A dynamic cable shape recognition system and submarine cable

Abstract

The invention discloses a dynamic cable morphology recognition system and a submarine cable, and relates to the field of cable technology. The system comprises: a detection area module, a dynamic cable detection area is set, a ring laser projection device and a plurality of image collectors are arranged in the dynamic cable detection area; a laser image acquisition module, a laser ring is emitted to the dynamic cable to be detected, and a plurality of dynamic cable surface laser images are obtained; a mapping and reconstruction module, an image mapping framework is built; an image analysis module, a plurality of dynamic cable surface laser images are preprocessed, and the laser image center point and laser image feature points are extracted; a surface reconstruction module, a three-dimensional reconstruction of the laser image feature points is performed, and the dynamic cable surface point cloud data is obtained; a fitting evaluation module, a cylindrical surface is fitted to the dynamic cable surface point cloud data, and the dynamic cable fitting cylindrical surface is obtained and compared with the standard cylindrical surface, and the roundness evaluation result is obtained. The technical effect of achieving efficient online detection and improving the cabling quality of dynamic cables is achieved.

Description

Technical Field

[0001] The present invention relates to the field of cable technology, and in particular to a dynamic cable shape identification system and a submarine cable. Background Art

[0002] Static cables are static submarine cables laid in the trench, and dynamic cables are submarine cables connecting wind turbines and static cable sections. Dynamic cables have at least one hump to reduce the impact of waves on the dynamic cable. In order to ensure that the hump can float, buoyancy blocks are installed on the hump. The trough is connected to an iron chain or a counterweight.

[0003] Compared with static cables, dynamic cables have higher roundness requirements. Once the roundness of dynamic cables is not enough, it will increase the difficulty of installing additional fixed buoyancy blocks and other components. The roundness of dynamic cables is mainly determined by the tightness of the core twisting and the position of the filler. In the actual production process, the roundness of the dynamic submarine cable is not high, and it is difficult to meet the requirements of the cable out-of-roundness. There are technical problems such as difficulty in continuous roundness detection and inaccurate roundness detection. Summary of the invention

[0004] The present invention provides a dynamic cable shape identification system and a submarine cable to solve the technical problems in the prior art that continuous roundness detection is difficult and roundness detection is inaccurate, and achieves the technical effect of efficient online detection and improving the quality of dynamic cable cabling.

[0005] In a first aspect, the present invention provides a dynamic cable shape identification system, wherein the system comprises:

[0006] A detection zone module, wherein the detection zone module is used to set a dynamic cable detection zone, arrange an annular laser projection device in the dynamic cable detection zone, arrange multiple image collectors around the circumference of the annular laser projection device, determine the calibration positions of the multiple image collectors, and establish communication connection with the image acquisition card;

[0007] A laser image acquisition module, wherein the laser image acquisition module is used to emit a laser ring to the dynamic cable to be detected based on the annular laser projection device, and to simultaneously trigger multiple image collectors to perform image acquisition through an image acquisition card to obtain multiple dynamic cable surface laser images;

[0008] A mapping reconstruction module, the mapping reconstruction module is used to establish a position mapping relationship of multiple image collectors according to the calibrated positions, and to build an image mapping framework according to the position mapping relationship;

[0009] An image analysis module, the image analysis module is used to preprocess the multiple dynamic cable surface laser images, obtain multiple preprocessed surface laser images, and extract laser image center points and laser image feature points from the multiple preprocessed surface laser images;

[0010] A surface reconstruction module, the surface reconstruction module is used to perform three-dimensional reconstruction on the characteristic points of the laser image based on the center point of the laser image and the image mapping framework to obtain dynamic cable surface point cloud data;

[0011] A fitting evaluation module, wherein the fitting evaluation module is used to perform cylindrical surface fitting on the dynamic cable surface point cloud data, obtain the dynamic cable fitting cylindrical surface, and compare the dynamic cable fitting cylindrical surface with a standard cylindrical surface to obtain a roundness evaluation result, wherein the standard cylindrical surface is constructed according to a roundness constraint.

[0012] In a second aspect, the present invention provides a submarine cable comprising: a core, a metal armor layer and an outer sheath, wherein the outer sheath is manufactured by real-time monitoring of roundness using the dynamic cable shape recognition system as described in the first aspect above.

