Pipeline defect weak magnetic signal visual construction method and system

Abstract

The invention relates to the technical field of pipeline defect detection, in particular to a pipeline defect weak magnetic signal visualization construction method and system. A pipeline signal acquisition system is constructed according to structural characteristics and functional characteristics of a pipeline, the pipeline is provided with a coating layer, and the pipeline signal acquisition system is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; weak magnetic signals on the surface or in the pipeline are obtained through combined use of multiple magnetic signal detection devices; signal acquisition sensors are arranged according to the characteristics of the S-shaped elbow, and the position layout of the sensors is adjusted through a genetic algorithm to improve the pipeline defect detection rate; the method comprises the following steps: matching magnetic signal data acquired by different equipment in a signal acquisition system according to position and time labels, performing data integration by utilizing different equipment characteristics to construct a complete visual model of a pipeline, and marking suspicious defect points; potential defects are found in time through the visual model, and a coping scheme is made.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline defect detection, and in particular to a method and system for visualizing weak magnetic signals of pipeline defects. Background Art

[0002] Pipeline weak magnetic signal detection technology, originating from magnetic flux leakage detection methods in the 1960s, has evolved from a single measurement method to a multi-dimensional, high-precision, and intelligent one. Early technologies primarily relied on strong magnetic field excitation and simple magnetic flux leakage signal detection, suitable for bare, uncoated pipelines. With the widespread use of coated pipelines, traditional technologies have faced limitations such as severe signal attenuation and high noise interference. Advances in sensing and digital signal processing technologies have led to the rise of low-field magnetic detection and weak magnetic signal enhancement, making it possible to detect weak magnetic signals through thick coatings. Weak magnetic signal detection on coated pipelines faces challenges such as weak signal penetration and high attenuation. Furthermore, interfaces between different material layers generate additional magnetic signals, which can easily blend with defect signals. Magnetic field distortion caused by geometric changes such as pipe bends and diameter changes can also interfere with defect signals. In this context, a method is needed to more accurately detect pipeline defects while maximizing detection coverage within limited inspection resources. Summary of the Invention

[0003] (1) Technical problems to be solved The purpose of the present invention is to provide a method and system for visualizing weak magnetic signals of pipeline defects, so as to solve the problem of the influence of additional magnetic interference generated by pipelines with coating layers on defect signal detection, as well as the interference of magnetic field distortion caused by geometric changes such as pipe bending and diameter change on defect signal detection.

[0004] (2) Technical solution To achieve the above objectives, the present invention provides a method for visualizing weak magnetic signals of pipeline defects, which detects weak magnetic signals of pipeline defects in S-shaped bend pipes with coatings. The method comprises: S1. Construct a pipeline signal acquisition system based on the structural characteristics and functional features of the pipeline. The pipeline has a coating layer and is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector, and a fixed monitoring station. A weak magnetic signal on the surface or inside of the pipeline is obtained by combining multiple magnetic signal detection devices.

[0005] S2. The pipeline is an S-shaped bend at some point, which includes two 45° bends and has partial diameter extrusion deformation in the middle section. Signal acquisition sensors are arranged according to the characteristics of the S-shaped bend, and the sensor position layout is adjusted using a genetic algorithm to improve the pipeline defect detection rate.

[0006] S3. Match the magnetic signal data collected by different devices in the signal acquisition system according to location and time tags, use the characteristics of different devices to integrate the data to build a complete visual model of the pipeline, and mark suspected defect points; use the visual model to promptly discover potential defects and develop response plans.

[0007] Furthermore, the method of adjusting the sensor position layout by using a genetic algorithm includes: Obtain the preset coverage radius of each magnetic signal sensor, set the sensor's location conditions and layout area conditions; obtain the number of sensors that can be arranged in the area based on the preset coverage radius and layout area conditions of the magnetic signal sensor , generating an initialized population of magnetic signal sensor layout schemes.

[0008] By calculating the fitness value of each individual in the initialized population, selecting individuals with higher fitness for reproduction, combining the sensor positions of two individuals using multi-point crossover, adjusting the positions of individual sensors, or increasing or decreasing the number of sensors, a new population is generated; iteratively repeating the evaluation, selection, crossover, and mutation until the fitness change is less than a preset threshold, a new magnetic field sensor layout is obtained.

