A pipeline detection method and system based on a UAV

Abstract

The application discloses a pipeline detection method and system based on a UAV, relates to the technical field of UAV detection, and comprises the following steps: considering the self condition of a to-be-detected pipeline to select a reasonable UAV collection point, providing a reliable basis for subsequent flight plan planning and pipeline detection of the UAV; considering the position environment of the pipeline and the demand for flight stability to formulate an initial flight plan, guaranteeing the adaptability of UAV flight; judging whether the second flight is needed based on the quality of the first detection information, judging whether the first flight is sufficient to support the quality demand of pipeline detection according to the quality of the detection information, considering the complementarity and intersection between the first detection information and the second detection information to output the detection result of the pipeline, improving the adaptability and accuracy of pipeline detection, guaranteeing the adaptability of the UAV flight condition and the pipeline environment, and meeting the demand of the complex environment of the pipeline for detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle detection, and in particular to a pipeline detection method and system based on unmanned aerial vehicles. BACKGROUND

[0002] In traditional pipeline inspection, manual inspection faces problems such as complex terrain, low efficiency, high cost, high risk, and the like, and is difficult to meet the demand for large-scale pipeline safety monitoring. With the rapid development of unmanned aerial vehicle technology and artificial intelligence, a pipeline detection scheme based on unmanned aerial vehicles has emerged. The scheme integrates technologies such as unmanned aerial vehicle flight platforms, high-definition sensors, AI image recognition, edge computing, and three-dimensional modeling, and can efficiently cover complex terrain and collect pipeline and surrounding environment data in real time. By carrying high-definition cameras, laser radars, infrared thermal imagers, and the like, the unmanned aerial vehicle can accurately identify defects such as pipeline surface cracks, corrosion, and leakage, and use AI algorithms to analyze data in real time. At the same time, three-dimensional real scene modeling technology can convert image data into high-precision three-dimensional models to assist in defect positioning and risk assessment. The scheme breaks through the geographical limitations of traditional inspection and significantly improves the detection efficiency and accuracy, providing an intelligent and digital solution for pipeline safe operation.

[0003] In the prior art, the influence of the unmanned aerial vehicle flight condition on the pipeline detection information is not considered, that is, the adaptability of the unmanned aerial vehicle flight condition to the pipeline detection environment is not considered, resulting in poor adaptability and accuracy of pipeline detection, which cannot meet the complex and variable pipeline detection requirements.

[0004] Therefore, how to improve the adaptability and accuracy of pipeline detection is a technical problem to be solved at present. SUMMARY

[0005] The present application aims to solve the problem in the prior art that the adaptability and accuracy of pipeline detection are poor due to the lack of consideration of the adaptability of the unmanned aerial vehicle flight condition to the pipeline detection environment, and proposes a pipeline detection method based on unmanned aerial vehicles, which comprises,

[0006] obtaining pipeline information, pipeline location environment information, a pipeline detection task, and unmanned aerial vehicle carried sensor information of a pipeline to be detected, setting a collection section of the pipeline according to the pipeline information and the pipeline detection task of the pipeline to be detected, and formulating a first flight plan based on the pipeline location environment information and the collection section of the pipeline.

[0007] The unmanned aerial vehicle flies according to the first flight plan, collects first detection information of the pipeline, and calculates the flight information deviation between the first actual flight and the first flight plan, evaluates the quality of the first detection information, and determines whether a second flight is needed based on the quality of the first detection information.

[0008] If a second flight is needed, a second flight plan is optimized according to the flight information deviation, and the unmanned aerial vehicle flies according to the second flight plan to collect second detection information of the pipeline;

[0009] The first detection information and the second detection information of the pipeline are integrated to output a detection result of the pipeline.

[0010] In some embodiments of the present application, the collection section of the pipeline is set according to pipeline information of the pipeline to be detected and a pipeline detection task, including,

[0011] The pipeline information includes pipeline length, pipeline direction, pipeline material and pipeline conveying medium;

[0012] The pipeline simple scene and the pipeline complex scene are defined according to the pipeline length and the pipeline direction, and for the pipeline simple scene, the whole pipeline to be detected is taken as the collection section of the pipeline to be detected.

[0013] For the pipeline complex scene, a pipeline model of the pipeline to be detected is established according to the pipeline length, the pipeline direction, the pipeline material and the pipeline conveying medium, all risk elements are collected through the pipeline detection task, and high-risk positions are marked on corresponding positions of the pipeline model of the pipeline to be detected by integrating all risk elements, and the high-risk positions are taken as the collection section of the pipeline to be detected.

[0014] In some embodiments of the present application, the pipeline simple scene and the pipeline complex scene are defined according to the pipeline length and the pipeline direction, including,

[0015] A plurality of key features on the pipeline direction are extracted, the pipeline is divided into a plurality of sections according to the pipeline length, the key feature density of each section is counted, the average value of the key feature density is obtained, the scene complexity is defined based on the pipeline length, the plurality of key features and the average value of the key feature density, and the pipeline simple scene and the pipeline complex scene are distinguished by the scene complexity.