[0013] The present invention discloses a dynamic cable morphology recognition system and a submarine cable, comprising: a detection zone module, which is used to set a dynamic cable detection zone, arrange an annular laser projection device in the dynamic cable detection zone, arrange multiple image collectors in the circumferential direction of the annular laser projection device, determine the calibration positions of the multiple image collectors, and communicate with the image acquisition card; a laser image acquisition module, which is used to emit a laser ring to the dynamic cable to be detected based on the annular laser projection device, and simultaneously trigger multiple image collectors to perform image acquisition through the image acquisition card to obtain multiple dynamic cable surface laser images; a mapping and reconstruction module, which is used to establish a position mapping relationship of multiple image collectors according to the calibration position, and build an image mapping framework according to the position mapping relationship; an image analysis module, which is used to analyze the multiple dynamic cable surface laser images Preprocessing, obtaining multiple preprocessed surface laser images, and extracting laser image center points and laser image feature points from the multiple preprocessed surface laser images; a surface reconstruction module, for performing three-dimensional reconstruction of the laser image feature points based on the laser image center point and the image mapping framework, and obtaining dynamic cable surface point cloud data; a fitting evaluation module, for performing cylindrical surface fitting on the dynamic cable surface point cloud data, obtaining the dynamic cable fitting cylindrical surface, and comparing the dynamic cable fitting cylindrical surface with the standard cylindrical surface to obtain a roundness evaluation result, wherein the standard cylindrical surface is constructed according to the roundness constraint. A dynamic cable morphology recognition system and submarine cable disclosed in the present invention solve the technical problems of difficulty in continuous roundness detection and inaccurate roundness detection, and achieve the technical effect of efficient online detection and improving the quality of dynamic cable cabling. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a structural schematic diagram of a dynamic cable shape recognition system of the present invention;

[0015] Figure 2 The figure is a schematic structural diagram of a submarine cable of the present invention. DETAILED DESCRIPTION

[0016] Some terms involved in the embodiments of the present invention are explained in advance below.

[0017] Dynamic cable: The dynamic cable refers to a dynamic submarine cable, which is used to connect wind turbines and static cable segment submarine cables. The dynamic cable is a multi-layer composite optical cable including a cable core, an optical unit, a special-shaped filling layer, a reinforcing wrapping tape, an inner sheath, an armor layer, and an outer sheath.

[0018] The ring laser projection device is a special form of laser that can directly output a ring laser beam. Compared with ordinary point or line lasers, it has the following characteristics: the ring laser outputs a hollow ring-shaped laser spot instead of a solid point or line. The optical axis of the ring laser beam coincides with the central axis of the laser body and has good coaxiality. The laser power output of the ring laser is evenly distributed along the radial direction of the ring, and the light intensity is consistent.

[0019] The image collector may be a white light camera, a black and white camera, an industrial camera, or other image collection devices that can achieve similar functional utilities.

[0020] The technical solution provided in the embodiments of the present invention is to solve the technical problems of the prior art that the continuous roundness detection is difficult and the roundness detection is inaccurate. The overall idea adopted is as follows:

[0021] First, an annular laser projection device is set in the dynamic cable detection area, and multiple image collectors are arranged around the circumferential direction of the annular laser projection device, their calibration positions are determined, and they are connected to the image acquisition card for communication. Then, based on the annular laser projection device, a laser ring is emitted to the dynamic cable to be detected, and multiple image collectors are triggered simultaneously through the image acquisition card to collect images, and multiple dynamic cable surface laser images are obtained. Then, the position mapping relationship of multiple image collectors is established according to the calibration position, and an image mapping framework is built. Then, the multiple dynamic cable surface laser images are preprocessed to extract the center point of the laser image and the laser image feature points. Finally, based on the center point of the laser image and the image mapping framework, the laser image feature points are three-dimensionally reconstructed to obtain the dynamic cable surface point cloud data, and then the point cloud data is fitted with a cylindrical surface to obtain the dynamic cable fitting cylindrical surface, and it is compared with the standard cylindrical surface to obtain the roundness evaluation result, wherein the standard cylindrical surface is constructed according to the roundness constraint.

[0022] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0023] Figure 1 The present invention is a structural schematic diagram of a dynamic cable shape recognition system; wherein, it includes:

[0024] A detection zone module 11, wherein the detection zone module 11 is used to set a dynamic cable detection zone, arrange an annular laser projection device in the dynamic cable detection zone, arrange multiple image collectors around the circumference of the annular laser projection device, determine the calibration positions of the multiple image collectors, and communicate with the image acquisition card;

[0025] Among them, the dynamic cable detection area refers to a specific area or section on the target dynamic cable production line used for dynamic cable roundness detection. Exemplarily, the dynamic cable detection area is located at the molding end of the target dynamic cable production line and after the outer sheath molding process of the target dynamic cable production line.

[0026] Optionally, the dynamic cable detection area is equipped with a ring-shaped laser projection device, which is a special form of laser as mentioned above, and can directly output a ring-shaped laser beam, and is used to output and project the ring-shaped laser beam to the outer surface of the target dynamic cable. The ring-shaped laser beam output by the ring-shaped laser projection device has high brightness and good directionality.