[0009] The sensor location condition is the limiting condition that each sensor location must meet, and the layout area condition is the limiting condition for the area where multiple sensors are installed together; the initialization population is composed of Individual Each individual is a magnetic signal sensor layout scheme, including the combination of sensor positions and sensor quantities in the area, expressed as: ,in Indicates the location of the sensor. Indicates the number of sensors.

[0010] According to the preset coverage radius of the magnetic signal sensor, the magnetic signal sensors that exceed the area limitation conditions in the individual are eliminated, and the magnetic signal sensors that overlap the coverage range in the individual are eliminated or rearranged to obtain a new magnetic signal sensor initialization population Individual .

[0011] Furthermore, the method for calculating the fitness value of each individual in the initialized population includes: An initialization population of magnetic signal sensors is generated according to the limited conditions of the magnetic signal sensor layout, and an iterative fitness function of the magnetic signal sensor population is designed to perform fitness evaluation. The fitness function is: ; in, is the distance from spatial point p to the position of the i-th sensor A is the distance where multiple sensors are installed together.

[0012] The fitness of each individual in the initialized population is calculated separately. The higher the sensor coverage rate among the individuals, the smaller the average value of the magnetic difference in the area and the fewer the number of sensors in the area, the higher the fitness of the individuals; individuals with higher fitness are selected for reproduction to form a new magnetic signal sensor population.

[0013] Furthermore, the method of repeatedly evaluating, selecting, crossing, and mutating until the fitness change is less than a preset threshold to obtain a new magnetic field sensor layout includes: The magnetic signal sensor positions of two individuals are combined by single-point crossover or multi-point crossover to generate a new magnetic signal sensor layout; the positions of individual sensors are then randomly adjusted to increase or decrease the number of sensors, increase the diversity of the population, and generate a new population. Individual .

[0014] Calculate the fitness of each individual in the new population, select excellent individuals based on the fitness, and then perform single-point crossover or multi-point crossover to combine the magnetic signal sensors of two individuals, adjust the position of individual sensors, increase or decrease the number of sensors, generate a new generation of individuals, and obtain a new generation of population. Individual .

[0015] Repeat the evaluation, selection, crossover and mutation to obtain a new population until the fitness change is less than the preset threshold. Then stop the iteration and obtain the optimal population under the preset conditions, which is the optimal layout plan of the sensors under the preset conditions in the area.

[0016] Furthermore, the method of obtaining weak magnetic signals on the surface or inside of a pipeline by combining multiple magnetic signal detection devices includes: The pipeline is coated, and the pipeline signal acquisition system is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; the inner pipeline magnetic data is scanned and detected from the inside of the pipeline by the inner pipeline intelligent crawling detection device; the outer pipeline detector is used to detect defects in the metal pipeline layer through the coating layer; the fixed monitoring station fixes the equipment at key nodes for long-term detection; the key nodes are pipeline interfaces, valves, bending points and extrusion deformation points; by arranging detection equipment inside, outside and at key points of the pipeline with a coating, a multi-device collaborative detection solution is formed.

[0017] Furthermore, the multi-device collaboration method includes: According to the characteristics of the pipeline shape, dynamic inspection trajectories are designed for the inner pipeline intelligent crawling detection device and the outer pipeline detector; the moving speed of the inner pipeline intelligent crawling detection device is reduced by 30% when passing through the S-shaped area of ​​the pipeline, and mobile detection is performed by adopting a spiral merging and turning movement mode. The number of sensor activations at the bends is increased to increase the number of magnetic signal detections in the S-shaped area of ​​the pipeline, while also increasing the sampling frequency.

[0018] The detection path of the external pipeline detector is designed using the pipeline structure, a hovering scanning point is designed for the extrusion deformation of the S-shaped bend, a high-power pulse magnetic field generator is set at the front end of the bend, a frequency-adjustable alternating magnetic field source is used in the pipe diameter change section, and a magnetic shielding device is set near the key support points to reduce interference.

[0019] Furthermore, the method of timely discovering potential defects through the visualization model and formulating response plans includes: By establishing a multi-device collaborative detection scheme to form a master-slave signal calibration mechanism, mutual interference is reduced according to the time-division multiplexing sampling scheme; the sensor array sensitivity and sampling frequency are dynamically adjusted according to the initial detection results, high-risk areas are automatically identified based on the detection data, additional detection resources are dynamically allocated, and detection equipment is dynamically adjusted in real time.