[0016] In some embodiments of the present application, the first flight plan is formulated based on the pipeline position environment information and the collection section of the pipeline, including,

[0017] The pipeline information and the pipeline position environment information are imported into the GIS software, and the obstacle distribution map in the pipeline position environment is superimposed, the collection section position of the pipeline is marked in the GIS software, a plurality of collection section positions are taken as passing points, a plurality of candidate paths are generated by a path planning algorithm, and each candidate path passes through all passing points;

[0018] Each candidate path is divided into a plurality of path units, and each candidate path is evaluated in terms of path cost and flight risk cost based on the path units, a comprehensive cost is generated by combining the path cost and the flight risk cost, and a comprehensive cost change curve varying with the path units is drawn;

[0019] The stable part curve and the fluctuation part curve are distinguished on the comprehensive cost change curve, the relative cost of the stable part curve and the fluctuation part curve is respectively calculated through the change of the curve slope, the relative comprehensive cost is determined, and a candidate path is screened out as the target path according to the relative comprehensive cost;

[0020] Flight information of the unmanned aerial vehicle is configured on the basis of the target path.

[0021] In some embodiments of the present application, flight information of the unmanned aerial vehicle is configured on the basis of the target path, including,

[0022] The target path is divided into a flight path and a collection path according to the position of the collection section of the pipeline;

[0023] For the flight path, the flight information of the unmanned aerial vehicle is configured by analyzing the demand of the environment for flight stability through the GIS software and combining the flight task demand;

[0024] For the collection path, the flight information of the unmanned aerial vehicle is configured based on the demand of the pipeline for flight stability, the demand of the environment for flight stability and the flight task demand by analyzing the demand of the pipeline for flight stability and the demand of the environment for flight stability through the GIS software.

[0025] In some embodiments of the present application, the flight information deviation between the first actual flight and the first flight plan is counted, including,

[0026] The deviations of the flight information of the flight path and the collection path are counted respectively to obtain the first deviation and the second deviation, the change curves of the first deviation and the second deviation with the position are drawn, the deviation points of the first deviation and the second deviation are identified on the change curves respectively, and the correlation between the two types of deviation points is analyzed.

[0027] In some embodiments of the present application, the second flight plan is optimized according to the flight information deviation, including,

[0028] The flight information deviation and the quality of the first detection information are spatially aligned through the position of the pipeline, and the second flight plan is optimized based on the deviation points and the correlation between the deviation points.

[0029] In some embodiments of the present application, the detection result of the pipeline is output by integrating the first detection information and the second detection information of the pipeline, including,

[0030] The intersection information part and the difference information part between the first detection information and the second detection information of the pipeline are determined according to the position of the pipeline;

[0031] The intersection information part of the first detection information and the second detection information is informationally fused to obtain fused information;

[0032] Based on the difference set information part and the fusion information output pipeline detection result.

[0033] Correspondingly, the application also provides a pipeline detection system based on a UAV, comprising,

[0034] The first module is used for acquiring pipeline information, pipeline location environment information, a pipeline detection task and UAV-carrying sensor information of a pipeline to be detected, setting a collection section of the pipeline according to the pipeline information and the pipeline detection task of the pipeline to be detected, and formulating a first flight plan based on the pipeline location environment information and the collection section of the pipeline.

[0035] The second module is used for flying the UAV according to the first flight plan, collecting first detection information of the pipeline, counting flight information deviation between the first actual flight and the first flight plan, evaluating the quality of the first detection information, and judging whether a second flight is needed based on the quality of the first detection information.

[0036] The third module is used for optimizing the second flight plan according to the flight information deviation if the second flight is needed, flying the UAV according to the second flight plan, and collecting second detection information of the pipeline.

[0037] The fourth module is used for outputting a detection result of the pipeline by comprehensively integrating the first detection information and the second detection information of the pipeline.

[0038] Compared with the prior art, the application has the following beneficial effects:

[0039] 1. The collection section of the pipeline is set according to the pipeline information and the pipeline detection task of the pipeline to be detected, the reasonable UAV collection point is selected by considering the self condition of the pipeline to be detected, and a reliable foundation is provided for the subsequent flight plan planning of the UAV and pipeline detection. The first flight plan is formulated based on the pipeline location environment information and the collection section of the pipeline, the initial flight plan is formulated by considering the location environment condition of the pipeline and the demand for flight stability, and the adaptability of the UAV flight is ensured.

[0040] 2. Whether the second flight is needed is judged based on the quality of the first detection information, the first flight is judged according to the quality of the detection information, the second flight is performed when the first flight is difficult to support, the second flight plan is optimized according to the flight information deviation, the loopholes of the first flight are filled, the reliability and adaptability of the second flight are ensured, the detection result of the pipeline is output by considering the complementarity and intersection between the first detection information and the second detection information, the adaptability and accuracy of the pipeline detection are improved, the adaptability of the UAV flight condition and the pipeline environment is ensured, and the demand of the complex environment of the pipeline for detection is met. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 A flowchart of a pipeline detection method based on a UAV proposed in the present application is shown in the figure.

[0042] Figure 2 A structural diagram of a pipeline detection system based on a UAV proposed in the present application is shown in the figure. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application.

[0044] Referring to Figure 1 A pipeline detection method based on a UAV, comprising the following steps:

[0045] In step S101, pipeline information, pipeline location environment information, pipeline detection tasks and UAV-carrying sensor information of a pipeline to be detected are acquired, a collection section of the pipeline is set according to the pipeline information and the pipeline detection tasks of the pipeline to be detected, and a first flight plan is formulated based on the pipeline location environment information and the collection section of the pipeline.