[0027] Optionally, the deployment of the annular laser projection device includes aligning the axis of the annular laser projection device with the axis of the dynamic cable conveying direction in the target dynamic cable production line, thereby ensuring that the plane formed by the annular laser beam projected on the target dynamic cable is perpendicular to the axis of the target dynamic cable to be detected.

[0028] Optionally, the layout of the annular laser projection device also includes the setting of working parameters such as line width, power uniformity, brightness, stability, etc. Among them, narrower laser line width, higher power uniformity and stability, and appropriate brightness are conducive to improving the accuracy of subsequent recognition.

[0029] Optionally, multiple image collectors are arranged around the circumferential direction of the annular laser projection device, including arranging multiple image collectors in multiple directions around the circumference of the annular laser projection device. These image collectors can be cameras or other image sensors, which are used to capture the image formed on the cable by the light beam projected by the annular laser projection device. Specifically, the arrangement of the multiple image collectors includes single-sided arrangement and double-sided arrangement relative to the circumferential direction of the annular laser projection device.

[0030] Optionally, multiple image collectors are arranged around the circumferential direction of the annular laser projection device, and the multiple image collectors arranged have overlapping collection ranges, and the effective collection viewing angles of the multiple image collectors are connected end to end, so that the effective collection angle formed by the multiple image collectors covers all angles of the circumference of the target dynamic cable.

[0031] Optionally, the arrangement of the annular laser projection device also includes setting the projection direction of the output annular laser beam. Specifically, the projection direction is perpendicular to the target dynamic cable axis to be detected or forms a certain angle with the target dynamic cable axis to be detected. Among them, the projection direction of the annular laser projection device is perpendicular to the target dynamic cable axis to be detected, which is conducive to the arrangement of multiple image collectors on the left and right sides of the annular laser projection device, and the projection direction of the annular laser projection device forms a certain angle with the target dynamic cable axis to be detected, which helps to provide more ample arrangement space for multiple image collectors.

[0032] Optionally, an image acquisition card is used to communicate with multiple image collectors, thereby realizing integrated control of multiple image collectors, and at the same time facilitating identification and marking of multi-angle images acquired by multiple image collectors. Exemplarily, the image acquisition card has multiple independent image acquisition input channels, and each channel has a corresponding independent number. The image collector, the independent image acquisition input channels, and the corresponding independent numbers constitute a mapping relationship between multiple image collectors and the calibration position.

[0033] In some embodiments, determining the calibration positions of the plurality of image collectors includes:

[0034] A calibration plate is arranged in the dynamic cable detection area, and a plurality of image collectors are controlled to capture images of the calibration plate to obtain a plurality of images of the calibration plate;

[0035] The feature points of multiple calibration plate images are extracted, and the mapping relationship between the image and the three-dimensional space is established according to the pixel coordinates of the feature points to determine the calibration positions of multiple image collectors.

[0036] Among them, the imaging process of multiple image collectors includes the transformation of four coordinate systems. Specifically, first, the world coordinate system is transformed into the camera coordinate system after rigid body transformation (such as rotation and translation), then transformed into the image coordinate system through perspective projection, and finally transformed into the pixel coordinate system through affine transformation.

[0037] Optionally, determining the calibration positions of multiple image collectors includes determining the internal parameter matrix and the external parameter matrix of the multiple image collectors. Preferably, the calibration of multiple image collectors is performed based on Zhang's chessboard calibration method. By shooting a chessboard of known shape from multiple perspectives, calibration is performed using a two-dimensional image instead of a three-dimensional image. Then, the calibration positions of multiple image collectors are determined.

[0038] Exemplarily, first, a calibration plate is placed at an appropriate position in the dynamic cable detection area. The calibration plate has a known checkerboard image for extracting image feature points. Then, the image collector is controlled to capture the image of the calibration plate. And ensure that the captured calibration plate image is complete and the image quality is good. Next, multiple calibration plate images are processed to extract feature points in the image. Feature points are corners, intersections or other significant features of the checkerboard image. Then, based on the pixel coordinates of the extracted feature points and the known size and shape of the calibration plate, the camera calibration theory is used to calculate the product of the internal parameter matrix and the external parameter matrix, solve the internal parameter matrix, and solve the external parameter matrix, thereby establishing a mapping relationship between the image and the three-dimensional space. Then, based on the established mapping relationship between the image and the three-dimensional space, the calibration position of each image collector is determined. Specifically, it includes the internal parameters of the camera (such as focal length, principal point offset, etc.) and external parameters (such as the position and direction of the camera). Through the above steps, the calibration positions of multiple image collectors can be accurately determined, thereby ensuring that the acquired image acquisition results can be accurately converted into coordinates and reconstructed in three dimensions, which is helpful for subsequent cable detection and analysis.