[0020] Based on the same inventive concept, the present invention also provides a pipeline defect weak magnetic signal visualization construction system, the system comprising: a first data acquisition module, a second layout optimization module, and a data integration module, each module being connected in sequence; The first data acquisition module is used to construct a pipeline signal acquisition system based on the structural characteristics and functional features of the pipeline. The pipeline has a coating layer. The pipeline signal acquisition system consists of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; weak magnetic signals on the surface or inside of the pipeline are obtained by combining multiple magnetic signal detection devices.

[0021] The second layout optimization module is used to layout signal acquisition sensors according to the characteristics of the S-shaped bend pipe, and adjust the sensor position layout through a genetic algorithm to improve the pipeline defect detection rate; a certain part of the pipeline is an S-shaped bend pipe, and the S-shaped bend pipe contains two 45° bends, and there is partial pipe diameter extrusion deformation in the middle section.

[0022] The data integration module is used to match the magnetic signal data collected by different devices in the signal acquisition system according to location and time tags, use the characteristics of different devices to integrate data to build a complete visual model of the pipeline, and mark suspected defect points; potential defects are discovered in a timely manner through the visual model, and response plans are formulated.

[0023] (3) Beneficial effects Compared with the existing technology, the beneficial effects of the present invention are as follows: a multi-device collaborative detection scheme is adopted for defect detection of pipes with coating layers, and a signal acquisition system is composed of an intelligent crawling detection device for the inner pipe, an outer pipe detector and a fixed monitoring station, thereby reducing the weakening of the coating layer on the signal and the interference of the multi-layer material on the signal; the detection coverage rate is improved by optimizing the layout of the magnetic signal sensor; and the interference of magnetic field distortion caused by geometric shape changes such as pipe bending and diameter change on defect signal detection is improved by designing a dynamic inspection trajectory, thereby improving the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a flowchart of a method for visualizing weak magnetic signals of pipeline defects according to Example 1 of the present invention; Figure 2 This is a schematic diagram of the module composition of a pipeline defect weak magnetic signal visualization system according to Example 2 of the present invention. DETAILED DESCRIPTION

[0025] 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.

[0026] Before giving examples, it is necessary to explain the application scenarios of the present invention. The present invention is a method and system for visualizing weak magnetic signals of pipeline defects, which are used to detect weak magnetic signal defects of pipelines.

[0027] Example 1: Figure 1 As shown, this embodiment provides a method for visualizing weak magnetic signals of pipeline defects, the method comprising: S1. Construct a pipeline signal acquisition system based on the structural characteristics and functional features of the pipeline. The pipeline has a coating layer and is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector, and a fixed monitoring station. A weak magnetic signal on the surface or inside of the pipeline is obtained by combining multiple magnetic signal detection devices.

[0028] The pipeline is coated, and the pipeline signal acquisition system is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; the inner pipeline magnetic data is scanned and detected from the inside of the pipeline by the inner pipeline intelligent crawling detection device; the outer pipeline detector is used to detect defects in the metal pipeline layer through the coating layer; the fixed monitoring station fixes the equipment at key nodes for long-term detection; the key nodes are pipeline interfaces, valves, bending points and extrusion deformation points; by arranging detection equipment inside, outside and at key points of the pipeline with a coating, a multi-device collaborative detection solution is formed.

[0029] It's important to note that the intelligent crawling detection device inside the pipeline scans and detects magnetic data from the inner layer of the pipeline; the external pipeline detector detects defects in the metal pipeline through the cladding layer; and the fixed monitoring station conducts long-term monitoring at key nodes (pipeline joints, valves, bends, and extrusion deformation points). This integrated internal and external, dynamic and static detection system effectively overcomes the signal attenuation effects of the cladding layer and the magnetic field distortion caused by complex geometric shapes, significantly improving the comprehensiveness and accuracy of detection.