[0046] In this embodiment, pipelines are important components of energy transportation, urban water supply and drainage systems, and their safety and stability are crucial. In order to timely detect potential problems in pipelines and ensure the normal operation of pipeline systems, using drones for pipeline detection has become an efficient and convenient method. Obtaining detailed information about the pipeline to be detected is the basis for developing a flight plan. This information includes the type of pipeline (such as oil and gas pipelines, natural gas pipelines, water supply and drainage pipelines, etc.), the material of the pipeline, the diameter and length of the pipeline, etc. Different types of pipelines may require different sensors and detection methods during detection, the material and diameter of the pipeline will affect the detection sensitivity of the sensor to pipeline defects, and the buried depth of the pipeline will affect the flight height of the drone and the penetration ability of the detection equipment. Understanding the environmental information of the location where the pipeline is located is crucial for developing a flight plan. This includes the topography of the pipeline's surroundings (such as mountains, plains, cities, etc.), weather conditions (such as wind speed, wind direction, temperature, humidity, etc.), the distribution of surrounding buildings and obstacles, etc. The topography will affect the selection of the flight route and landing points of the drone, the weather conditions will affect the flight safety of the drone and the accuracy of the detection data, and the surrounding buildings and obstacles need to be avoided by the drone during flight to ensure flight safety. Clearly defining the pipeline detection task is crucial for determining the detection target and requirements. Detection tasks may include detecting whether the pipeline has defects such as leaks, corrosion, deformation, and rupture, as well as determining the location, size, and severity of the defects. Different detection tasks require different sensors and detection methods, and have different requirements for the accuracy and completeness of the detection data. Understanding the performance and characteristics of the sensors carried by the drone is an important basis for developing a flight plan. Sensors may include high-resolution cameras, thermal imagers, laser radars, gas sensors, etc. Different sensors have different detection principles and application ranges, for example, high-resolution cameras can be used to detect the deformation and damage of the pipeline surface, thermal imagers can be used to detect pipeline leaks and abnormal temperature distribution, laser radars can be used to obtain three-dimensional terrain information around the pipeline, and gas sensors can be used to detect whether there is a flammable or toxic gas leak around the pipeline.

[0047] In some embodiments of the present application, the collection section of the pipeline is set according to the pipeline information and the pipeline detection task of the pipeline to be detected, including,

[0048] The pipeline information includes the length of the pipeline, the direction of the pipeline, the material of the pipeline, and the medium transported in the pipeline;

[0049] The simple pipeline scenario and the complex pipeline scenario are defined according to the length of the pipeline and the direction of the pipeline. For the simple pipeline scenario, the entire pipeline to be detected is taken as the collection section of the pipeline to be detected;

[0050] For a complex pipeline scene, a pipeline model of the pipeline to be tested is established according to the pipeline length, pipeline direction, pipeline material and pipeline conveying medium, all risk elements are collected through the pipeline detection task, high-risk positions are marked on the corresponding positions of the pipeline model of the pipeline to be tested by comprehensively considering all risk elements, and the high-risk positions are used as the collection section of the pipeline to be tested.

[0051] In this embodiment, the following information is extracted from the pipeline design drawings or database:

[0052] Pipeline length (L): unit is meter (m), used for dividing sections.

[0053] Pipeline direction: describes the extension direction of the pipeline through geographic coordinates or mileage coordinates.

[0054] Pipeline material (such as steel pipe, plastic pipe): affects the corrosion rate and defect type.

[0055] Pipeline conveying medium (such as oil, natural gas, water): affects the leakage risk and detection focus.

[0056] Key features such as bends, branches, undulations, slope changes and intersection points are extracted from the pipeline direction. Simple scenes and complex scenes are divided. For a simple scene, the entire pipeline can be used as a collection section. For a complex scene, in order to balance the efficiency and accuracy of detection, part of the pipeline needs to be selected as a collection section.

[0057] Risk element collection: according to the pipeline detection task, possible risk elements (such as stress concentration at the bend, leakage risk at the branch, material corrosion sensitive area, etc.) are collected.

[0058] High-risk position marking: map the risk elements to the pipeline model and mark the high-risk positions (such as bends, branches, material aging areas, etc.).

[0059] Collection section setting: the high-risk positions and their surrounding areas are used as the collection section, and the collection density is increased (such as reducing the flight height and increasing the collection angle).

[0060] In some embodiments of the application, the pipeline simple scene and the pipeline complex scene are defined according to the pipeline length and the pipeline direction, including,

[0061] A plurality of key features on the pipeline direction are extracted, the pipeline is divided into a plurality of sections according to the pipeline length, the density of the key features in each section is counted, the average value of the key feature density is obtained, the scene complexity is defined based on the pipeline length, the plurality of key features and the average value of the key feature density, and the pipeline simple scene and the pipeline complex scene are distinguished through the scene complexity.

[0062] In this embodiment, the scene complexity calculation formula is as follows:

[0063] ;

[0064] wherein, is the scene complexity, is the pipe length coefficient, is the number of key features, is the combination weight of the th key feature, is the complexity of the th key feature, the key features can include bends, branches, and undulations, and the complexity can be the angle of the bend, the branch level, and the undulation degree, is the density average (average frequency of occurrence under the section) of the th key feature, is the first constant of the th key feature, represents the correction of the density average of the key feature to the sum of the complexity of the key feature.

[0065] Key feature extraction: extracting key features such as bends, branches, undulations, slope changes, and intersection points from the pipe route.

