based on the actual vibrations occurring during pigging inspection. This allows for precise definition of the time-domain trigger threshold for each zone. Specifically, after 2-4 weeks of observation, fine-tune the coarse time-domain trigger threshold based on the mean of the observed samples. Adjust the threshold higher in areas with significant external interference and lower in areas with minimal external interference, thereby fine-tuning the time-domain trigger threshold for each zone. Based on the time domain trigger threshold of each defense zone, the event defense zone suspected of vibration events is locked through time domain triggering; Conduct correlation and refinement analysis on adjacent defense zones of the event defense zone and calculate the vibration characteristics of adjacent defense zones; Image recognition technology is used to convert vibration features into image features. 3D multi-convolutional neural network technology is then used to perform deep learning on these features to determine the travel position of the pigging inspection equipment. The travel speed of the pigging detection equipment is calculated based on the travel position, and an abnormal warning is issued when the travel speed is abnormal.

2. The method for monitoring the progress of a pigging detection device according to claim 1, characterized in that: The method of continuously collecting vibration signal data in the pipeline by using a Φ-OTDR sensor installed in the pipeline specifically includes: Multiple Φ-OTDR sensors installed in the pipeline are connected to the computer via optical cables laid in the same trench as the pipeline; The Φ-OTDR sensor continuously collects vibration signal data in the pipeline for 7-15 days.

3. The method for monitoring the progress of a pigging detection device according to claim 2, characterized in that: The performing of correlation and refinement analysis on adjacent defense zones of the event defense zone and calculating vibration characteristics of the adjacent defense zones specifically includes: The adjacent defense zones of the event defense zone are subjected to correlation and refinement analysis, and four vibration characteristics of the adjacent defense zones, namely, square difference, short-time Fourier transform, short-time level crossing rate and disturbance duration, are calculated. The calculation formula of the average level crossing rate L in each time period is: N is the number of vibration signal data points in each time period; I(n) is the amplitude of the nth vibration signal data point; α is the set level threshold; ψ is the indicator function, which is 1 when the condition in the brackets is met and 0 otherwise.

4. The method for monitoring the progress of a pigging detection device according to claim 3, characterized in that: The method of converting vibration features into image features using image recognition technology and combining it with 3D multi-convolutional neural network technology to perform deep learning on the image features to determine the travel position of the pigging detection equipment specifically includes: The four vibration eigenvalues of each defense zone are normalized to represent the vibration intensity of each defense zone; Use points of different colors to represent the vibration intensity of each defense zone, and then plot the vibration intensity of each defense zone on the entire optical cable into a waterfall chart based on time and space; The vibration intensity of different color depths on the waterfall chart is used as image features, and deep learning is performed using a 3D multi-convolutional neural network to determine the travel position of the pipe cleaning inspection equipment.

5. A traveling monitoring system for a pigging detection device, characterized in that: include: A data acquisition module is used to continuously collect vibration signal data in the pipeline through a Φ-OTDR sensor installed in the pipeline; A data mining module is used to analyze the vibration signal data of each defense zone in the pipeline using data mining technology to determine the time domain trigger threshold of each defense zone; The method of analyzing the vibration signal data of each defense zone in the pipeline using data mining technology to determine the time domain trigger threshold of each defense zone specifically includes: The variance of the vibration signal data for each zone in the pipeline Perform statistical analysis. By analyzing the variance distribution of each defense zone for 7-15 consecutive days, a rough time-domain trigger threshold for each defense zone is set using Gaussian normal distribution. Specifically, a mean is calculated using Gaussian statistics, and this mean is then used as the rough time-domain trigger threshold for triggering the positioning calculation of the pigging detection equipment. Where X is the single vibration signal data of the defense zone; μ is the overall mean of all vibration signal data in the defense zone; and F is the total number of cases. Fine-tune the coarse time-domain trigger threshold based on the actual vibrations occurring during pigging inspection. This allows for precise definition of the time-domain trigger threshold for each zone. Specifically, after 2-4 weeks of observation, fine-tune the coarse time-domain trigger threshold based on the mean of the observed samples. Adjust the threshold higher in areas with significant external interference and lower in areas with minimal external interference, thereby fine-tuning the time-domain trigger threshold for each zone. The time domain trigger module is used to lock the event zone suspected of vibration events through time domain triggering based on the time domain trigger threshold of each zone; Adjacent defense zone association and vibration feature extraction module, used to perform association and refinement analysis on the adjacent defense zones of the event defense zone and calculate the vibration features of the adjacent defense zones; The path tracking module uses image recognition technology to convert vibration features into image features, and combines 3D multi-convolutional neural network technology to perform deep learning on image features to determine the travel position of the pigging inspection equipment; The travel status monitoring and early warning module is used to calculate the travel speed of the pipe cleaning detection equipment according to the travel position, and issue an abnormal early warning when the travel speed is abnormal.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for monitoring the progress of the pigging detection equipment according to any one of claims 1 to 4 is implemented.

7. The electronic device according to claim 6, wherein: The memory is a non-transitory computer-readable storage medium.

Images