[0030] For example, a 120-kilometer-long LNG pipeline with a diameter of 1.2 meters features a multi-layer composite structure (an inner layer of special alloy steel pipe, a middle thermal insulation layer, an outer polyethylene anti-corrosion layer, and an outermost concrete weighting layer). During on-site inspections, a 300-meter-long S-shaped section (including two 45° bends and a 15% pipe diameter extrusion deformation zone) was found to be extremely challenging to inspect, as signal noise significantly interfered with defect identification. Based on the pipeline's structural characteristics, a three-component signal acquisition system was constructed: an internal pipeline intelligent crawler (PIG-S600), an external underwater robot (ROV-M320), and three fixed monitoring stations (FMS-850).

[0031] S2. The pipeline is an S-shaped bend at some point, which includes two 45° bends and has partial diameter extrusion deformation in the middle section. Signal acquisition sensors are arranged according to the characteristics of the S-shaped bend, and the sensor position layout is adjusted using a genetic algorithm to improve the pipeline defect detection rate.

[0032] Obtain the preset coverage radius of each magnetic signal sensor, set the sensor's location conditions and layout area conditions; obtain the number of sensors that can be arranged in the area based on the preset coverage radius and layout area conditions of the magnetic signal sensor , generating an initialized population of magnetic signal sensor layout schemes.

[0033] By calculating the fitness value of each individual in the initialized population, selecting individuals with higher fitness for reproduction, combining the sensor positions of two individuals using multi-point crossover, adjusting the positions of individual sensors, or increasing or decreasing the number of sensors, a new population is generated; iteratively repeating the evaluation, selection, crossover, and mutation until the fitness change is less than a preset threshold, a new magnetic field sensor layout is obtained.

[0034] The sensor location condition is the limiting condition that each sensor location must meet, and the layout area condition is the limiting condition for the area where multiple sensors are installed together; the initialization population is composed of Individual Each individual is a magnetic signal sensor layout scheme, including the combination of sensor positions and sensor quantities in the area, expressed as: ,in Indicates the location of the sensor. Indicates the number of sensors.

[0035] For example, experimental data showed that the initial detection rate in the S-shaped section was only 71%, far lower than the 85% in other sections. Therefore, the team decided to optimize the sensor layout to improve detection accuracy. They first determined the effective parameters of various sensors in this pipeline environment: the coverage radius of fluxgate sensors was measured to be 25 cm (±1.2 cm), that of Hall effect sensors to be 18 cm (±0.8 cm), and that of giant magnetoresistive sensors to be 30 cm (±1.5 cm). Based on the spatial constraints of the S-shaped section, installation requirements were set: a minimum distance of 5 cm between the sensor and the pipe wall, and a minimum spacing of 10 cm between adjacent sensors. Coverage was required to be ≥95% for every 45-degree arc in the bend, and ≥98% in the squeeze zone. The average of three independent calculations determined that a theoretical maximum of 108 sensors could be deployed in this area. The experimental team generated an initial population of 50 different layouts, each containing 76 to 108 sensors at different locations. Inspection using a laser measurement system revealed that 12 of these layouts had sensor locations outside the installation requirements, and another 8 had significant overlap (>60% coverage overlap). After adjustments, 48 ​​valid layouts were obtained.

[0036] According to the preset coverage radius of the magnetic signal sensor, the magnetic signal sensors that exceed the area limitation conditions in the individual are eliminated, and the magnetic signal sensors that overlap the coverage range in the individual are eliminated or rearranged to obtain a new magnetic signal sensor initialization population Individual .

[0037] An initialization population of magnetic signal sensors is generated according to the limited conditions of the magnetic signal sensor layout, and an iterative fitness function of the magnetic signal sensor population is designed to perform fitness evaluation. The fitness function is: ; in, is the distance from spatial point p to the position of the i-th sensor A is the distance where multiple sensors are installed together.

[0038] It should be noted that the fitness function mathematical model ensures the scientific nature of individual fitness evaluation, giving priority to individuals with high sensor coverage, small average magnetic difference in the region, and a small number of sensors, thereby optimizing resource allocation while ensuring detection quality.

[0039] The fitness of each individual in the initialized population is calculated separately. The higher the sensor coverage rate among the individuals, the smaller the average value of the magnetic difference in the area and the fewer the number of sensors in the area, the higher the fitness of the individuals; individuals with higher fitness are selected for reproduction to form a new magnetic signal sensor population.