[0066] Section division: dividing the pipe into multiple continuous sections according to the pipe length (e.g., one section per 100 meters).

[0067] Feature density calculation: counting the number of key features in each section and calculating the key feature density (number of features / section length).

[0068] Quantify the scene complexity through the key feature density and the pipe length, avoid subjective judgment, and improve the scientificity of the collection section setting. In a complex scene, high-risk positions (such as bends, branches, and material aging areas) are high-defect areas and need to be collected.

[0069] In some embodiments of the present application, the first flight plan is formulated based on the pipe location environment information and the collection section of the pipe, including,

[0070] Import the pipe information and the pipe location environment information into the GIS software, superimpose the obstacle distribution map in the pipe location environment, mark the collection section position of the pipe in the GIS software, take multiple collection section positions as passing points, and generate multiple candidate paths through the path planning algorithm, each candidate path passing through all passing points;

[0071] Split each candidate path into multiple path units, evaluate the path cost and flight risk cost of each candidate path based on the path units, combine the path cost and flight risk cost to generate a comprehensive cost, and draw a comprehensive cost change curve varying with the path units;

[0072] Distinguish between the stable and fluctuating portions of the overall cost curve. Calculate the relative cost of each portion of the curve by changing the slope of the curve, thereby determining the relative overall cost. Then, select a candidate path as the target path based on the relative overall cost.

[0073] Configure the drone's flight information based on the target path.

[0074] In this embodiment, the path planning algorithm uses A* algorithm, RRT (fast random tree) or ant colony algorithm to generate multiple candidate paths, ensuring that each path passes through all waypoints.

[0075] Obstacle overlay: Overlay obstacle distribution maps (such as buildings, trees, high-voltage lines, etc.) in GIS.

[0076] Collection segment labeling: Mark the collection segment locations of the pipeline (such as high-risk locations or complex scene sections) as transit points.

[0077] Path unit splitting: Each candidate path is split into multiple path units (e.g., each unit is 10 meters).

[0078] Path cost calculation:

[0079] Energy consumption cost: Calculate energy consumption (unit: joules / meter) based on the drone's flight speed, altitude, and payload.

[0080] Risk cost calculation:

[0081] Obstacle distance cost: The closer you are to an obstacle, the higher the cost (e.g., when the distance is less than 5 meters, the cost is 100; when the distance is between 5 and 10 meters, the cost is 50).

[0082] Because the two types of costs change at different locations, cost evaluation is based on path units. A comprehensive cost is generated by combining path costs and flight risk costs (this can be achieved through weighted summation, etc.).

[0083] Input data: Comprehensive cost based on path unit decomposition (path cost + flight risk cost).

[0084] Curve plotting: Plot a continuous curve of changes in comprehensive cost with the path unit as the horizontal axis (X-axis) and the comprehensive cost as the vertical axis (Y-axis).

[0085] The stable part curve is a curve of a part where the comprehensive cost appears more frequently, and the fluctuant part curve is a curve of a part where the comprehensive cost appears less frequently. The average value of the slope change of the stable part curve and the fluctuant part curve is calculated, and the representative value of the comprehensive cost of the stable part curve and the fluctuant part curve is obtained according to the frequency of each comprehensive cost in the comprehensive cost interval of the stable part curve and the fluctuant part curve respectively. The relative comprehensive cost is obtained based on the representative value of the comprehensive cost and the average value of the slope change, and the calculation formula is as follows:

[0086] ;

[0087] wherein, is the i th candidate path (corresponding to the comprehensive cost change curve), is the i th candidate path (corresponding to the comprehensive cost change curve), , is the cost weight of the stable part curve and the fluctuant part curve respectively. Generally, the cost weight of the stable part curve is higher than the cost weight of the fluctuant part curve, , is the representative value of the comprehensive cost of the stable part curve and the fluctuant part curve of the i th candidate path respectively, , is the average value of the slope change of the stable part curve and the fluctuant part curve of the i th candidate path respectively, , is the average value of the slope change of the stable part curve and the fluctuant part curve of the i th candidate path respectively, , is the second constant and the third constant corresponding to the i th candidate path respectively, , and indicate the correction of the average value of the slope change to the representative value of the comprehensive cost. The more stable the change is, the greater the credibility is. The second constant and the third constant are used to balance the size of the correction function.

[0088] In some embodiments of the present application, the flight information of the unmanned aerial vehicle is configured on the basis of the target path, including,

[0089] The target path is divided into a flight path and a collection path according to the collection segment position of the pipeline;

[0090] For the flight path, the flight information of the unmanned aerial vehicle is configured by analyzing the environmental demand for flight stability through GIS software and combining the flight task demand;

[0091] For the collection path, the flight information of the unmanned aerial vehicle is configured based on the pipeline demand for flight stability, the environmental demand for flight stability and the flight task demand by analyzing the pipeline demand for flight stability and the environmental demand for flight stability through GIS software.

[0092] In this embodiment, the flight path: the unmanned aerial vehicle in the target path only needs to fly to the transition section between the collection sections (without performing the collection task).

[0093] The collection path: the section in the target path where the unmanned aerial vehicle needs to collect data of the pipeline (such as shooting and detection).

[0094] The environmental stability requirement: analyze the terrain (such as plains and mountainous areas), weather (such as wind speed and precipitation), and obstacle distribution (such as buildings and high-voltage lines) of the flight path through GIS software.

[0095] Flight task requirements: such as path requirements and time requirements.