[0039] Optionally, determining the calibration positions of the plurality of image collectors also includes iteratively optimizing the parameters by the LM method. This method combines the steepest descent method with a linearization method (Taylor series), because the steepest descent method is applicable to the early stages of the iteration when the parameter estimates are far from the optimal value, while the linearization method, such as the Gauss-Newton method, is applicable to the late stages of the iteration when the parameter estimates are close to the optimal value. Combining these two methods can find the optimal value more quickly.

[0040] A laser image acquisition module 12, the laser image acquisition module 12 is used to emit a laser ring to the dynamic cable to be detected based on the annular laser projection device, and trigger multiple image collectors to perform image acquisition through an image acquisition card to obtain multiple dynamic cable surface laser images;

[0041] Optionally, after completing the layout of the dynamic cable detection area and the calibration and connection of multiple image collectors, the image acquisition card is controlled to synchronously activate multiple input channels through the control circuit, thereby simultaneously triggering multiple image collectors to capture images of the target dynamic cable.

[0042] Optionally, multiple image collectors perform image capture on the target dynamic cable based on a preset capture granularity, which defines capture parameters of the image collector such as capture interval and capture clarity, and the preset capture granularity is determined based on the roundness control requirement of the target dynamic cable. A higher roundness control requirement corresponds to a finer capture granularity. In other words, for a target dynamic cable with stricter roundness control requirements, the corresponding capture interval is shorter and the capture clarity is higher.

[0043] A mapping reconstruction module 13, the mapping reconstruction module 13 is used to establish a position mapping relationship of multiple image collectors according to the calibrated positions, and to build an image mapping framework according to the position mapping relationship;

[0044] Preferably, the mapping and reconstruction module 13 performs position mapping of multiple image collectors according to the camera intrinsic parameter matrix and extrinsic parameter matrix determined by the calibration position, thereby realizing reconstruction of the three-dimensional coordinate system and obtaining the three-dimensional reconstructed coordinate system. The above-mentioned image mapping framework is the obtained three-dimensional reconstructed coordinate system.

[0045] Exemplarily, a three-dimensional reconstruction coordinate system is constructed with the axis of the target dynamic cable as the X-axis of the image mapping frame and the plane formed by the annular laser beam projected on the target dynamic cable as the YZ plane; then, through mathematical methods, including affine transformation and perspective transformation, the calibrated positions of multiple image collectors are mapped to a unified coordinate system, thereby determining the spatial relationship between the image collectors.

[0046] The acquisition of the image mapping framework helps to subsequently merge and process the images collected by multiple collectors to obtain a complete dynamic cable surface laser image, wherein the above-mentioned complete dynamic cable surface laser image refers to a panoramic image of the projection position of the target dynamic cable annular laser beam.

[0047] An image analysis module 14, the image analysis module is used to preprocess the multiple dynamic cable surface laser images, obtain multiple preprocessed surface laser images, and extract laser image center points and laser image feature points from the multiple preprocessed surface laser images;

[0048] The multiple dynamic cable surface laser images are preprocessed to reduce redundant information and noise in the multiple dynamic cable surface laser images, thereby improving the availability of the multiple dynamic cable surface laser images in subsequent extraction of laser image center points and laser image feature points.

[0049] Exemplarily, multiple dynamic cable surface laser images are preprocessed, including denoising, contrast enhancement, ROI interception, grayscale threshold selection and other operations to improve image quality and feature visualization, while achieving image data reduction for subsequent analysis and processing.

[0050] Preferably, the contrast enhancement preprocessing of the multiple dynamic cable surface laser images is performed by segmented grayscale transformation or nonlinear grayscale transformation.

[0051] In some embodiments, extracting the center point of the laser image includes:

[0052] Binarizing the plurality of preprocessed surface laser images, extracting a plurality of surface laser profiles, calculating the geometric centers of the plurality of surface laser profiles, and obtaining a plurality of surface laser center lines;

[0053] In the dynamic cable detection process, the pixel coordinates of the center points of the plurality of pre-processed surface laser images are extracted to obtain a plurality of pixel coordinates;

[0054] The plurality of pixel coordinates are determined as the pixel coordinates of the center point of the laser image, and the laser image center points of the plurality of preprocessed surface laser images are determined.

[0055] Specifically, through binarization processing, multiple surface laser profiles are extracted from multiple pre-processed surface laser images based on an edge detection algorithm. The multiple surface laser profiles reflect the shape characteristics of the annular laser beam projected on the target dynamic cable, specifically, including the distortion and width of the annular laser beam.

[0056] Optionally, the shape characteristics of the annular laser beam are analyzed based on the extracted multiple surface laser profiles, and the geometric centers of the multiple surface laser profiles are calculated. Specifically, first, the geometric center points of the profiles are calculated segmentally to obtain multiple geometric centers of multiple positions of the surface laser profiles, and then these geometric centers are connected and fitted based on mathematical methods to obtain the surface laser center line. The acquisition of the surface laser center line is conducive to improving the accuracy of subsequent laser image analysis, thereby ensuring the dynamic cable shape recognition effect.