[0040] For example, the detection system divided the S-shaped segment into 3,600 evenly distributed grid points, and tested the coverage of each layout scheme on an actual pipeline model. Records show that layout scheme 32 (78 sensors) achieved a measured coverage rate of 96.2%, a monitoring intensity of 0.19 at the weakest point (located on the outer edge of the second bend), and a standard deviation of magnetic field variation within the area of ​​0.18 millitesla. In contrast, layout scheme 15 (82 sensors) achieved a coverage rate of 94.8%, a monitoring intensity of 0.16 at the weakest point, and a standard deviation of magnetic field variation of 0.23 millitesla. Using the fitness calculation formula, scheme 32 achieved a fitness value of 0.78, significantly better than scheme 15's 0.65. The top 20 fitness schemes (ranging from 0.64 to 0.78) were selected for iterative optimization.

[0041] The magnetic signal sensor positions of two individuals are combined by single-point crossover or multi-point crossover to generate a new magnetic signal sensor layout; the positions of individual sensors are then randomly adjusted to increase or decrease the number of sensors, increase the diversity of the population, and generate a new population. Individual .

[0042] Calculate the fitness of each individual in the new population, select excellent individuals based on the fitness, and then perform single-point crossover or multi-point crossover to combine the magnetic signal sensors of two individuals, adjust the position of individual sensors, increase or decrease the number of sensors, generate a new generation of individuals, and obtain a new generation of population. Individual .

[0043] Repeat the evaluation, selection, crossover and mutation to obtain a new population until the fitness change is less than the preset threshold. Then stop the iteration and obtain the optimal population under the preset conditions, which is the optimal layout plan of the sensors under the preset conditions in the area.

[0044] For example, a multi-point crossover operation was performed between solution 32 and solution 8 (fitness 0.72). The 38 sensors in solution 32's compression zone (coordinate range x:120-185m, y:15-42m) were retained, while the 34 sensors in solution 8's first bend (coordinate range x:20-95m, y:10-38m) were adopted to form a new composite solution. Subsequently, fine-tuning was performed: the three sensors at coordinates (132,28), (156,32), and (175,26) were moved toward the high-stress zone by 12 cm, 8 cm, and 15 cm, respectively. Two giant magnetoresistive sensors were added at coordinates (215,22) and (232,25) on the inside of the second bend. Experimental records show that the first iteration generated a total of 42 new layout solutions. The optimal solution (numbered N1-17) achieved a fitness value of 0.84, a coverage rate of 97.1%, and utilized 80 sensors. The system ran for eight consecutive iterations (a total of 316 hours of computing time). The difference in fitness between the seventh and eighth iterations was measured to be 0.009, falling below the preset threshold of 0.01, and the iterations were automatically terminated. The final optimized solution (numbered N8-03) achieved a fitness of 0.94 and a measured coverage of 99.3%, using 74 sensors. Based on the optimized layout, three sets of collaborative testing equipment were deployed: a PIG-S600 with 128 fluxgate sensors (2.94 cm spacing) installed around its perimeter, sampling at 250 Hz; an ROV-M320 equipped with six superconducting quantum interference devices (10^-12 Tesla sensitivity) and eight pulsed eddy current probes; and three FMS-850 fixed monitoring stations, each equipped with 16-24 GMR sensors in an optimized layout, were deployed at two bends (coordinates (75, 27) and (220, 24)) and the center of the squeeze zone (coordinates (158, 30)).

[0045] S3. Match the magnetic signal data collected by different devices in the signal acquisition system according to location and time tags, use the characteristics of different devices to integrate the data to build a complete visual model of the pipeline, and mark suspected defect points; use the visual model to promptly discover potential defects and develop response plans.

[0046] According to the characteristics of the pipeline shape, dynamic inspection trajectories are designed for the inner pipeline intelligent crawling detection device and the outer pipeline detector; the moving speed of the inner pipeline intelligent crawling detection device is reduced by 30% when passing through the S-shaped area of ​​the pipeline, and mobile detection is performed by adopting a spiral merging and turning movement mode. The number of sensor activations at the bends is increased to increase the number of magnetic signal detections in the S-shaped area of ​​the pipeline, while also increasing the sampling frequency.

[0047] The detection path of the external pipeline detector is designed using the pipeline structure, a hovering scanning point is designed for the extrusion deformation of the S-shaped bend, a high-power pulse magnetic field generator is set at the front end of the bend, a frequency-adjustable alternating magnetic field source is used in the pipe diameter change section, and a magnetic shielding device is set near the key support points to reduce interference.