[0096] Configuration parameters:

[0097] Flight height and attitude: adjusted according to the height of obstacles (such as flying above high-voltage lines with a height of >20 meters).

[0098] Flight speed: adjusted according to wind speed (such as speed = 8 m / s when wind speed <3 m / s, and speed = 5 m / s when wind speed >5 m / s).

[0099] Obstacle avoidance strategy: enable radar or visual obstacle avoidance system.

[0100] Pipeline stability requirement: pipeline type (such as buried and overhead) and collection accuracy requirement (such as low-speed flight for high-precision collection).

[0101] Environmental stability requirement: same as flight path analysis.

[0102] Configuration parameters:

[0103] Flight height: adjusted according to the position of the pipeline (such as flying at a height of 10 meters for overhead pipelines).

[0104] Flight speed: adjusted according to the collection accuracy (such as speed = 3 m / s for high-precision collection).

[0105] Shooting angle: adjusted according to the direction of the pipeline (such as vertical shooting or inclined shooting).

[0106] Obstacle avoidance strategy: enable obstacle avoidance system to ensure the safety of the collection equipment.

[0107] In step S102, the unmanned aerial vehicle flies according to the first flight plan, collects the first detection information of the pipeline, calculates the flight information deviation between the first actual flight and the first flight plan, evaluates the quality of the first detection information, and determines whether a second flight is needed based on the quality of the first detection information.

[0108] In this embodiment, the flight task is started: the UAV executes the flight task according to the first flight plan (including target path, flight height, speed, acquisition parameters, etc.) generated in advance.

[0109] Data acquisition: in the acquisition path segment, the UAV acquires the first detection information (such as surface defects, temperature distribution, deformation data) of the pipeline through the sensors (such as high-definition cameras, infrared thermal imagers, laser radars) carried by the UAV.

[0110] Deviation type:

[0111] Path deviation: lateral / longitudinal deviation (unit: meters) between the actual flight trajectory and the target path.

[0112] Position deviation: position deviation (unit: meters) between the actual arrival point and the planned point.

[0113] Attitude deviation: deviation (unit: degrees) between the actual flight attitude (pitch angle, roll angle, yaw angle) and the planned attitude.

[0114] Speed deviation: deviation (unit: meters / second) between the actual flight speed and the planned speed.

[0115] Deviation statistical method:

[0116] Compare the flight log (such as GPS data, IMU data) recorded by the UAV flight control system with the planned parameters to calculate the mean and maximum of each deviation value.

[0117] Quality evaluation of first detection information (list, not exhaustive)

[0118] Evaluation index:

[0119] Data integrity: whether the acquired pipeline data covers all planned acquisition segments (such as missing segment length ratio).

[0120] Data clarity: whether the resolution of image or point cloud data meets the detection requirements (such as the proportion of blurred images).

[0121] Judgment rule:

[0122] Flight deviation impact: if the path deviation or attitude deviation leads to data loss or blur (such as the proportion of blurred images > 5%), a second flight is required.

[0123] Data quality threshold: if the data integrity < 90% or the data accuracy (false detection rate + missed detection rate) > 5%, a second flight is required.

[0124] In some embodiments of the present application, the flight information deviation between the first actual flight and the first flight plan is counted, including,

[0125] The deviations of flight information of the flight path and the collection path are respectively counted to obtain a first deviation and a second deviation, and the variation curves of the first deviation and the second deviation with respect to positions are drawn, deviation points of the first deviation and the second deviation are respectively identified on the variation curves, and the correlation between the two types of deviation points is analyzed.

[0126] In this embodiment, the deviation types are:

[0127] Flight path deviation (first deviation): The deviation of actual flight parameters (such as path, position, attitude, speed) of the flight path segment from planned parameters is counted.

[0128] Collection path deviation (second deviation): The deviation of actual flight parameters of the collection path segment from planned parameters is counted.

[0129] Statistical method:

[0130] The flight log is derived from the unmanned aerial vehicle flight control system, and actual flight data (GPS coordinates, IMU attitude, speed, etc.) are extracted.

[0131] The actual data and the planned data are aligned according to time or position, and the deviation value (such as actual position-planned position) is calculated.

[0132] Curve drawing rule:

[0133] The path position (such as the number of points or the cumulative distance) is taken as the horizontal axis (X-axis), and the deviation value is taken as the vertical axis (Y-axis).

[0134] The flight path deviation curve and the collection path deviation curve are respectively drawn.

[0135] Identification rule:

[0136] Deviation point definition: The point whose deviation value exceeds the threshold value (such as path deviation>1.0 meter, attitude deviation>2.0°).

[0137] Identification method: The deviation point is marked by a sliding window or a peak detection algorithm.

[0138] Correlation analysis of two types of deviation points

[0139] Analysis method:

[0140] Cause correlation: The common causes of deviation points (such as sudden change of wind speed, complex terrain leading to unstable flight attitude) are analyzed.

[0141] Causal correlation, the first type of deviation leads to the second type of deviation:

[0142] To determine whether the flight path deviation point (first type of deviation point) leads to the collection path deviation point (second type of deviation point), the following aspects need to be considered:

[0143] Spatial correlation: Whether two types of deviation points occur in close locations or the same flight segment.

[0144] Temporal correlation: Whether two types of deviation points occur consecutively in time (e.g., flight path deviation followed by immediate collection path deviation).

