Method, device and equipment for determining speed of detector in pipeline

Abstract

The embodiment of the invention relates to the technical field of pipeline interior detection, in particular to a method, device and equipment for determining the speed of a detector in a pipeline, and the method comprises the steps: obtaining the first angle variation of an odometer wheel of the detector at the current moment; the first angle variable quantity comprises a first angle variable quantity of the odometer wheel at the current moment; calculating a first speed value of the odometer wheel at the current moment according to the first angle variation; acquiring a second angle variation of the odometer wheel support arm at the current moment and first acceleration data of the inner detector at the current moment; judging whether the first speed value is abnormal or not according to the first angle variation, the second angle variation and the first acceleration data; and if the first speed value is abnormal, correcting the first speed value by using the first acceleration data to obtain a second speed value of the inner detector at the current moment.

Description

Technical Field

[0001] The embodiments of this specification relate to the technical field of in-pipeline detection, and more particularly to a method, device, and apparatus for determining the velocity of an in-pipeline detector. Background Art

[0002] Oil and gas pipelines serve as the "main arteries" of energy transmission. Their main transport media include crude oil, refined oil, natural gas and other special fluids with flammable, explosive and high-pressure characteristics. In-pipeline detection is one of the important technologies to ensure the safe operation of pipelines. A stable operating speed of the internal detector is an important condition for ensuring detection accuracy and the validity of detection data. In order to achieve stable control of the operating speed of the in-pipeline detector, it is necessary to ensure accurate perception of the operating speed of the in-pipeline detector. The speed of the in-pipeline detector can be calculated by the angle difference generated by the odometer wheel within a certain period of time. However, when the in-pipeline detector is running, the odometer wheel passes through welds, depressions or slips, resulting in errors in the operating speed data calculated based on the rotation angle of the rotor. It is urgent to use a correction algorithm to correct the erroneous speed of the odometer wheel to improve the accuracy of the speed calculated by the odometer wheel.

[0003] Existing methods for determining the speed of in-pipeline detectors generally adopt the following two technical routes: one is the optimization method of three-way odometer wheel signals, which selects the "optimal" speed value from the three signals through simple logic. Although this method can suppress some anomalies, it cannot correct systematic deviations (such as synchronous slippage of multiple odometer wheels), and it relies on manually set thresholds, which has poor adaptability. The other is the fusion of multiple sensor data, which fuses auxiliary data such as acceleration and gyroscopes with odometer wheel signals according to fixed weights. Although this method improves fault tolerance, it only reduces the impact of abnormal signals through weighting, and does not directly correct the original odometer wheel data, resulting in residual errors and poor reliability.

[0004] Therefore, how to overcome the problems of low accuracy, poor adaptability and reliability in the existing methods for determining the velocity of in-pipe detectors and propose a method for determining the velocity of in-pipe detectors with high accuracy, adaptability and reliability is a key issue that needs to be solved urgently. Summary of the Invention

[0005] The purpose of the embodiments of this specification is to provide a method, device and equipment for determining the velocity of a detector in a pipeline, so as to overcome the problems of low accuracy, poor adaptability and reliability existing in the existing methods for determining the velocity of a detector in a pipeline.

[0006] On the one hand, an embodiment of the present specification provides a method for determining the speed of a detector in a pipeline, including: obtaining a first angle change of the odometer wheel of the internal detector at the current moment; calculating a first speed value of the odometer wheel at the current moment based on the first angle change; collecting a second angle change of the odometer wheel support arm at the current moment and first acceleration data of the internal detector at the current moment; judging whether the first speed value is abnormal based on the first angle change, the second angle change and the first acceleration data; if the first speed value is abnormal, correcting the first speed value using the first acceleration data to obtain the second speed value of the internal detector at the current moment.

[0007] On the other hand, an embodiment of the present specification provides a device for determining the speed of a detector in a pipeline, including: an acquisition module for acquiring a first angle change of the odometer wheel of the internal detector at the current moment; a calculation module for calculating a first speed value of the odometer wheel at the current moment based on the first angle change; an acquisition module for acquiring a second angle change of the odometer wheel support arm at the current moment and first acceleration data of the internal detector at the current moment; a judgment module for judging whether the first speed value is abnormal based on the first angle change, the second angle change and the first acceleration data; and a correction module for correcting the first speed value using the first acceleration data if the first speed value is abnormal, to obtain the second speed value of the internal detector at the current moment.

[0008] In another aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-mentioned method for determining the velocity of the in-pipeline detector.

[0009] As can be seen from the technical solutions provided in the embodiments of this specification, the embodiments of this specification can obtain the first angle change of the odometer wheel of the internal detector at the current moment; calculate the first velocity value of the odometer wheel at the current moment based on the first angle change; collect the second angle change of the odometer wheel support arm at the current moment and the first acceleration data of the internal detector at the current moment; determine whether the first velocity value is abnormal based on the first angle change, the second angle change, and the first acceleration data; and if the first velocity value is abnormal, correct the first velocity value using the first acceleration data to obtain the second velocity value of the internal detector at the current moment. Compared to existing methods, the embodiments of this specification can determine whether the odometer wheel has passed through a pipeline defect, slipped, or idled based on the first angle change of the odometer wheel at the current moment, the second angle change of the odometer wheel support arm, and the first acceleration data of the internal detector. Furthermore, the first acceleration data of the internal detector can be used to correct the abnormal first velocity value of the odometer wheel in these situations to obtain the second velocity value of the internal detector, significantly improving the accuracy and reliability of the internal detector's velocity measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art.

[0011] Figure 1 This is a flow chart of a method for determining the velocity of a detector in a pipeline provided by an embodiment of this specification;

[0012] Figure 2 It is a flow chart of a method for determining the speed of a pipeline detector when the pipeline detector includes a three-way odometer provided in an embodiment of this specification;

[0013] Figure 3 This is a schematic diagram of the workflow of the system for determining the speed of a pipeline detector when the pipeline detector provided by an embodiment of this specification includes a three-way odometer;

[0014] Figure 4 This is a flow chart of a method for determining whether a first speed value of an odometer wheel is abnormal, provided in an embodiment of this specification;

[0015] Figure 5 This is an overall logic flow chart of a method for determining whether the first speed value of an odometer wheel is abnormal or not provided in an embodiment of this specification;

[0016] Figure 6 This is a flow chart of a method for fusing odometer wheel speeds provided in an embodiment of this specification;

[0017] Figure 7 This is a schematic diagram of the structure of a device for determining the velocity of a detector in a pipeline provided in an embodiment of this specification;

[0018] Figure 8 It is a schematic diagram of the structural composition of the computer device provided in the embodiment of this specification. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this specification.

[0020] In some embodiments, the in-pipe detector can be a special device for detecting the internal condition of the pipeline, which can perform comprehensive detection of defects such as corrosion, cracks, deformation, and sediment in the pipeline. By equipping detection units such as sensors and cameras, and combining them with a data processing system, the in-pipe detector can achieve real-time monitoring and evaluation of the health status of the pipeline. The in-pipe detector can move in the pipeline by relying on the pressure difference of the medium in the pipeline or its own power device (such as a motor drive). During the movement, the detection unit continuously collects data such as images of the inner wall of the pipeline, ultrasonic signals, electromagnetic signals, etc., and records the detection position through positioning devices such as encoders and odometers, and finally generates a positioning report of the pipeline defect.

[0021] In some implementations, the in-pipe detector may be provided with one or more sensors, including an acceleration sensor, a first angle sensor, a second angle sensor, and a third angle sensor. The acceleration sensor may be used to obtain first acceleration data of the in-pipe detector at the current moment and second acceleration data at the previous moment. The first angle sensor may be used to obtain a first angle change of the odometer wheel at the current moment and a fourth angle change at the previous moment. The second angle sensor may be used to obtain a second angle change of the odometer wheel support arm at the current moment. The third angle sensor may be used to obtain a third angle value between the odometer wheel support arm and the vertical line at the previous moment.