[0048] By establishing a multi-device collaborative detection scheme to form a master-slave signal calibration mechanism, mutual interference is reduced according to the time-division multiplexing sampling scheme; the sensor array sensitivity and sampling frequency are dynamically adjusted according to the initial detection results, high-risk areas are automatically identified based on the detection data, additional detection resources are dynamically allocated, and detection equipment is dynamically adjusted in real time.

[0049] It should be noted that the adaptive adjustment mechanism enables the system to optimize resource allocation according to the actual detection situation, improve detection efficiency and accuracy, realize closed-loop optimization of the detection process, and ensure the maximum detection of potential defects under complex pipeline structure conditions.

[0050] For example, during on-site implementation, a dedicated dynamic inspection trajectory was designed for the equipment. Speedometer recordings showed that the PIG-S600 automatically reduced its speed from 1.2 m / s to a precise 0.84 m / s (with an accuracy of ±0.03 m / s) upon entering the S-shaped area. Travel recorders revealed that it employed different movement patterns in different areas: At the first bend (0-95 meters), it rotated 15 degrees for every 5 centimeters of forward movement, forming a spiral pattern and increasing the number of sampling points per unit area by 2.8 times. The compression deformation zone (95-190 meters) performed three round-trip scans, each shifting the sensor array angle by approximately 12 degrees. At the second bend (190-300 meters), all 128 sensors were simultaneously activated, increasing the sampling frequency to 400 Hz. For the ROV-M320, a pulsed magnetic field generator with a measured power of 487 watts was installed at the first bend entrance (coordinates (5, 25)). Five hovering points were arranged at the extrusion deformation site (with dwell times of 92, 88, 94, 90, and 91 seconds, respectively). An alternating magnetic field source with a measured frequency range of 48-203 Hz was used in the pipe diameter change section. Magnetic shielding devices installed near the support points were tested to reduce environmental interference by 56.8%. Data from the first complete inspection showed that the optimized signal-to-noise ratio increased from 2.7 to 3.92, and the detection rate of small defects (<2 mm) increased from 56% to 83.2%. During system operation, a real-time dynamic optimization mechanism was established. Data transmission records show that the PIG-S600, acting as the master device, generates a standard magnetic field distribution map after each inspection. The ROV-M320 and FMS-850, acting as slave devices, use this map to calibrate their sensors, reducing the calibration error from the original ±12% to ±0.6%. Time records show that signal acquisition from the three devices was precisely scheduled at intervals of 7.5 minutes, 12.3 minutes, and 8.6 minutes, eliminating 98.7% of mutual interference. System logs record that during the second inspection (April 18, 2023), a magnetic signal anomaly was detected at coordinates (62, 31) on the inside of the first bend (deviation value of 0.38 millitesla, exceeding the threshold of 0.25 millitesla). The system automatically increased the sensitivity of all sensors within a 30-centimeter radius of the area by 25.3%, increasing the sampling frequency from 400 Hz to 652 Hz, and triggering the emergency detection procedure at 2:35 PM. The ROV completed a detailed scan in 2 hours and 12 minutes, confirming a developing microcrack (precisely measured at 1.83 mm in length and 0.64 mm in depth). Analysis of cumulative data from six months of operation revealed that the system automatically identified stress concentration points in the extrusion zone (coordinate range (145-165, 25-35)) as high-risk areas. Inspection frequency in this area was increased from monthly to weekly, and two additional portable giant magnetoresistive sensors were deployed for specialized monitoring. Data from a full year of operation showed that the system discovered 27 early-stage defects (less than 3 mm in diameter), 12 more than traditional solutions, and preemptively addressed three potentially serious defects, effectively ensuring safe pipeline operation.

[0051] Example 2: Based on the same inventive concept, Figure 2 As shown, this embodiment also provides a pipeline defect weak magnetic signal visualization construction system, including: a first data acquisition module, a second layout optimization module, and a data integration module, and each module is connected in sequence; The first data acquisition module is used to construct a pipeline signal acquisition system based on the structural characteristics and functional features of the pipeline. The pipeline has a coating layer. The pipeline signal acquisition system consists of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; weak magnetic signals on the surface or inside of the pipeline are obtained by combining multiple magnetic signal detection devices.