[0057] Optionally, during the dynamic cable detection process, the pixel coordinates of the center points of multiple preprocessed surface laser images are extracted, and these pixel coordinates are determined as the center point pixel coordinates of the laser image. In other words, the extracted center point pixel coordinates are the center point of the imaging plane corresponding to the laser image.

[0058] Optionally, based on the pixel coordinates of multiple laser image center points and combined with the focal length information of multiple image collectors, the laser image center points of multiple preprocessed surface laser images are obtained, and the laser image center point is the optical center of the preprocessed surface laser image.

[0059] A surface reconstruction module 15, which is used to perform three-dimensional reconstruction on the characteristic points of the laser image based on the center point of the laser image and the image mapping framework to obtain dynamic cable surface point cloud data;

[0060] Optionally, based on the principle of laser triangulation, combined with the position mapping relationship of the image collector and the center point of the laser image, the three-dimensional coordinates of the laser image feature points are calculated. Exemplarily, first, the lines connecting the pixel points corresponding to the multiple laser image feature points on the surface laser center line and the laser image center point are obtained to form a ray set; then, based on the incident angle and position information of the laser beam projected by the annular laser projection device, the intersection of the laser projection surface of the annular laser projection device and the above ray set is obtained to determine the three-dimensional coordinates of the laser image feature points; then, using the image mapping framework, the pixel coordinates of the laser image feature points are mapped to the three-dimensional coordinate system, and the positions of the multiple laser image feature points on the surface laser center line in the three-dimensional space are obtained, so as to realize the three-dimensional reconstruction of the laser image feature points and form the dynamic cable surface point cloud data.

[0061] The fitting evaluation module 16 is used to perform cylindrical surface fitting on the dynamic cable surface point cloud data, obtain the dynamic cable fitting cylindrical surface, and compare the dynamic cable fitting cylindrical surface with the standard cylindrical surface to obtain a roundness evaluation result, wherein the standard cylindrical surface is constructed according to the roundness constraint.

[0062] Optionally, the surface point cloud data of the dynamic cable obtained based on multiple laser image feature points on the surface laser centerline is fitted, which means fitting multiple groups of dynamic cable surface point cloud data continuously obtained at multiple moments to form a continuous and real-time updated dynamic cable fitting cylindrical surface. The dynamic cable fitting cylindrical surface reflects the surface shape characteristics of the target dynamic cable passing through the dynamic cable detection area. It provides a data basis for the online continuous morphological recognition of the roundness of the target dynamic cable production line.

[0063] Exemplarily, a surface fitting algorithm such as the least square method is used to fit a dynamic cable fitting cylindrical surface, and the dynamic cable fitting cylindrical surface is a cylindrical surface that best represents the surface shape of the dynamic cable.

[0064] Optionally, a standard cylindrical surface model is constructed according to the roundness constraint, and the standard cylindrical surface is used as a reference standard. The roundness evaluation is performed by comparing the dynamic cable fitting cylindrical surface with the standard cylindrical surface. The roundness constraint is determined according to the actual quality control requirements and standard specifications of the target dynamic cable production line. Preferably, the roundness constraint requires that the cable out-of-roundness is not greater than 2%.

[0065] Optionally, the dynamic cable fitting cylindrical surface is compared with the standard cylindrical surface to obtain a roundness evaluation result, wherein the roundness evaluation result includes the non-roundness satisfaction and the non-roundness value. Exemplarily, the parameters of the dynamic cable fitting cylindrical surface are compared with the parameters of the standard cylindrical surface to calculate the difference between them. The comparison method includes calculating the difference in the radius of the cylindrical surface. Specifically, first, the maximum diameter and the minimum diameter on the cross section of the dynamic cable fitting cylindrical surface are obtained; then, the intersection of the maximum diameter and the minimum diameter is used as the center of the cross section of the dynamic cable fitting cylindrical surface; then, the center of the standard cylindrical surface cross section is aligned with the center of the cross section of the dynamic cable fitting cylindrical surface, and the inner deviation extreme value and the outer deviation extreme value of the cross section of the dynamic cable fitting cylindrical surface and the standard cylindrical surface cross section are obtained; finally, based on the sum of the above inner deviation extreme value and the outer deviation extreme value, the non-roundness of the dynamic cable fitting cylindrical surface is calculated, and a threshold judgment is performed according to the above roundness constraint, and the threshold judgment result and the non-roundness value are output to generate a roundness evaluation result.

[0066] Optionally, the roundness evaluation result also includes the outer deviation azimuth and the inner deviation azimuth of the cross section of the dynamic cable fitting cylinder compared with the standard cylindrical cross section.