[0022] In some embodiments, the current moment may be a moment that is separated from the previous moment by a preset fixed time interval. For example, the preset fixed time interval may be 0.05 seconds. If the previous moment is 5.05 seconds, the current moment is 5.10 seconds.

[0023] In some embodiments, an odometer wheel can be the core positioning component of an in-pipe detector. By contacting the inner wall of the pipe and rotating with the movement of the detector, it converts the detector's axial displacement into rotational motion, thereby accurately measuring the detected position. The odometer wheel can consist of a metal hub, a rubber tire, bearings, and an encoder. The hub's outer diameter is designed to match the pipe's inner diameter to ensure close contact with the pipe wall; the rubber tire provides friction and adapts to pipe bends; the bearings reduce rotational resistance; and the encoder converts the rotation angle into an electrical signal, achieving odometer measurement through pulse counting. Technical parameters for the odometer wheel may include measurement accuracy, applicable pipe diameter, and operating temperature. Measurement accuracy is affected by wheel diameter, encoder resolution, and installation accuracy, with a typical accuracy of ±0.1% FS (full scale). Applicable pipe diameters range from DN50 to DN1200 (nominal diameters 50 mm to 1200 mm). The operating temperature range is -20°C to 80°C to accommodate diverse media environments.

[0024] In some embodiments, the odometer wheel support arm can be a mechanical structure that connects the odometer wheel to the detector body in the pipeline. It is responsible for pressing the odometer wheel against the inner wall of the pipeline and providing the necessary support and adjustment functions to ensure the stable operation of the odometer wheel. The support arm can be composed of components such as springs, connecting rods, hinges, etc., and the spring preload force is used to keep the odometer wheel close to the pipe wall. When the inner diameter of the pipeline changes, the support arm can adaptively adjust the position of the odometer wheel to avoid slipping or overloading due to excessive clearance. Some support arms are designed with damping mechanisms to reduce the impact of vibration on measurement accuracy. Specifically, the odometer wheel support arm can adopt a spring-connecting rod mechanism to achieve dynamic contact between the odometer wheel and the pipe wall. When encountering an obstacle, the odometer wheel support arm can automatically retract to prevent damage to the odometer wheel.

[0025] In some embodiments, as the in-pipe detector moves within the pipeline, the odometry wheel is pressed against the pipe wall via a support arm, converting axial displacement into rotational motion. An encoder records the odometry wheel's rotation angle in real time and transmits this information to the in-pipe detector control system via a pulse signal. The control system combines this odometry data with information collected by the detection unit to generate a precise location report for the pipeline defect. The in-pipe detector, odometry wheel, and odometry wheel support arm together form the core system for pipeline defect locating. By optimizing the odometry wheel's mechanical structure and encoder technology, combined with the support arm's adaptive adjustment capabilities, the accuracy and reliability of in-pipeline detection can be significantly improved, providing technical support for safe pipeline operation.

[0026] Figure 1 This is a flow chart of a method for determining the velocity of a detector in a pipeline provided in an embodiment of this specification. Figure 2 This is a flow chart of a method for determining the speed of a pipeline detector when the pipeline detector includes a three-way odometer provided in an embodiment of this specification. Figure 3 This is a schematic diagram of the workflow of a system for determining the speed of a pipeline detector when the pipeline detector includes a three-way odometer provided in an embodiment of this specification. When implemented, the system includes the following steps:

[0027] S101: Obtain a first angle change of the odometer wheel of the inner detector at the current moment.

[0028] In some embodiments, a first angle change of the odometer wheel of the inner detector at the current moment may be obtained.

[0029] By acquiring the first angle change of the odometer wheel of the inner detector at the current moment, a data foundation is laid for calculating the first speed value of the odometer wheel at the current moment.

[0030] The first angle change can be the angular change of the odometry wheel from the previous moment to the current moment. The odometry wheel can be equipped with a high-precision encoder (such as a photoelectric encoder or a magnetic encoder) to measure the odometry wheel's rotation angle in real time. The encoder outputs a pulse signal for each rotation of a certain angle, with the number of pulses proportional to the rotation angle. A data acquisition card or embedded system (such as a microcontroller) can be used to acquire the encoder's pulse signals and convert them into the first angle change. Specifically, the encoder pulse signals can be counted to record the number of pulses per unit time. The pulse count is converted into the first angle change based on the encoder's resolution (i.e., the number of pulses per revolution). For example, if the encoder outputs 1000 pulses per revolution, each pulse corresponds to a rotation angle of 0.36°. At the current time T, the odometry wheel's current angle θ(T) is recorded. This is compared with the angle θ(T-1) at the previous time T-1 to calculate the first angle change: Δθ = θ(T) - θ(T-1).

[0031] S102: Calculate a first speed value of the odometer wheel at the current moment according to the first angle change.

[0032] In some embodiments, based on the first angle change, the following formula can be used to calculate the first speed value of the odometer at the current moment:

[0033]

[0034] Where V1 represents the first speed value of the odometer wheel in m / s, Dq represents the angle rotated by the odometer wheel in time Dt, that is, the first angle change, D represents the diameter of the odometer wheel in mm, and Dt represents the time interval between the current moment and the previous moment.

[0035] The first angle change can accurately represent the angle change value of the odometer wheel in a normal state. The angle change of the odometer wheel in a normal state can be accurately converted into the linear velocity of the odometer wheel through the above formula.

[0036] By measuring the angle Δθ that the odometer wheel rotates over a period of time Δt, the angular velocity ω can be calculated based on the relationship between the rotation angle and time: ω = Δθ / Δt. Furthermore, considering the relationship between the linear velocity v, the angular velocity ω, and the radius r (i.e., the radius of the odometer wheel) as v = ω × r, and considering the odometer wheel's circumference C = 2πr, the radius r = C / (2π). Substituting r into the linear velocity formula yields v = (Δθ / Δt) × (C / (2π)). Therefore, for a normal odometer wheel, by measuring the rotation angle Δθ over a period of time Δt and the known odometer wheel circumference C, the linear velocity v of the odometer wheel can be accurately calculated.

[0037] S103: Collect the second angle change of the odometer support arm at the current moment and the first acceleration data of the internal detector at the current moment.

[0038] In some embodiments, the second angle change of the odometer support arm at the current moment and the first acceleration data of the internal detector at the current moment may be collected.

[0039] Abnormal conditions such as changes in the pressure difference before and after the internal detector or collision with defects will cause changes in the acceleration of the entire internal detector. Introducing the acceleration signal into the speed correction of the mileage wheel can greatly improve the reliability of the mileage wheel speed.

[0040] The second angle change can be the angle change of the odometer arm from the previous moment to the current moment. The rotation angle of the odometer arm, or the second angle change, can be measured using an angle sensor integrated into the joint of the odometer arm. When the odometer or internal detector moves within the pipeline and strikes a defect, the arm comes into contact with the inner wall of the pipeline. At this point, the angle sensor, using Hall effect or encoder technology (such as a photoelectric encoder or magnetic encoder), provides real-time feedback on the contact angle between the arm and the inner wall, thereby determining the second angle change. Specifically, the encoder outputs a pulse signal for each rotation of a certain angle, with the number of pulses proportional to the rotation angle. A data acquisition card or embedded system (such as a microcontroller) can be used to acquire the encoder pulse signal. The pulse number can be converted into the second angle change based on the encoder's resolution (i.e., the number of pulses per revolution). This design enables the system to more accurately calculate the linear velocity or position of the odometer, improving the accuracy and reliability of internal detection.