[0145] Causal logic: Whether the first type of deviation point directly or indirectly causes the second type of deviation point through physical mechanisms (e.g., attitude loss of control, sudden speed change).

[0146] Spatial correlation analysis

[0147] Method:

[0148] Mark flight path deviation points and collection path deviation points on the flight trajectory diagram and check if they overlap or are adjacent.

[0149] Temporal correlation analysis

[0150] Method:

[0151] According to the flight log timestamp, check if two types of deviation points occur consecutively in a short period of time.

[0152] Causal logic analysis

[0153] Possible mechanisms:

[0154] Path deviation → attitude deviation:

[0155] Flight path deviation (e.g., deviating from the planned route) may cause the UAV to adjust its attitude to correct the trajectory, thereby causing attitude deviation.

[0156] Path deviation → speed deviation → attitude deviation:

[0157] Path deviation may trigger speed adjustment (e.g., slowing down to correct the trajectory), and sudden speed change may cause attitude instability.

[0158] Environmental factors superimposed:

[0159] Path deviation may be caused by environmental interference (e.g., sudden wind speed change), and the same environmental interference may also cause collection path deviation.

[0160] Step S103, if a second flight is needed, optimize the second flight plan according to the flight information deviation, and the UAV flies according to the second flight plan to collect the second detection information of the pipeline.

[0161] In some embodiments of the present application, optimizing the second flight plan according to the flight information deviation includes,

[0162] The flight information deviation and the quality of the first detection information are spatially aligned by pipeline position, and the second flight plan is optimized based on the deviation points and the correlation between the deviation points.

[0163] In this embodiment, the pipeline position mapping maps the flight information deviation (path deviation, attitude deviation, etc.) and the quality of the first detection information (data integrity, clarity, accuracy) to the same spatial coordinate system based on the actual position of the pipeline (such as the milepost number, GPS coordinates). A "deviation-quality" joint distribution diagram is generated to clearly show the spatial correspondence between the deviation and the quality problem.

[0164] Optimization strategy:

[0165] Path optimization:

[0166] Increase the density of passing points in the high-deviation-point section (such as milepost numbers 2-3 km), reduce the flight interval, and improve the trajectory control accuracy.

[0167] Adjust the flight height or speed to reduce environmental interference (such as reducing the height to avoid high-rise building wind disturbance).

[0168] Attitude control optimization:

[0169] Enable more stringent attitude stabilization algorithms (such as PID control parameter adjustment) in the high-attitude-deviation section (such as near milepost number 5 km).

[0170] Increase the attitude sensor redundancy (such as dual-IMU configuration) to improve the reliability of attitude data.

[0171] Collection strategy optimization:

[0172] Increase the number of repeated collections (such as flying back and forth twice) in the low-quality-point section (such as at milepost number 5 km) to ensure data coverage.

[0173] Adjust the sensor parameters (such as camera exposure time, laser radar point cloud density) to improve data quality.

[0174] Both deviation and quality problems are related to pipeline position, and need to be analyzed in the same spatial dimension to establish causal relationship. Avoid isolating deviation and quality problems, and improve optimization targeting. Through correlation analysis, the causal chain of deviation and quality problems can be identified to avoid blind optimization. From the three dimensions of path, attitude, and collection, collaborative optimization is carried out to ensure flight stability and data quality.

[0175] Step S104, the first detection information and the second detection information of the pipeline are integrated to output the detection result of the pipeline.

[0176] In some embodiments of the present application, the first detection information and the second detection information of the pipeline are integrated to output the detection result of the pipeline, which includes,

[0177] According to the position of the pipeline, the intersection information part and the difference set information part between the first detection information and the second detection information are determined;

[0178] The intersection information part of the first detection information and the second detection information is information fused to obtain fused information;

[0179] Based on the difference set information part and the fused information, the detection result of the pipeline is output.

[0180] In this embodiment, the actual position of the pipeline (such as the milepost number, GPS coordinates) is taken as the reference, and the first detection information and the second detection information are mapped to the same spatial coordinate system.

[0181] The intersection information part is defined as the pipeline section (such as the milepost number 0km-10km) covered by both detections.

[0182] The difference set information part is defined as the pipeline section (such as the milepost number 10km-15km covered only by the first detection).

[0183] Data weighted fusion: The two detection results of the same section are given weights (such as the second detection weight is higher, because the quality is more reliable after optimization).

[0184] Difference set information processing:

[0185] The section covered only by the first detection (such as the milepost number 10km-15km): directly use the first detection result.

[0186] The process of outputting the detection result of the pipeline according to the detection information:

[0187] The pipeline information collected by the unmanned aerial vehicle usually includes image data (visible light / infrared / laser radar), position data (GPS / RTK), attitude data (roll / pitch / yaw angle), etc. The analysis process needs to combine multi-source data, through four steps of preprocessing, feature extraction, defect identification and result integration to get the detection conclusion.

[0188] 1. Data preprocessing

[0189] Image data:

[0190] Noise removal: Gaussian filter, median filter are used to remove image noise.

[0191] Geometric correction: correct image distortion according to unmanned aerial vehicle attitude data (such as pitch angle deviation).

[0192] Radiometric correction: compensate for uneven illumination (such as solar radiation interference of infrared image).

[0193] Position data:

[0194] Coordinate Conversion: Convert GPS coordinates to pipeline mile markers for alignment with historical data.

[0195] Trajectory Smoothing: Correct abnormal points in flight trajectory (e.g., jumps caused by GPS signal loss) using Kalman filtering.