[0067] Through the above steps, the dynamic cable fitting cylindrical surface can be compared with the standard cylindrical surface, and then the roundness evaluation result can be obtained, which provides a reference for evaluating the surface quality of the dynamic cable.

[0068] In some embodiments, the system further comprises a roundness compensation unit, configured to:

[0069] A roundness compensation area is provided after the dynamic cable detection area;

[0070] When the roundness evaluation result does not satisfy the roundness constraint, extracting cabling parameters of the dynamic cable to be detected, wherein the cabling parameters include twisting parameters and filling parameters;

[0071] Compensation extrusion parameters are generated according to the cabling parameters and the roundness constraint, and roundness compensation is performed on the dynamic cable to be detected.

[0072] Among them, twisting refers to the process of twisting multiple strands of wire or rope together to form a cable, and the twisting parameters refer to the way and degree of twisting of the conductors and armor layers in the cable. The twisting parameters describe the twisting method, twisting angle, twisting density and other information, which have an important impact on the strength and flexibility of the cable. Filling is used to fill the gaps between twisted wires to increase the density and stability of the cable. The filling parameters describe the type of filling material, filling density, and filling amount of each part of the target dynamic cable cross-section, which have an important impact on the shape and roundness of the cable.

[0073] Exemplarily, a roundness compensation area is set after the dynamic cable detection area to compensate for the detected cables that do not meet the roundness constraints. First, the cabling parameters of the dynamic cable to be detected are extracted. The cabling parameters include twisting parameters and filling parameters. The above cabling parameters affect the roundness of the dynamic cable. Then, based on the extracted cabling parameters and roundness constraints, compensating extrusion parameters are generated. Among them, the compensating extrusion parameters include extrusion force, extrusion footage, extrusion direction angle, etc. Finally, the generated compensating extrusion parameters are used to compensate for the roundness of the dynamic cable to be detected, and the cable is re-extruded or adjusted to improve its roundness. This includes extruding the filling position corresponding to the outer deviation angle of the roundness, and extruding the filling positions corresponding to both sides of the inner deviation angle of the roundness, so that they meet the roundness requirements.

[0074] Through the above steps, the roundness compensation unit can compensate for the dynamic cable that does not meet the roundness constraint and improve its quality and performance.

[0075] In some implementations, generating a compensating extrusion parameter according to the cabling parameter and the roundness constraint includes:

[0076] Collecting samples of the dynamic cable to be detected to obtain a cabling parameter sample set and a dynamic cable sample set;

[0077] Performing a roundness mark on the dynamic cable sample set according to the roundness constraint, and constructing a compensation extrusion parameter sample set through the roundness mark and the cabling parameter sample set;

[0078] Constructing a roundness compensator based on the cabling parameter sample set, the roundness constraint, and the compensation extrusion parameter sample set;

[0079] The cabling parameters and the roundness constraint are input into a roundness compensator to generate compensation extrusion parameters.

[0080] Optionally, the cabling parameter sample set is a set of cabling parameters that are manually fine-tuned and verified to be effective by professional technicians or an expert system based on the above-mentioned roundness identification, and each set of cabling parameter samples in the cabling parameter sample set has corresponding roundness constraints and roundness identifications.

[0081] Optionally, the roundness compensator is an end-to-end roundness compensation decision model built based on a machine learning algorithm, and the roundness compensator is trained through a cabling parameter sample set, a roundness constraint, and a compensation extrusion parameter sample set. The parameters of the model are adjusted by an optimization algorithm (such as gradient descent) so that the roundness compensator can obtain the mapping relationship between the cabling parameters and the roundness identifier and the compensation extrusion parameter. Exemplarily, the roundness compensator is built based on a machine learning model or a deep learning model such as a regression model, a support vector machine, and a neural network.

[0082] In some embodiments, the system further comprises a cabling optimization control unit configured to:

[0083] Determining a value range of the cabling parameter according to the roundness constraint, and constructing a parameter optimization space based on the value range;

[0084] When the roundness evaluation result does not satisfy the roundness constraint, generating a roundness deviation according to the roundness evaluation result;

[0085] Based on the cabling parameters of the dynamic cable to be detected, parameter optimization is performed in the parameter optimization space according to the roundness deviation to obtain optimized cabling parameters;

[0086] The optimized cabling parameters are applied to a cabling control area to perform dynamic cabling roundness control, wherein the cabling control area is before the dynamic cabling detection area.

[0087] Optionally, the value range of the cabling parameters is determined according to the roundness constraint, and the parameter optimization space is constructed. The value range of the cabling parameters is an empirical range obtained by mathematical statistics based on historical dynamic cable production records or an experimental result based on dynamic cable production experimental data. The cabling parameters within this range can meet the roundness requirement.