[0041] The first acceleration data may include the acceleration value of the internal detector at the current moment. The first acceleration data can be obtained by an acceleration sensor provided in the main body of the detector in the pipeline. Specifically, the acceleration sensor can be installed in the core part of the main body of the detector in the pipeline, such as between the drive module and the battery compartment, to ensure that it is synchronized with the axial movement of the entire machine. A flexible circuit board (FPC) can be used to connect the acceleration sensor and the main control board to reduce the stress transfer caused by the rigid connection. The acceleration sensor can measure the axial acceleration of the entire detector in the pipeline. This value reflects the dynamic acceleration change of the detector along the axial direction during the movement of the detector in the pipeline, including positive acceleration (such as propeller start-up), negative acceleration (such as braking) and instantaneous acceleration fluctuations caused by vibration. By monitoring the first acceleration data, an accurate description of the motion state of the detector can be achieved, providing basic data for subsequent speed correction and fault diagnosis.

[0042] S104: Determine whether the first velocity value is abnormal based on the first angle change, the second angle change, and the first acceleration data.

[0043] When the odometer wheel passes through a pipeline defect or slips or idles, the first speed value calculated using the first angle change will no longer be accurate, that is, the first speed value is abnormal. Figure 4 This is a flow chart of a method for determining whether the first speed value of an odometer wheel is abnormal or not, provided in this specification. Figure 5 This is a logic flow chart of a method for determining whether the first speed value of an odometer wheel is abnormal. The method includes the following steps when implemented:

[0044] S1041: Obtain a third angle value between the odometer arm and the vertical line at the previous moment; calculate a first height value of the pipeline defect contour that the odometer passes through at the current moment based on the first angle change and the third angle value; if the first height value is greater than a preset height threshold, determine that the first speed value is abnormal.

[0045] In some embodiments, a third angle value between the odometer arm and the vertical line at the previous moment can be obtained; based on the first angle change and the third angle value, a first height value of the pipeline defect contour passed by the odometer at the current moment can be calculated; if the first height value is greater than a preset height threshold, it can be determined that the first speed value is abnormal.

[0046] By comparing the first height value with a preset height threshold, it can be determined whether there has been a sudden change in the odometer arm angle between the previous moment and the current moment, that is, whether the vehicle has passed through a pipeline defect at the current moment. If so, the odometer wheel's first speed value at the current moment can be corrected, helping to improve the reliability of the odometer wheel's speed measurement.

[0047] The third angle value may include the angle value between the odometer wheel support arm and the vertical line at the last moment. The angle value between the odometer wheel support arm and the vertical line can be measured by an angle sensor preset at the odometer wheel support arm, which will not be described in detail here.

[0048] A relationship between the first height value and the first angle change of the pipeline defect profile at each moment the odometer wheel passes can be pre-established. This relationship can be based on deep neural networks, machine learning models, or empirical formulas. The first angle change at the current moment can be substituted into this relationship to obtain the first height value of the pipeline defect profile at the current moment the odometer wheel passes.

[0049] At the current moment T, if the odometer wheel passes through a pipeline defect, the obtained first angle variation is a sudden change, so the first speed value calculated based on the first angle variation will no longer be accurate.

[0050] The first height value can be used to determine whether the odometry wheel has passed through a pipeline defect at the current moment. Specifically, if the odometry wheel has passed through a pipeline defect, the first height value will undergo a sudden change. A height threshold (e.g., 3 mm) can be set. If the first profile height value of an odometry wheel in the pipeline detector at the current moment exceeds the preset height threshold, it can be determined that the odometry wheel is affected by the pipeline defect at the current moment. In other words, the first velocity value of the odometry wheel, calculated from the first angle change of the odometry wheel obtained by the angle sensor, is abnormal.

[0051] S1042: If the first height value is less than or equal to a preset height threshold, obtain a second height value of the pipeline defect contour that the odometer passed at the previous moment; if the second height value is greater than the preset height threshold, determine that the first speed value is abnormal.

[0052] In some embodiments, if the first height value is less than or equal to a preset height threshold, a second height value of the pipeline defect contour passed by the odometer at the previous moment can be obtained; if the second height value is greater than the preset height threshold, it can be determined that the first speed value is abnormal.

[0053] By comparing the second height value with a preset height threshold, it can be determined whether a sudden change in the odometer arm angle has occurred between the previous moment and the current moment, indicating whether the vehicle passed through a pipeline defect. If so, the odometer wheel's first speed value at the current moment can be corrected, helping to improve the reliability of the odometer wheel's speed measurement.

[0054] At the current time T, if the first altitude value is less than or equal to the preset altitude threshold, the first speed value cannot be determined to be normal. Although the odometer wheel did not pass through the pipeline defect at the current time T, the obtained first angle change may still have undergone a sudden change. Specifically, at the previous time T-1, if the odometer wheel passed through the pipeline defect, then the first angle change would also have undergone a sudden change between the previous time T-1 and the current time T. In this case, the first speed value calculated based on the first angle change would no longer be accurate.

[0055] Therefore, if the first height value is less than or equal to a preset height threshold, a second height value of the pipeline defect outline that the odometer wheel passed at the previous moment can be obtained. If the second height value is greater than a preset height threshold (e.g., 3 mm), it can be determined that a sudden change in the first angle occurred between the previous moment and the current moment. In other words, the first speed value of the odometer wheel, calculated from the first angle change of the odometer wheel obtained by the angle sensor, is abnormal.

[0056] S1043: If the second altitude value is less than or equal to a preset altitude threshold, obtain the second acceleration data of the internal detector at the previous moment and the fourth angle change of the odometer at the previous moment; if the odometer meets the preset judgment conditions, determine that the first velocity value is abnormal; the preset judgment conditions include that the first acceleration data is greater than the second acceleration data and the first angle change is less than the fourth angle change, or that the first acceleration data is zero and the absolute difference between the first angle change and the fourth angle change is greater than a preset deviation threshold.

[0057] In some embodiments, if the second height value is less than or equal to a preset height threshold, the second acceleration data of the internal detector at the previous moment and the fourth angle change of the odometer at the previous moment can be obtained; if the odometer meets the preset judgment conditions, the first speed value can be judged to be abnormal; the preset judgment conditions may include that the first acceleration data is greater than the second acceleration data and the first angle change is less than the fourth angle change, or that the first acceleration data is zero and the absolute difference between the first angle change and the fourth angle change is greater than a preset deviation threshold.

[0058] By using pre-set judgment conditions, you can quickly and accurately determine whether the odometer wheel is slipping or spinning at the current moment. If slipping or spinning occurs, the odometer wheel's first speed value at that moment can be corrected, helping to improve the reliability of the odometer wheel speed measurement.

[0059] The second acceleration data may include the acceleration value of the internal detector at the previous moment, and the second acceleration data may be obtained by an acceleration sensor disposed in the detector body in the pipeline, which will not be described in detail here.

[0060] The fourth angle change may include an angle change value of the odometer from the moment before the last moment to the last moment. The third angle change may be obtained by an angle sensor provided on the odometer, which will not be described in detail here.

[0061] If both the first and second altitude values ​​are less than or equal to the preset altitude threshold, it indicates that no sudden change in the first angle change occurred between the previous time T-1 and the current time T. However, the first speed value cannot be determined to be normal, as the odometer wheel may have slipped or spun during the time T-1. Odometer wheel slip can be manifested as a continuous increase in the overall acceleration of the internal detector and a continuous decrease in the odometer wheel angle change. Specifically, the overall acceleration of the internal detector at the current time T is greater than the overall acceleration of the internal detector at the previous time T-1, and the first odometer wheel angle change at the current time T is less than the third odometer wheel angle change at the previous time T-1. Odometer wheel spun can be manifested as the overall acceleration of the internal detector becoming zero and a sudden change in the odometer wheel angle change. Specifically, the overall acceleration of the internal detector at the current time T is zero, and the absolute difference between the first odometer wheel angle change at the current time T and the third odometer wheel angle change at the previous time T-1 is greater than a preset deviation threshold.