[0196] 2. Feature Extraction and Defect Identification

[0197] Image Feature Extraction:

[0198] Texture Features: Extract pipeline surface texture (e.g., roughness of corrosion areas) using Gray Level Co-occurrence Matrix (GLCM).

[0199] Geometric Features: Identify geometric defects (e.g., cracks, deformations) through edge detection (e.g., Canny algorithm).

[0200] Deep Learning Models:

[0201] Object Detection: Identify pipeline defects (e.g., corrosion, cracks, deformations) using models like YOLO, Faster R-CNN.

[0202] Semantic Segmentation: Segment defect areas and calculate area proportion using models like U-Net, DeepLab.

[0203] Multi-source Data Fusion:

[0204] Laser Radar Point Cloud: Extract three-dimensional deformation data (e.g., diameter changes, bending) of the pipeline.

[0205] Infrared Thermal Imaging: Identify leakage points (e.g., temperature anomalies caused by natural gas leaks).

[0206] 3. Defect Classification and Quantification

[0207] Defect Classification:

[0208] By Type: Corrosion, cracks, deformations, leaks, etc.

[0209] By Severity:

[0210] Mild: Defect area proportion <5%, no structural risk.

[0211] Moderate: 5% ≤ defect area proportion <15%, requires regular monitoring.

[0212] Severe: Defect area proportion ≥15%, requires immediate repair.

[0213] Quantitative Indicators:

[0214] Corrosion Depth: Calculated through image three-dimensional reconstruction or laser radar point cloud.

[0215] Crack length: calculated by image edge detection and geometric measurement.

[0216] 4. Result integration and report generation

[0217] Spatial alignment:

[0218] Align the defect location with the pipeline milepost number to generate a "defect-location" mapping table.

[0219] Correspondingly, the application also provides a pipeline detection system based on a UAV, as shown in Figure 2 which comprises,

[0220] The first module is configured to acquire pipeline information, pipeline location environment information, a pipeline detection task, and UAV-carried sensor information of a pipeline to be detected, set a collection section of the pipeline according to the pipeline information and the pipeline detection task, and formulate a first flight plan based on the pipeline location environment information and the collection section of the pipeline.

[0221] The second module is configured to make the UAV fly according to the first flight plan, collect first detection information of the pipeline, count a flight information deviation between a first actual flight and the first flight plan, evaluate the quality of the first detection information, and determine whether a second flight is needed based on the quality of the first detection information.

[0222] The third module is configured to optimize a second flight plan according to the flight information deviation if the second flight is needed, make the UAV fly according to the second flight plan, and collect second detection information of the pipeline.

[0223] The fourth module is configured to output a detection result of the pipeline by comprehensively integrating the first detection information and the second detection information of the pipeline.

[0224] Compared with the prior art, the application has the following beneficial effects:

[0225] 1. The collection section of the pipeline is set according to the pipeline information and the pipeline detection task of the pipeline to be detected, the reasonable UAV collection point is selected by considering the self conditions of the pipeline to be detected, and a reliable foundation is provided for the subsequent flight plan planning of the UAV and the pipeline detection. The first flight plan is formulated based on the pipeline location environment information and the collection section of the pipeline, the initial flight plan is formulated by considering the location environment conditions of the pipeline and the demand for flight stability, and the adaptability of the UAV flight is ensured.

[0226] 2. Based on the quality of the first detection information, it is judged whether the second flight is needed, and based on the quality of the detection information, it is judged whether the first flight is sufficient to support the quality requirement of pipeline detection. When it is difficult to support, the second flight is performed. According to the flight information deviation, the second flight plan is optimized, the loopholes of the first flight are supplemented, and the reliability and adaptation of the second flight are ensured. The detection result of the pipeline is output by comprehensively considering the first detection information and the second detection information of the pipeline, and the complementarity and intersection between the first detection information and the second detection information are considered to output the detection result of the pipeline, which improves the adaptability and accuracy of pipeline detection, ensures the adaptability of the unmanned aerial vehicle flight condition and the pipeline environment, and meets the detection requirements of the complex environment of the pipeline.

[0227] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by hardware, or by means of software and necessary general hardware platform. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.), and includes a plurality of instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments of the present application.

[0228] Those skilled in the art can understand that the drawings are only a schematic diagram of a preferred embodiment, and the modules or processes in the drawings are not necessarily required for implementing the present application.