[0052] The second layout optimization module is used to layout signal acquisition sensors according to the characteristics of the S-shaped bend pipe, and adjust the sensor position layout through a genetic algorithm to improve the pipeline defect detection rate; a certain part of the pipeline is an S-shaped bend pipe, and the S-shaped bend pipe contains two 45° bends, and there is partial pipe diameter extrusion deformation in the middle section.

[0053] The data integration module is used to match the magnetic signal data collected by different devices in the signal acquisition system according to location and time tags, use the characteristics of different devices to integrate data to build a complete visual model of the pipeline, and mark suspected defect points; potential defects are discovered in a timely manner through the visual model, and response plans are formulated.

[0054] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated on here.

[0055] Finally, it should be noted that although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for visualizing weak magnetic signals of pipeline defects, which is used to detect weak magnetic signals of pipeline defects in S-shaped bend pipes with coatings, characterized in that: The method comprises: A pipeline signal acquisition system is constructed based on the structural and functional characteristics of the pipeline. The pipeline is coated and consists of an inner pipeline intelligent crawling detection device, an outer pipeline detector, and a fixed monitoring station. A combination of multiple magnetic signal detection devices is used to obtain weak magnetic signals on the surface or inside of the pipeline. The pipeline has an S-shaped bend at a certain point, which includes two 45-degree bends and has a partial diameter extrusion deformation in the middle section. Signal acquisition sensors are arranged according to the characteristics of the S-shaped bend, and the sensor position layout is adjusted using a genetic algorithm to improve the pipeline defect detection rate. The magnetic signal data collected by different devices in the signal acquisition system are matched according to location and time tags. The data characteristics of different devices are integrated to build a complete visual model of the pipeline and mark suspected defect points. Potential defects can be discovered in a timely manner through the visual model and response plans can be formulated.

2. A method for visualizing weak magnetic signals of pipeline defects according to claim 1, characterized in that: The method for adjusting the sensor position layout by using a genetic algorithm includes: Obtain the preset coverage radius of each magnetic signal sensor, set the sensor's location conditions and layout area conditions; obtain the number of sensors that can be arranged in the area based on the preset coverage radius and layout area conditions of the magnetic signal sensor , generating an initialization population of magnetic signal sensor layout schemes; By calculating the fitness value of each individual in the initialized population, selecting individuals with higher fitness for reproduction, combining the sensor positions of two individuals using multi-point crossover, adjusting the positions of individual sensors, or increasing or decreasing the number of sensors, a new population is generated; iteratively repeating the evaluation, selection, crossover, and mutation until the fitness change is less than a preset threshold, thereby obtaining a new magnetic field sensor layout; The sensor location condition is the limiting condition that each sensor location must meet, and the layout area condition is the limiting condition for the area where multiple sensors are installed together; the initialization population is composed of Individual Each individual is a magnetic signal sensor layout scheme, including the combination of sensor positions and sensor quantities in the area, expressed as: ,in Indicates the location of the sensor. Indicates the number of sensors; According to the preset coverage radius of the magnetic signal sensor, the magnetic signal sensors that exceed the area limitation conditions in the individual are eliminated, and the magnetic signal sensors that overlap the coverage range in the individual are eliminated or rearranged to obtain a new magnetic signal sensor initialization population Individual .

3. A method for visualizing weak magnetic signals of pipeline defects according to claim 2, characterized in that: The method for calculating the fitness value of each individual in the initialized population includes: An initialization population of magnetic signal sensors is generated according to the limited conditions of the magnetic signal sensor layout, and an iterative fitness function of the magnetic signal sensor population is designed to perform fitness evaluation. The fitness function is: ； in, is the distance from spatial point p to the position of the i-th sensor A is the area limitation condition for the joint installation of multiple sensors; The fitness of each individual in the initialized population is calculated separately. The higher the sensor coverage rate among the individuals, the smaller the average value of the magnetic difference in the area and the fewer the number of sensors in the area, the higher the fitness of the individuals; individuals with higher fitness are selected for reproduction to form a new magnetic signal sensor population.