[0062] Therefore, a pre-set condition for determining slip / spin can be established: the first acceleration data is greater than the second acceleration data and the first angle change is less than the fourth angle change, or the first acceleration data is zero and the absolute difference between the first and fourth angle changes is greater than a preset deviation threshold. If the odometer wheel meets this condition, it indicates that the odometer wheel is slipping or spinning. In this case, the first angle change of the odometer wheel obtained by the angle sensor is abnormal, and the first speed value of the odometer wheel calculated from the first angle change is also abnormal.

[0063] S1044: If the odometer wheel does not meet the preset determination condition, it is determined that the first speed value is normal.

[0064] In some embodiments, if the odometer wheel does not meet the preset determination criteria, the first speed value is determined to be normal. If both the first altitude value and the second altitude value are less than or equal to the preset altitude threshold, this indicates that no sudden change in the first angle change occurred between the previous time T-1 and the current time T. Furthermore, if the odometer wheel does not meet the preset determination criteria, this indicates that the odometer wheel has not slipped or spun at the current time T, the first angle change is normal, and therefore the first speed value of the odometer wheel calculated from the first angle change is also normal.

[0065] For each odometer wheel in the detector at the current moment, whether it has passed a defect, whether it is slipping, and whether it is idling can be determined independently in sequence, thereby determining whether the first speed value corresponding to each odometer wheel is abnormal.

[0066] S105: If the first velocity value is abnormal, the first velocity value is corrected using the first acceleration data to obtain a second velocity value of the internal detector at the current moment.

[0067] In some embodiments, a third speed value of the internal detector at the previous moment can be obtained; based on the third speed value and the first acceleration data, the first speed value can be corrected to obtain a fourth speed value of the odometer at the current moment; based on the fourth speed value and multiple preset support thresholds, a support matrix of the odometer can be constructed; based on the maximum modulus eigenvalue and corresponding eigenvector of the support matrix, the weight value of the odometer can be determined; based on the fourth speed value and weight value of the odometer, the second speed value of the internal detector can be calculated.

[0068] Based on the third velocity value of the internal detector at the previous moment and the first acceleration data of the internal detector at the current moment, the abnormal first velocity value of the odometry wheel can be corrected, improving the accuracy of the odometry wheel speed. Furthermore, based on the corrected odometry wheel speed values, a support matrix between multiple odometry wheels can be constructed using multiple preset support thresholds. This allows the selection of odometry wheels with greater similarity and closer speed to the internal detector. By increasing the weight assigned to the speed data of these odometry wheels, their weight in the final fused speed is increased, resulting in a more accurate final speed.

[0069] After determining that a particular odometry wheel in the pipeline detector has passed through a pipeline defect, slipped, or idled, the velocity of the odometry wheel can be calculated using the acceleration of the entire pipeline detector and the velocity value of the pipeline detector at the previous moment. Specifically, the time interval between the current moment and the previous moment can be obtained and multiplied by the first acceleration data to obtain the velocity change value. The velocity change value is added to the velocity value of the pipeline detector at the previous moment to obtain the fourth velocity value of the odometry wheel at the current moment, i.e., V t =V0+aDt, where V t represents the fourth velocity value, V0 represents the third velocity value, a represents the first acceleration data, and Dt represents the time interval between the current moment and the previous moment. If an odometer wheel in the pipeline detector has neither passed through a pipeline defect nor experienced slippage or idling, that is, the first velocity value of the odometer wheel is normal, then the first velocity value of the odometer wheel can be directly used as the fourth velocity value.

[0070] For each odometry wheel in the current internal detector, the similarity between the fourth velocity values ​​of each odometry wheel can be calculated. Similarity can be calculated using methods such as absolute error or absolute error change rate. Based on the similarity between the fourth velocity values ​​of each odometry wheel, these similarities can be further processed using a preset support threshold to obtain the support between the fourth velocity values ​​of each odometry wheel. Based on the support between the fourth velocity values ​​of each odometry wheel, a support matrix for multiple odometry wheels can be constructed. The maximum modulus eigenvalue and corresponding eigenvector of this support matrix can be calculated, and the eigenvector corresponding to this maximum modulus eigenvalue can be normalized to determine the weight value of each odometry wheel in the internal detector. The weighted average of the fourth velocity values ​​and weight values ​​of the multiple odometry wheels can be calculated to obtain the second velocity value of the internal detector.

[0071] Figure 6 This is a flow chart of a method for fusing odometer wheel speeds provided in an embodiment of this specification. When implemented, the method includes the following steps:

[0072] S1051: Obtain the third speed value of the internal detector at the previous moment.

[0073] In some embodiments, based on an extended Kalman filter algorithm, historical speed values ​​of the inner detector are analyzed to obtain a third speed value of the inner detector at a previous moment.

[0074] When analyzing historical velocity data from the internal detector, the extended Kalman filter algorithm processes each velocity observation point in chronological order, continuously revising the state estimate and reducing the error covariance. This process effectively suppresses the effects of measurement and process noise, providing a smoother and more accurate velocity sequence than the raw data. Ultimately, the algorithm outputs the third velocity value at the previous moment based not only on direct observations at that moment but also on the dynamic trends of historical data, resulting in greater reliability and robustness.

[0075] Based on the extended Kalman filter algorithm, a kinematic model of the internal detector can be established. The state-space equation can be used to describe the change law of the position, velocity and other state quantities of the internal detector. Taking into account that the actual pipeline system often has nonlinear characteristics (such as acceleration changes, external disturbances, etc.), the extended Kalman filter algorithm can be used to locally linearize the nonlinear kinematic model (i.e. calculate the Jacobian matrix), so as to estimate the state and covariance at the current moment in the prediction step. In the update step, the extended Kalman filter algorithm can be used to weightedly fuse the latest observed velocity data with the predicted value to obtain the third velocity value of the internal detector at the previous moment. By dynamically adjusting the confidence of the prediction and observation through the Kalman gain, the estimated third velocity value can be closer to the true value.

[0076] S1052: Correcting the first speed value according to the third speed value and the first acceleration data to obtain a fourth speed value of the odometer at the current moment;

[0077] In some embodiments, the first acceleration data is integrated using an inertial navigation system algorithm to obtain a speed trend value at the current moment; based on the Markov random field, the speed trend value is corrected using historical speed trend values ​​to obtain a fourth speed value of the odometer at the current moment.

[0078] The introduction of Markov random fields enables the system to flexibly model motion correlations across different timescales, adapting to a variety of dynamic scenarios. Furthermore, without the need for additional external observation equipment, effective error suppression can be achieved solely by relying on historical data. This Markov random field-based correction method is more suitable for handling non-Gaussian noise and nonlinear motion patterns, and can better maintain the continuity and consistency of velocity estimates.

[0079] An inertial navigation system algorithm can be used to preprocess the third velocity value and the first acceleration data, including eliminating system errors such as sensor bias and temperature drift. Numerical integration algorithms such as the trapezoidal method or the Runge-Kutta method can be used for integration to reduce the computational errors introduced by the discretized velocity and acceleration values. Furthermore, the accumulated errors caused by accelerometer noise and the integration process can be estimated and compensated in real time. However, due to the inherent characteristics of the inertial navigation system algorithm, the velocity trend value will still gradually deviate from the true value over time. Therefore, a Markov random field model can be introduced to address this issue. The Markov random field model constructs velocity trend values ​​at multiple consecutive moments into a graph structure with Markov properties, where each node represents the velocity state at a moment, and the edges represent the spatiotemporal constraints between states. By defining the velocity trend function based on a preset sliding window, the Markov random field model can capture the smoothness and continuity of velocity changes, transforming physical laws into probabilistic constraints.