[0229] Those skilled in the art can understand that the modules in the system in the embodiments can be distributed in the system in the embodiments according to the description of the embodiments, or can be changed and located in one or more systems different from the embodiments. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0230] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can make equivalent replacements or changes to the technical solutions and inventive concepts of the present application within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for pipeline inspection based on a UAV, characterized in that, The application relates to a pipeline detection method and device based on unmanned aerial vehicle (UAV). The application comprises the following steps: acquiring pipeline information, pipeline location environment information, a pipeline detection task and sensor information carried by an unmanned aerial vehicle, setting a collection section of the pipeline according to the pipeline information and the pipeline detection task of the pipeline to be detected, and formulating a first flight plan based on the pipeline location environment information and the collection section of the pipeline; the unmanned aerial vehicle flies according to the first flight plan, collects first detection information of the pipeline, counts flight information deviation between the first actual flight and the first flight plan, evaluates the quality of the first detection information, and judges whether the second flight is needed based on the quality of the first detection information; if the second flight is needed, the second flight plan is optimized according to the flight information deviation, the unmanned aerial vehicle flies according to the second flight plan, and the second detection information of the pipeline is collected; 2.The unmanned aerial vehicle based pipeline inspection method of claim 1, wherein, the first detection information and the second detection information of the pipeline are comprehensively output to obtain a detection result of the pipeline. The application comprises the following steps: the pipeline information comprises pipeline length, pipeline direction, pipeline material and pipeline conveying medium; pipeline simple scenes and pipeline complex scenes are defined according to the pipeline length and the pipeline direction, the whole pipeline to be detected is taken as the collection section of the pipeline to be detected for the pipeline simple scenes; 3.The UAV-based pipeline inspection method of claim 2, wherein, for the pipeline complex scenes, a pipeline model of the pipeline to be detected is established according to the pipeline length, the pipeline direction, the pipeline material and the pipeline conveying medium, all risk elements are collected through the pipeline detection task, high-risk positions are marked on corresponding positions of the pipeline model of the pipeline to be detected according to all the risk elements, and the high-risk positions are taken as the collection section of the pipeline to be detected. The pipeline simple scenes and the pipeline complex scenes are defined according to the pipeline length and the pipeline direction, which comprises the following steps: 4.The UAV-based pipeline inspection method of claim 1, wherein, a plurality of key features on the pipeline direction are extracted, the pipeline is divided into a plurality of sections according to the pipeline length, the key feature density of each section is counted, the average value of the key feature density is obtained, the scene complexity is defined based on the pipeline length, the plurality of key features and the average value of the key feature density, and the pipeline simple scenes and the pipeline complex scenes are distinguished through the scene complexity. The first flight plan is formulated based on the pipeline location environment information and the collection section of the pipeline, which comprises the following steps: the pipeline information and the pipeline location environment information are imported into GIS software, an obstacle distribution map is superimposed in the pipeline location environment, the collection section positions of the pipeline are marked in the GIS software, a plurality of collection section positions are taken as passing points, a plurality of candidate paths are generated through a path planning algorithm, and each candidate path passes through all the passing points; each candidate path is divided into a plurality of path units, the path cost evaluation and the flight risk cost evaluation are carried out on each candidate path based on the path units, the comprehensive cost is generated by combining the path cost and the flight risk cost, and the comprehensive cost change curve varying with the path units is drawn; the stable part curve and the fluctuation part curve are distinguished on the comprehensive cost change curve, the relative cost of the stable part curve and the fluctuation part curve is respectively calculated through the curve slope change, so that the relative comprehensive cost is determined, one candidate path is selected as a target path according to the relative comprehensive cost, and the flight information of the unmanned aerial vehicle is configured based on the target path. 5.The UAV-based pipeline inspection method of claim 4, wherein, On the basis of the target path, flight information of the unmanned aerial vehicle is configured, Comprising, The target path is divided into a flight path and a collection path according to the position of the collection section of the pipeline; For the flight path, the environment is analyzed for flight stability requirements by GIS software, and the flight information of the unmanned aerial vehicle is configured in combination with the flight task requirements; For the collection path, the pipeline is analyzed for flight stability requirements and the environment is analyzed for flight stability requirements by GIS software, and the flight information of the unmanned aerial vehicle is configured based on the pipeline flight stability requirements, the environment flight stability requirements and the flight task requirements. 6.The UAV-based pipeline inspection method of claim 5, wherein, The flight information deviation between the first actual flight and the first flight plan is counted, including, The flight information deviation of the flight path and the collection path is counted respectively to obtain the first deviation and the second deviation, and the change curves of the first deviation and the second deviation with the position are drawn, the deviation points of the first deviation and the second deviation are identified respectively, and the correlation between the two types of deviation points is analyzed. 7.The UAV-based pipeline inspection method of claim 6, wherein, According to the flight information deviation, the second flight plan is optimized, Including, The flight information deviation and the quality of the first detection information are spatially aligned by the position of the pipeline, and the second flight plan is optimized based on the deviation points and the correlation between the deviation points. 8.The UAV-based pipeline inspection method of claim 4, wherein, The detection results of the pipeline are output by integrating the first detection information and the second detection information of the pipeline, including, The intersection information part and the difference information part between the first detection information and the second detection information of the pipeline are determined according to the position of the pipeline; The intersection information part of the first detection information and the second detection information is fused to obtain fused information; The detection results of the pipeline are output based on the difference information part and the fused information.

9. A drone-based pipeline inspection system, comprising: Including, The first module is used for acquiring pipeline information, pipeline position environment information, pipeline detection task and unmanned aerial vehicle sensor information of the pipeline to be detected, setting the collection section of the pipeline according to the pipeline information and the pipeline detection task of the pipeline to be detected, and formulating the first flight plan based on the pipeline position environment information and the collection section of the pipeline; The second module is used for the unmanned aerial vehicle to fly according to the first flight plan, collect the first detection information of the pipeline, count the flight information deviation between the first actual flight and the first flight plan, evaluate the quality of the first detection information, and determine whether the second flight is needed based on the quality of the first detection information; The third module is used for optimizing the second flight plan according to the flight information deviation if the second flight is needed, and the unmanned aerial vehicle flies according to the second flight plan to collect the second detection information of the pipeline; The fourth module is used for outputting the detection results of the pipeline by integrating the first detection information and the second detection information of the pipeline.

Images