[0088] Optionally, the determined cabling parameter value range is converted into a parameter optimization space. The parameter optimization space can be regarded as a multidimensional space, in which each dimension represents a cabling parameter, and its value range is the range of the dimension. The entire multidimensional space contains all possible cabling parameter combinations.

[0089] Optionally, when the roundness evaluation result does not meet the roundness constraint, parameter optimization is performed. First, the roundness deviation is calculated according to the roundness evaluation result and the roundness constraint. The roundness deviation represents the difference between the roundness under the current cabling parameters and the required roundness; then, based on the cabling parameters of the dynamic cable to be tested, an optimization algorithm (such as a genetic algorithm, a particle swarm algorithm, etc.) is used to perform parameter optimization, and the optimal cabling parameter combination is searched in the parameter optimization space to minimize the roundness deviation.

[0090] Optionally, the cost function in the optimization algorithm evaluates the roundness of the cabling parameter combination during the parameter optimization process based on the simulation results of the simulation model. Exemplarily, first, based on the twin simulation technology, combined with the finite element analysis software, a simulation model simulating the dynamic cable shape is established. The simulation model simulates the cabling process through the inherent parameters of the cabling parameters and the target dynamic cable, and outputs the cabling simulation results. Among them, the cabling simulation results include the roundness of the simulated dynamic cable product.

[0091] Optionally, the optimized cabling parameters are applied to the cabling control area for dynamic cable production control. The cabling control area is located before the dynamic cable detection area, and the shape of the dynamic cable is controlled by adjusting the cabling parameters to achieve roundness requirements.

[0092] In some implementations, obtaining optimized cabling parameters includes:

[0093] The parameter optimization is a segmented backtracking optimization of parameters;

[0094] Determining the cabling parameters of the dynamic cable to be detected as initial cabling parameters, and determining the twisting parameters and the filling parameters as initial twisting parameters and initial filling parameters respectively;

[0095] The initial twisting parameters are locked, and the initial filling parameters are optimized in the parameter optimization space to obtain the filling parameters to be tested;

[0096] The filling parameters to be tested are locked, and the initial twisting parameters are optimized in the parameter optimization space to obtain the twisting parameters to be tested;

[0097] Continue to iterate and optimize the twisting parameters to be tested and the filling parameters to be tested to obtain optimized cabling parameters.

[0098] Optionally, obtaining the optimized cabling parameters includes segmented backtracking optimization of the initial twisting parameters and the initial filling parameters. In other words, after the cabling parameters of the dynamic cable to be detected are determined as the initial cabling parameters, the initial twisting parameters and the initial filling parameters are alternately optimized. Specifically, first lock any one of the initial twisting parameters and the initial filling parameters, then optimize the other unlocked parameter, and combine the locked parameter with the other optimized parameter to generate a combined parameter for overall evaluation.

[0099] Optionally, if the evaluation result of the combined parameters is optimized compared with the evaluation result of the initial cabling parameters, iterative optimization is performed based on the combined parameters; specifically, the twisting parameters or initial filling parameters that were not locked in the previous cycle in the combined parameters are locked, and the initial filling parameters or twisting parameters that were locked in the previous cycle in the combined parameters are optimized. Repeat the above combination and overall evaluation.

[0100] Optionally, if the evaluation result of the combined parameters does not have an optimization effect compared with the evaluation result of the initial cabling parameters, then the other unlocked parameter in the combined parameters is updated and optimized until the evaluation result of the updated combined parameters has an optimization effect compared with the evaluation result of the initial cabling parameters, and then the locked parameters and the unlocked parameters are exchanged based on the aforementioned method for iterative optimization.

[0101] Optionally, optimized cabling parameters are obtained, including a jump-out judgment for iterative optimization based on preset optimization constraints; illustratively, the preset optimization constraints include: if the change in the optimization constraints is small enough or the optimization algorithm has reached a preset maximum number of iterations, then the current iteration is jumped out and it is considered that the optimization has converged.

[0102] In some embodiments, the system further comprises a secondary detection unit for:

[0103] Setting a secondary detection area in the roundness compensation area based on the dynamic cable detection area;

[0104] After the roundness compensation is performed on the dynamic cable to be tested, the dynamic cable to be tested is transferred to the secondary testing area, and a secondary roundness test is performed on the dynamic cable to obtain a compensated roundness evaluation result;

[0105] The dynamic cable to be detected is marked according to the compensation roundness evaluation result.

[0106] Optionally, the secondary detection area is arranged based on the same method and principle as the aforementioned dynamic cable detection area, and also includes arranging a ring laser projection device, arranging multiple image collectors around the circumference of the ring laser projection device, determining the calibration positions of the multiple image collectors, and communicating with the secondary image acquisition card. It should be understood that for the sake of brevity of the specification, no further expansion is given here.