[0080] Specifically, a joint probability distribution model can be established that incorporates the current velocity trend value and historical velocity trend values. This model considers both the observation likelihood of the inertial navigation integration results and the prior motion laws reflected by historical data. Using inference algorithms such as maximum a posteriori probability estimation or belief propagation, the optimal velocity state estimate under this probability model is obtained. This correction method essentially exploits the spatiotemporal correlations of the velocity sequence and uses historical information to constrain the current estimate, thereby effectively suppressing the divergence of inertial navigation errors. In particular, when the odometer state undergoes a sudden change, the Markov random field model can adjust the weights of different time nodes to maintain responsiveness while avoiding abnormal jumps. This ensures that the corrected fourth velocity value both promptly reflects the actual motion changes and maintains a reasonably smooth transition. Taking the fourth velocity value estimate of an abnormal odometer as an example, the third velocity value and first acceleration data can be preprocessed based on inertial navigation. Then, a fourth-order Runge-Kutta method can be used for numerical integration to obtain the velocity trend value sequence. At the same time, a sliding window containing the speed trend values ​​of 30 consecutive moments before the previous moment can be maintained to construct a spatiotemporal Markov random field model, and probabilistic reasoning can be performed through a parallelized belief propagation algorithm to output the corrected fourth speed value.

[0081] S1053: Constructing a support matrix of the odometer according to the fourth speed value and a plurality of preset support thresholds.

[0082] In some embodiments, based on the fourth speed value and a plurality of preset support thresholds, the following formula may be used to construct a support matrix of the odometry wheel:

[0083]

[0084] Where r ij is the element in the i-th row and j-th column of the support matrix, which represents the support of the j-th mileage round of the inner detector at the current moment for the i-th mileage round, j∈[1,N], i∈[1,N], N represents the total number of mileage rounds in the inner detector, N33, v j Indicates the corrected speed value of the jth odometer at the current moment, v i represents the corrected speed value of the ith odometer at the current moment, d1, d2 and d3 represent the preset first support threshold, second support threshold and third support threshold respectively and satisfy 0￡d1 <d3<d2￡1。

[0085] By using multiple preset support thresholds to construct a support matrix between multiple odometry wheels, we can select odometry wheels with greater similarity and closer to the inner detector speed. By increasing the weight assigned to the speed data of these odometry wheels, their weight in the final fused speed increases, making the final inner detection second speed value more accurate.

[0086] The support of each odometry wheel for any odometry wheel can be determined based on the preset first, second, and third support thresholds (e.g., 0, 0.3, and 0.15, respectively). Specifically, the absolute difference between the fourth speed value of each odometry wheel and the fourth speed value of any odometry wheel can be calculated. If the absolute difference is less than or equal to the preset first support threshold, the support is 1. If the absolute difference is greater than the preset first support threshold and less than or equal to the preset third support threshold, the support range is [0.5, 1]. If the absolute difference is greater than the preset third support threshold and less than or equal to the preset second support threshold, the support range is (0, 0.5). If the absolute difference is greater than or equal to the preset second support threshold, the support is 0.

[0087] Based on the processed pairwise support data, a complete support matrix can be constructed. This N×N symmetric matrix has clear mathematical properties: the diagonal elements are always 1, indicating the full support of each mileage wheel with itself; the non-diagonal elements store the support values ​​between the corresponding wheel pairs. Sparse matrix storage technology can also be used to optimize computational efficiency, especially when the number of mileage wheels is large. The support matrix not only reflects the current operating consistency of the system, but its eigenvalues ​​and eigenvectors also contain deep system status information. By performing spectral clustering analysis on the matrix, the subset of mileage wheels that may have faults can be identified; and the change in the rank of the matrix can reflect the overall coordination level of the system. This matrix representation method provides a powerful mathematical tool for subsequent fault diagnosis and predictive maintenance.

[0088] In some embodiments, multiple support thresholds may be dynamically adjusted based on a hierarchical adaptive strategy.

[0089] After obtaining the mileage wheel similarity data (such as absolute difference), a preset support threshold can be introduced for data conversion. A multi-level support threshold system can be designed taking into account the system fault tolerance requirements and historical operation data. For example, the primary threshold is used to distinguish between complete matches and approximate matches, the intermediate threshold is used to identify short-term suspicious deviations, and the advanced threshold corresponds to long-term trend adjustments. Each similarity value will be mapped to a corresponding support value according to its threshold interval. This mapping process can use linear interpolation or use nonlinear functions such as Sigmoid to achieve a smooth transition. Through this processing, the original numerical difference is converted into a more explanatory support indicator, and its value range is usually standardized between 0 and 1, with 1 indicating full support and 0 indicating no support at all.

[0090] Specifically, a hierarchical adaptive strategy can be used to dynamically maintain these support thresholds to improve environmental adaptability in different internal detectors. The hierarchical adaptive strategy can include three control levels: the base layer maintains the default threshold settings, the middle layer makes small adjustments based on short-term data fluctuations, and the top layer implements large-scale corrections based on long-term operating trends. The adaptive process can be achieved by monitoring the support distribution characteristics: when the system detects that the support is generally high, the threshold can be appropriately tightened to increase sensitivity; when there is a general decrease in support, the threshold can be relaxed to prevent false alarms. This dynamic adjustment can be achieved through a PID control algorithm or an intelligent adjustment method based on reinforcement learning, which will not be discussed here.

[0091] S1054: Determine the weight value of the mileage wheel according to the maximum modulus eigenvalue and the corresponding eigenvector of the support matrix.

[0092] In some embodiments, the initial weight value of the mileage wheel is determined based on the maximum modulus eigenvalue and the corresponding eigenvector of the support matrix; the initial weight value is fine-tuned according to a preset first sliding window; the support matrix and the fine-tuned initial weight value can be input into a preset second sliding window to obtain the weight value of the mileage wheel; the length of the first sliding window is less than the length of the second sliding window.

[0093] The support matrix of the odometry wheel can be viewed as an undirected weighted graph, where each odometry wheel represents a node, and the edge weights between nodes are determined by the corresponding values ​​in the support matrix. Based on this graph structure, different centrality indices can be calculated, including degree centrality, closeness centrality, and eigenvector centrality, to measure the importance of each odometry wheel in the overall internal detector system. Specifically, the weighted degree of each node (i.e., the sum of the support of all edges connected to the node) can be calculated. The higher the support of the node, the greater its degree centrality, indicating that the data of the odometry wheel is more reliable. The average "distance" of a node from other nodes (i.e., the inverse of the support) can be measured. A node with high centrality means that its speed value is more consistent with other nodes and may be more reliable.

[0094] Based on matrix eigendecomposition, the maximum modulus eigenvalue and corresponding eigenvector of the support matrix are determined. The eigenvector corresponding to this maximum modulus eigenvalue is normalized to assign node weights, giving nodes with high support connections higher scores. These centrality metrics can provide a preliminary estimate of the relative credibility of each milepost. However, since the support matrix may be affected by noise or short-term fluctuations, relying solely on graph theory methods may not be adaptable to complex and dynamic environments. Therefore, deep neural networks can be combined for more refined weight optimization.

[0095] Since the operating status of internal detectors may change over time (e.g., due to wheelset wear, changes in pipeline conditions, etc.), weight distribution that relies on static calculations cannot adapt to complex pipeline environments. A sliding window + adaptive filtering method can be used to fine-tune the initial weight values ​​calculated from degree centrality, proximity centrality, or eigenvector centrality for different mileage wheels. For example, within a short time window (e.g., 1 second), the weights are fine-tuned based on the latest support matrix to avoid the influence of mutation noise. Within a long time window (e.g., 1 minute), the DNN is trained in combination with historical data to ensure the long-term stability of the weight distribution.