4. A method for visualizing weak magnetic signals of pipeline defects according to claim 1, characterized in that: The method of repeatedly evaluating, selecting, crossing, and mutating until the fitness change is less than a preset threshold to obtain a new magnetic field sensor layout includes: The magnetic signal sensor positions of two individuals are combined by single-point crossover or multi-point crossover to generate a new magnetic signal sensor layout; the positions of individual sensors are then randomly adjusted to increase or decrease the number of sensors, increase the diversity of the population, and generate a new population. Individual ; Calculate the fitness of each individual in the new population, select excellent individuals based on the fitness, and then perform single-point crossover or multi-point crossover to combine the magnetic signal sensors of two individuals, adjust the position of individual sensors, increase or decrease the number of sensors, generate a new generation of individuals, and obtain a new generation of population. Individual ; Repeat the evaluation, selection, crossover and mutation to obtain a new population until the fitness change is less than the preset threshold. Then stop the iteration and obtain the optimal population under the preset conditions, which is the optimal layout plan of the sensors under the preset conditions in the area.

5. The method for visualizing weak magnetic signals of pipeline defects according to claim 1, characterized in that: The method for obtaining weak magnetic signals on or inside a pipeline by combining multiple magnetic signal detection devices includes: The pipeline is coated, and the pipeline signal acquisition system is composed of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station; the inner pipeline magnetic data is scanned and detected from the inside of the pipeline by the inner pipeline intelligent crawling detection device; the outer pipeline detector is used to detect defects in the metal pipeline layer through the coating layer; the fixed monitoring station fixes the equipment at key nodes for long-term detection; the key nodes are pipeline interfaces, valves, bending points and extrusion deformation points; by arranging detection equipment inside, outside and at key points of the pipeline with a coating, a multi-device collaborative detection solution is formed.

6. A method for visualizing weak magnetic signals of pipeline defects according to claim 5, characterized in that: The multi-device collaboration method includes: Dynamic inspection trajectories are designed for the inner pipeline intelligent crawling detection device and the outer pipeline detector based on the characteristics of the pipeline shape. The movement speed of the inner pipeline intelligent crawling detection device is reduced by 30% when passing through the S-shaped area of ​​the pipeline. A spiral merging and folding movement mode is used for mobile detection. The number of sensor activations at the bends is increased to increase the number of magnetic signal detections in the S-shaped area of ​​the pipeline, while also increasing the sampling frequency. The detection path of the external pipeline detector is designed using the pipeline structure, a hovering scanning point is designed for the extrusion deformation of the S-shaped bend, a high-power pulse magnetic field generator is set at the front end of the bend, a frequency-adjustable alternating magnetic field source is used in the pipe diameter change section, and a magnetic shielding device is set near the key support points to reduce interference.

7. The method for visualizing weak magnetic signals of pipeline defects according to claim 1, characterized in that: The method of timely discovering potential defects through visual models and formulating response plans includes: By establishing a multi-device collaborative detection scheme to form a master-slave signal calibration mechanism, mutual interference is reduced according to the time-division multiplexing sampling scheme; the sensor array sensitivity and sampling frequency are dynamically adjusted according to the initial detection results, high-risk areas are automatically identified based on the detection data, additional detection resources are dynamically allocated, and detection equipment is dynamically adjusted in real time.

8. A pipeline defect weak magnetic signal visualization construction system, used to execute the method according to any one of claims 1 to 7, characterized in that: The system comprises: a first data acquisition module, a second layout optimization module, and a data integration module, each module being connected in sequence; The first data acquisition module is used to construct a pipeline signal acquisition system based on the structural characteristics and functional features of the pipeline. The pipeline has a coating layer. The pipeline signal acquisition system consists of an inner pipeline intelligent crawling detection device, an outer pipeline detector and a fixed monitoring station. The weak magnetic signal on the surface or inside of the pipeline is acquired by combining multiple magnetic signal detection devices. The second layout optimization module is used to arrange the signal acquisition sensors according to the characteristics of the S-shaped bend pipe, and adjust the sensor position layout through a genetic algorithm to improve the pipeline defect detection rate; a certain part of the pipeline is an S-shaped bend pipe, and the S-shaped bend pipe includes two 45° bends, and a portion of the pipe diameter is squeezed and deformed in the middle section; The data integration module is used to match the magnetic signal data collected by different devices in the signal acquisition system according to location and time tags, use the characteristics of different devices to integrate data to build a complete visual model of the pipeline, and mark suspected defect points; potential defects are discovered in a timely manner through the visual model, and response plans are formulated.

Images