[0096] Specifically, for short-term window adjustments, based on the latest support matrix, a lightweight algorithm (such as exponential smoothing or Kalman filtering) can be used to adjust the initial weights of each mileage round, so that nodes with high support receive higher confidence in the short term. If the support of a mileage round drops sharply in a short period of time (for example, exceeding a threshold), its weight is temporarily reduced to prevent outliers from contaminating the fusion results.

[0097] For long-term window adjustments, weight optimization can be performed using a deep neural network (DNN). Historical speed data for each odometry wheel can be input, and network parameters are optimized with the goal of minimizing speed fusion error, outputting the long-term weights for each odometry wheel. The network can receive a support matrix, historical speed data for each odometry wheel, real-time environmental parameters (such as vibration and temperature), and weight values ​​fine-tuned over a short time window. A multi-layer perceptron (MLP) or graph neural network (GNN) architecture can be used to learn high-order dependencies between different odometry wheels. For example, a GNN can aggregate support information from adjacent nodes to enhance robustness against local outliers. Normalized dynamic weights can be generated to ensure that the sum of the weights of all odometry wheels is 1, and softmax or adaptive weighting methods can be used to prioritize nodes with high support. Neural network training can employ reinforcement learning (RL) or online learning strategies. If RL is used, weight assignment can be formulated as a policy optimization problem, with the stability of the final fusion speed (e.g., minimizing variance) serving as a reward signal to dynamically adjust network parameters. If online learning is used, the network can be regularly fine-tuned with the latest data through a sliding window mechanism to adapt to changes in the operating environment.

[0098] In short, the short-term mechanism is similar to a "quick reflex nerve," capable of handling sudden noise; the long-term mechanism is similar to a "memory learning system," helping to ensure global stability. The weights fine-tuned in the short term serve as one of the input features of the DNN, while the long-term weights output by the DNN serve as the baseline for short-term adjustments, forming a closed-loop control system.

[0099] S1055: Calculate the second speed value of the inner detector according to the fourth speed value and the weight value of the odometer.

[0100] In some embodiments, a weighted Bayesian fusion is performed on the fourth speed value of the odometer and the weight value to obtain the second speed value of the inner detector.

[0101] The weighted Bayesian fusion method is used to intelligently integrate the fourth speed values ​​of multiple odometer wheels, fully considering the credibility of each data source and the correlation between them. It can effectively improve the robustness of the system while ensuring calculation accuracy.

[0102] Weighted Bayesian fusion first treats the fourth velocity value of each odometry wheel as a probability distribution, typically assumed to follow a normal distribution with the measured value as the mean and the inverse of the weight as the variance. Based on Bayes' theorem, these distributions are then fused layer by layer: first, the posterior distribution is calculated for each pair, and then the result is fused with the third distribution. This process is repeated until all odometry wheel data has been included in the calculation. During each fusion, the weight coefficients of each distribution are dynamically adjusted based on the real-time support matrix, giving higher-confidence data sources a greater weight in the fusion process. This fusion approach offers significant advantages: first, the Bayesian framework naturally handles measurement uncertainty, preserving the confidence interval of the velocity estimate through the form of a probability distribution; second, the weighting mechanism ensures that outliers do not have a decisive influence on the final result. The resulting second velocity value, calculated after fusion, not only contains the velocity information of each odometry wheel but also reflects the system's confidence in each data source through weighted distribution. This ensures that the second velocity value output by the internal detector is both highly accurate and maintains stable performance.

[0103] In some embodiments, the acquired sensor signals may be pre-processed.

[0104] The preprocessing process can simultaneously process signals from the odometer's angle sensor, the internal detector's acceleration sensor, and the odometer's arm's angle sensor. These raw signals often suffer from noise, inconsistent dimensions, and time series asynchrony. They must undergo a systematic preprocessing process to be converted into structured data suitable for algorithmic use. Preprocessing can include signal noise reduction, data normalization, feature enhancement, time series alignment, and outlier handling. Targeted algorithmic strategies can be employed at each stage to ensure the quality and consistency of input data.

[0105] For signal noise reduction, digital filtering techniques can be used to process the original signal. Depending on the characteristics of different signals, appropriate filter combinations are selected: for the odometer wheel speed signal, a low-pass filter with a cutoff frequency of 10Hz is primarily used. This effectively eliminates high-frequency mechanical vibration interference caused by track irregularities while preserving the true speed characteristics of the internal detector. For the acceleration signal, a bandpass filter (e.g., 0.5-50Hz) is used to eliminate baseline drift and ultra-high frequency noise. For the angle signal, a combination of median filtering and sliding average filtering is used to smooth random fluctuations while maintaining signal edge characteristics. All filters can use zero-phase filtering technology to avoid introducing additional phase distortion. This targeted noise reduction process can greatly improve the output signal-to-noise ratio of each signal, providing a clean signal foundation for subsequent analysis.

[0106] Data normalization is an important step in resolving dimensional differences in multi-source signals. A physical-based normalization method can be used: uniformly convert the odometer speed to linear velocity (m / s), the acceleration signal to standard gravity acceleration (g), and the angle signal to radians. All signals undergo z-score normalization, which involves subtracting the mean and dividing by the standard deviation to ensure that each feature is within a similar numerical range. This process not only eliminates dimensional effects but also allows for direct comparison and fusion of signals from different sensors. At the same time, the normalization parameters can be dynamically updated to adapt to changes in signal characteristics under different operating conditions. Standardized data is more conducive to the training and convergence of algorithms such as neural networks, and can also improve the generalization ability of traditional machine learning models.

[0107] The feature enhancement process focuses on extracting key information from the signal while compressing redundant data. Wavelet transform technology can be used to perform multi-resolution analysis of the signal, selecting the characteristic frequency band that best reflects the operating status of the internal detector for reconstruction. For odometry signals with significant periodicity, synchronous averaging technology is also used to enhance periodic features and suppress random noise. For data compression, an importance-based sampling strategy is adopted, reducing the sampling density in areas of gentle signal changes and maintaining a high sampling rate in areas of rapid changes. This method preserves key features while reducing the amount of data. The feature-enhanced signal not only contains the main information of the original signal but also removes irrelevant details, greatly improving the efficiency of subsequent processing algorithms.

[0108] Timing synchronization is a core challenge in processing multi-source signals. High-precision timestamp alignment technology can be employed, combining hardware triggering with software compensation to ensure that the sampling time error of the three odometer and acceleration signals is within 1ms. Specifically, GPS or PTP protocols are used to synchronize the time of each acquisition node. Interpolation algorithms are then used to time-align the asynchronously sampled signals. This time-aligned signal maintains a strict time correspondence, providing the foundation for multi-sensor data fusion.

[0109] Outlier detection and elimination can be achieved using a combination of multi-level intelligent algorithms. Based on the sliding window statistical method, the historical mean and variance of each signal channel are calculated in real time. The confidence interval is dynamically set using the 3σ principle to automatically identify and eliminate outliers that exceed the reasonable range. Different detection logic can be set for different types of anomalies: for odometer signals, physical consistency verification can be performed in combination with the arm angle information to identify and eliminate abnormal speeds caused by idling or slipping; for acceleration signals, frequency domain analysis can be used to distinguish between real impact events and measurement anomalies. All eliminated outliers will be reasonably filled with adjacent normal data through linear interpolation or spline interpolation to ensure data continuity.

[0110] Through this systematic preprocessing process, raw sensor signals are converted into high-quality, structured input data. This data, with a unified format and time scale, eliminates measurement noise and anomalies, highlights key features, and provides ideal input for subsequent algorithms such as speed calculation and fault diagnosis.

[0111] In some embodiments, based on the first angle change and the third angle value, the following formula can be used to calculate a first height value of the pipeline defect contour that the odometer passes through at the current moment:

[0112] Dh i =H i (1-cosDa i +tana i sinDa i );

[0113] Where Dh i Indicates the first height value, H i Da represents the height of the detector's ith odometer arm joint from the pipe wall at the last moment. i represents the first angle change of the i-th odometer wheel support arm, and a represents the third angle value between the i-th odometer wheel support arm and the vertical line at the previous moment.

[0114] By constructing a change relationship formula between the first height value of the pipeline defect contour passed by the odometer wheel and the first angle change based on trigonometric cosine theory, the first height value of the pipeline defect contour passed by the odometer wheel can be quickly and accurately determined according to the first angle change.

[0115] For the angle value a between the i-th odometer support arm and the vertical line at the last moment, the height value H of the i-th odometer support arm joint of the detector from the pipe wall at the last moment i And the length L of the i-th odometer arm. According to the trigonometric cosine theory, we can get:

[0116] Lcosa=H i ;

[0117] At the current moment, the angle between the i-th odometer support arm and the vertical line changes from a to a+Da, and the height of the i-th odometer support arm joint from the pipe wall changes from H i Change to H i -Dh i According to trigonometric cosine theory, we can get:

[0118] Lcos(a+Da)=H i -Dh i ;

[0119] Simultaneous formula Lcosa=H i And the formula Lcos(a+Da)=H i -Dh i , that is, the relationship between the first height value and the first angle change of the pipeline defect contour that the odometer passes at the current moment can be obtained:

[0120] Dh i =H i (1-cosDa i +tana i sinDa i );

[0121] The first angle variation and the third angle value may be substituted into the above formula to obtain a first height value of the pipeline defect contour that the odometer wheel passes by at the current moment.

[0122] In some embodiments, when the odometer wheel passes over a pipeline defect, a first height value can be obtained by measuring the vertical height change (in millimeters or microns) using a preset sensor. The preset sensor can specifically be a displacement sensor, a contact sensor, or a 3D imaging sensor. Displacement sensors can include laser rangefinders, eddy current sensors, capacitive sensors, and the like. When the odometer wheel passes over the pipeline defect, the displacement sensor can non-contactly measure the defect height to obtain a first profile height value. The contact sensor can be an encoder built into the odometer wheel or a microswitch. The contact sensor indirectly measures height changes through contact deformation to obtain a first profile height value. The 3D imaging sensor can be a structured light scanner or a laser scanner. When the odometer wheel passes over the pipeline defect, the 3D imaging sensor can directly generate a three-dimensional profile of the pipeline defect, thereby obtaining a first profile height value. Pipeline defects can include corrosion, deformation, and weld defects. Corrosion can include uniform corrosion and pitting, and the specific type of corrosion can be determined by the gradient of the first profile height value. Deformation can include concavity or bulging in the pipeline, which can be determined by abnormal fluctuations in the first profile height value. Weld defects may include weld excess height or misalignment, which can be determined by the periodic height change of the first profile height value.

[0123] The pipeline internal detector velocity determination method provided in the embodiments of this specification can obtain a first angle change of the internal detector's odometer at the current moment; calculate the odometer's first velocity at the current moment based on the first angle change; collect a second angle change of the odometer's support arm at the current moment and first acceleration data of the internal detector at the current moment; determine whether the first velocity value is abnormal based on the first angle change, the second angle change, and the first acceleration data; and if the first velocity value is abnormal, correct the first velocity value using the first acceleration data to obtain the internal detector's second velocity value at the current moment. Compared to existing methods, the embodiments of this specification can determine whether the odometer has passed through a pipeline defect, slipped, or idled based on the current first angle change of the odometer, the second angle change of the odometer's support arm, and the first acceleration data of the entire internal detector. Furthermore, the abnormal odometer's first velocity value in these situations can be corrected using the first acceleration data of the entire internal detector to obtain the internal detector's second velocity value, significantly improving the accuracy and reliability of internal detector velocity determination.

[0124] Based on the above-mentioned method for determining the velocity of a detector in a pipeline, this specification also proposes an embodiment of a device for determining the velocity of a detector in a pipeline. Figure 7 As shown, the in-pipe detector velocity determination device 700 may specifically include the following modules:

[0125] An acquisition module 701 may be used to acquire a first angle change of the odometer wheel of the inner detector at a current moment;

[0126] A calculation module 702 may be configured to calculate a first speed value of the odometer at a current moment based on the first angle change;

[0127] The acquisition module 703 may be used to acquire the second angle change of the odometer support arm at the current moment and the first acceleration data of the internal detector at the current moment;

[0128] The judgment module 704 may be configured to judge whether the first velocity value is abnormal based on the first angle change, the second angle change, and the first acceleration data;

[0129] The correction module 705 may be configured to correct the first velocity value using the first acceleration data if the first velocity value is abnormal, to obtain a second velocity value of the internal detector at the current moment.

[0130] In some embodiments, the above-mentioned judgment module 704 can be specifically used to obtain the third angle data of the odometer; the third angle data includes the angle value between the support arm of the odometer and the vertical line at the previous moment; based on the first angle change and the second angle change, calculate the first height value of the pipeline defect contour passed by the odometer at the current moment; if the first height value is greater than a preset height threshold, it is judged that the first speed value is abnormal.

[0131] In some embodiments, the above-mentioned judgment module 704 can also be specifically used to obtain a second height value of the pipeline defect contour passed by the odometer at the previous moment if the first height value is less than or equal to a preset height threshold; if the second height value is greater than the preset height threshold, it is determined that the first speed value is abnormal.

[0132] In some embodiments, the above-mentioned judgment module 704 can also be specifically used to obtain the second acceleration data of the internal detector and the fourth angle change of the odometer if the second altitude value is less than or equal to a preset altitude threshold; if the odometer meets the preset judgment conditions, judge that the first speed value is abnormal; the preset judgment conditions include that the first acceleration data is greater than the second acceleration data and the first angle change is less than the fourth angle change, or that the first acceleration data is zero and the absolute difference between the first angle change and the fourth angle change is greater than a preset deviation threshold.

[0133] In some embodiments, the judgment module 704 may be further configured to judge that the first speed value is normal if the odometer does not meet a preset judgment condition.

[0134] In some embodiments, the correction module 705 can be specifically used to obtain the third speed value of the internal detector at the previous moment; based on the third speed value and the first acceleration data, the first speed value is corrected to obtain the fourth speed value of the odometer at the current moment; based on the fourth speed value and multiple preset support thresholds, the support matrix of the odometer is constructed; based on the maximum modulus eigenvalue and the corresponding eigenvector of the support matrix, the weight value of the odometer is determined; based on the fourth speed value and the weight value of the odometer, the second speed value of the internal detector is calculated.

[0135] In some embodiments, the correction module 705 may be further configured to calculate a fourth velocity value according to the third velocity value and the first acceleration data using the following formula:

[0136] V t =V0+aDt;

[0137] Where V t represents the fourth speed value, V0 represents the third speed value, a represents the first acceleration data, and Dt represents the time interval between the current moment and the previous moment.

[0138] In some embodiments, the correction module 705 may be further configured to construct a support matrix of the odometry wheel using the following formula according to the fourth speed value and a plurality of preset support thresholds:

[0139]

[0140] Where r ij is the element in the i-th row and j-th column of the support matrix, which represents the support of the j-th mileage round of the inner detector at the current moment for the i-th mileage round, j∈[1,N], i∈[1,N], N represents the total number of mileage rounds in the inner detector, N33, v j Indicates the corrected speed value of the jth odometer at the current moment, v i represents the corrected speed value of the ith odometer at the current moment, d1, d2 and d3 represent the preset first support threshold, second support threshold and third support threshold respectively and satisfy 0￡d1 <d3<d2￡1。

[0141] The pipeline internal detector speed determination device provided in the embodiments of this specification can obtain a first angle change of the internal detector's odometer at the current moment; calculate the odometer's first velocity at the current moment based on the first angle change; collect a second angle change of the odometer's support arm at the current moment and first acceleration data of the internal detector at the current moment; determine whether the first velocity value is abnormal based on the first angle change, the second angle change, and the first acceleration data; and if the first velocity value is abnormal, correct the first velocity value using the first acceleration data to obtain the internal detector's second velocity value at the current moment. Compared to existing methods, the embodiments of this specification can determine whether the odometer has passed through a pipeline defect, slipped, or idled based on the first angle change of the odometer, the second angle change of the odometer's support arm, and the first acceleration data of the entire internal detector. Furthermore, the abnormal odometer's first velocity value in these situations can be corrected using the first acceleration data of the entire internal detector to obtain the internal detector's second velocity value, significantly improving the accuracy and reliability of internal detector speed determination.

[0142] It should be noted that the units, devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described in terms of functions and are divided into various modules and described separately. Of course, when implementing this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0143] An embodiment of the present specification also provides a computer device for a method for determining the speed of a detector in a pipeline, including a processor and a memory for storing instructions executable by the processor. When the processor is implemented, it can perform the following steps according to the instructions: obtain a first angle change of the odometer wheel of the internal detector at the current moment; calculate a first speed value of the odometer wheel at the current moment based on the first angle change; collect a second angle change of the odometer wheel support arm at the current moment and first acceleration data of the internal detector at the current moment; determine whether the first speed value is abnormal based on the first angle change, the second angle change and the first acceleration data; if the first speed value is abnormal, use the first acceleration data to correct the first speed value to obtain the second speed value of the internal detector at the current moment.

[0144] In order to complete the above instructions more accurately, refer to Figure 8 As shown, the embodiment of this specification also provides another specific computer device 800, wherein the computer device 800 includes a network communication port 801, a processor 802 and a memory 803, and the above structures are connected through internal cables so that each structure can perform specific data interaction.

[0145] The processor 802 can be specifically used to obtain a first angle change of the odometer wheel of the internal detector at the current moment; calculate a first speed value of the odometer wheel at the current moment based on the first angle change; collect a second angle change of the odometer wheel support arm at the current moment and first acceleration data of the internal detector at the current moment; determine whether the first speed value is abnormal based on the first angle change, the second angle change and the first acceleration data; if the first speed value is abnormal, correct the first speed value using the first acceleration data to obtain the second speed value of the internal detector at the current moment.

[0146] The memory 803 may be specifically used to store corresponding instruction programs.

[0147] In this embodiment, the network communication port 801 can be a virtual port that is bound to different communication protocols, thereby being capable of sending or receiving different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0148] In this embodiment, the processor 802 may be implemented in any suitable manner. For example, the processor may take the form of a microprocessor or a processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, an embedded microcontroller, etc. This specification is not intended to limit this.

[0149] In this embodiment, the memory 803 includes volatile memory and non-volatile memory. The memory 803 can include multiple levels. In digital systems, anything that can store binary data can be considered a memory. In integrated circuits, a circuit with a storage function that does not have a physical form is also called a memory, such as RAM and FIFO. In systems, a physical storage device is also called a memory, such as a memory stick or TF card.

[0150] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0152] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0154] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. 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 determining the velocity of a detector in a pipeline, characterized in that: The method comprises: Obtain the first angle change of the odometer wheel of the inner detector at the current moment; Calculate the first speed value of the odometer wheel at the current moment according to the first angle change; Collecting a second angle change of the odometer support arm at the current moment and first acceleration data of the internal detector at the current moment; determining whether the first velocity value is abnormal according to the first angle change, the second angle change, and the first acceleration data; If the first velocity value is abnormal, the first velocity value is corrected using the first acceleration data to obtain a second velocity value of the internal detector at the current moment.

2. The method according to claim 1, characterized in that The determining whether the first velocity value is abnormal according to the first angle change, the second angle change, and the first acceleration data includes: Obtain the third angle value between the odometer support arm and the vertical line at the previous moment; Calculating a first height value of the pipeline defect contour that the odometer wheel passes through at the current moment based on the first angle change and the third angle value; If the first altitude value is greater than a preset altitude threshold, it is determined that the first speed value is abnormal.

3. The method according to claim 2, characterized in that The determining whether the first velocity value is abnormal based on the first angle change, the second angle change and the first acceleration data further includes: If the first height value is less than or equal to a preset height threshold, obtaining a second height value of the pipeline defect contour that the mileage wheel passed at the previous moment; If the second altitude value is greater than a preset altitude threshold, it is determined that the first speed value is abnormal.

4. The method according to claim 3, characterized in that The determining whether the first velocity value is abnormal based on the first angle change, the second angle change and the first acceleration data further includes: If the second height value is less than or equal to the preset height threshold, obtaining the second acceleration data of the internal detector at the previous moment and the fourth angle change of the odometer at the previous moment; If the odometer meets a preset judgment condition, the first speed value is determined to be abnormal; the preset judgment condition includes that the first acceleration data is greater than the second acceleration data and the first angle change is less than the fourth angle change, or the first acceleration data is zero and the absolute difference between the first angle change and the fourth angle change is greater than a preset deviation threshold.

5. The method according to claim 4, characterized in that: The determining whether the first velocity value is abnormal based on the first angle change, the second angle change and the first acceleration data further includes: If the odometer does not meet the preset determination condition, it is determined that the first speed value is normal.

6. The method according to claim 1, characterized in that The method of correcting the first velocity value by using the first acceleration data to obtain the second velocity value of the internal detector at the current moment includes: Obtain the third speed value of the inner detector at the previous moment; Correcting the first speed value according to the third speed value and the first acceleration data to obtain a fourth speed value of the odometer at the current moment; constructing a support matrix of the odometer according to the fourth speed value and a plurality of preset support thresholds; Determining the weight value of the mileage wheel according to the maximum modulus eigenvalue and the corresponding eigenvector of the support matrix; A second speed value of the inner detector is calculated according to the fourth speed value of the odometer wheel and the weight value.

7. The method according to claim 6, characterized in that The method of correcting the first speed value according to the third speed value and the first acceleration data to obtain the fourth speed value of the odometer at the current moment includes: The fourth velocity value is calculated according to the third velocity value and the first acceleration data using the following formula: V t =V0+aDt; Where V t represents the fourth speed value, V0 represents the third speed value, a represents the first acceleration data, and Dt represents the time interval between the current moment and the previous moment.

8. The method according to claim 6, characterized in that The step of constructing a support matrix of the odometer according to the fourth speed value and a plurality of preset support thresholds includes: According to the fourth speed value and a plurality of preset support thresholds, the support matrix of the odometer is constructed using the following formula: Where r ij is the element in the i-th row and j-th column of the support matrix, which represents the support of the j-th mileage round of the inner detector at the current moment for the i-th mileage round, j∈[1,N], i∈[1,N], N represents the total number of mileage rounds in the inner detector, N≥3, v j Indicates the corrected speed value of the jth odometer at the current moment, v i represents the corrected speed value of the ith odometer at the current moment, d1, d2 and d3 represent the preset first support threshold, second support threshold and third support threshold respectively and satisfy 0￡d1 <d3<d2￡1。 9. A device for determining the velocity of a detector in a pipeline, characterized in that: The device comprises: An acquisition module, configured to acquire a first angle change of the odometer wheel of the inner detector at a current moment; a calculation module, configured to calculate a first speed value of the odometer at a current moment according to the first angle change; An acquisition module, configured to acquire a second angle change of the odometer support arm at a current moment and first acceleration data of the internal detector at a current moment; a judging module, configured to judge whether the first velocity value is abnormal based on the first angle change, the second angle change and the first acceleration data; The correction module is configured to correct the first velocity value using the first acceleration data if the first velocity value is abnormal, so as to obtain a second velocity value of the internal detector at the current moment.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images