[0107] Optionally, based on the compensation roundness evaluation result, the dynamic cable to be tested after roundness compensation is marked; specifically, the dynamic cable to be tested whose compensation roundness evaluation result does not meet the roundness constraint is marked as an unqualified dynamic cable. The dynamic cable to be tested whose compensation roundness evaluation result meets the roundness constraint is marked as a qualified dynamic cable.

[0108] Optionally, according to the identification result of the dynamic cable to be tested, the qualified dynamic cable is subjected to subsequent production processes, including performance testing, labeling, packaging, etc. of the dynamic cable, to ensure the safety and stability of the dynamic cable.

[0109] In summary, the dynamic cable shape recognition system provided by the present invention has the following technical effects:

[0110] A dynamic cable detection area is set up, an annular laser projection device is arranged in the dynamic cable detection area, multiple image collectors are arranged around the annular laser projection device in the circumferential direction, the calibration positions of the multiple image collectors are determined, and they are communicated with the image acquisition card; based on the annular laser projection device, a laser ring is emitted to the dynamic cable to be detected, and the image acquisition card is used to simultaneously trigger multiple image collectors to perform image acquisition, and multiple dynamic cable surface laser images are obtained; a position mapping relationship of multiple image collectors is established according to the calibration position, and an image mapping framework is built according to the position mapping relationship; multiple dynamic cable surface laser images are preprocessed to obtain multiple preprocessed surface laser images, and multiple A preprocessed surface laser image is used to extract the laser image center point and the laser image feature point; based on the laser image center point and the image mapping framework, the laser image feature point is three-dimensionally reconstructed to obtain the dynamic cable surface point cloud data; a cylindrical surface is fitted to the dynamic cable surface point cloud data to obtain the dynamic cable fitting cylindrical surface, and the dynamic cable fitting cylindrical surface is compared with the standard cylindrical surface to obtain the roundness evaluation result, wherein the standard cylindrical surface is constructed according to the roundness constraint. A dynamic cable morphology recognition method disclosed in the present invention solves the technical problems of difficulty in continuous roundness detection and inaccurate roundness detection, and achieves the technical effect of efficient online detection and improving the quality of dynamic cable cabling.

[0111] Figure 2 The schematic diagram of the structure of a submarine cable of the present invention comprises: a core 21, a metal armor layer 22 and an outer sheath 23. The outer sheath 23 is manufactured by real-time monitoring of roundness using a dynamic cable shape recognition system.

[0112] Optionally, a plurality of wire cores 21 constitute a cable core of the dynamic cable, and the outer sheath 23 is arranged on the outermost side of the cable core.

[0113] In some embodiments, the wire core 21 includes a copper conductor 211, a metal shielding layer 212 and an inner sheath 213 from the inside to the outside. The inner sheath 213 is made by using a dynamic cable shape recognition system to perform real-time monitoring of roundness.

[0114] Optionally, the roundness deviation value of the inner sheath 213 and the outer sheath 23 is in the range of [0.001, 0.005] from the maximum radial distance difference between the actual contour and the ideal circle, and the unit is mm.

[0115] In summary, the submarine cable provided by the present invention has the following technical effects: a dynamic cable shape recognition system is used to perform real-time monitoring of roundness to make a wire core, a metal armor layer and an outer sheath. A submarine cable monitoring method disclosed in the present invention solves the technical problems of difficulty in continuous monitoring of roundness and inaccurate roundness detection, ensuring that the roundness of the dynamic cable is always in a high standard state.

[0116] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A dynamic cable shape recognition system, characterized in that: The system comprises: A detection zone module, wherein the detection zone module is used to set a dynamic cable detection zone, arrange an annular laser projection device in the dynamic cable detection zone, arrange multiple image collectors around the circumference of the annular laser projection device, determine the calibration positions of the multiple image collectors, and establish communication connection with the image acquisition card; A laser image acquisition module, wherein the laser image acquisition module is used to emit a laser ring to the dynamic cable to be detected based on the annular laser projection device, and to simultaneously trigger multiple image collectors to perform image acquisition through an image acquisition card to obtain multiple dynamic cable surface laser images; A mapping reconstruction module, the mapping reconstruction module is used to establish a position mapping relationship of multiple image collectors according to the calibrated positions, and to build an image mapping framework according to the position mapping relationship; An image analysis module, the image analysis module is used to preprocess the multiple dynamic cable surface laser images, obtain multiple preprocessed surface laser images, and extract laser image center points and laser image feature points from the multiple preprocessed surface laser images; A surface reconstruction module, the surface reconstruction module is used to perform three-dimensional reconstruction on the characteristic points of the laser image based on the center point of the laser image and the image mapping framework to obtain dynamic cable surface point cloud data; A fitting evaluation module, wherein the fitting evaluation module is used to perform cylindrical surface fitting on the dynamic cable surface point cloud data, obtain the dynamic cable fitting cylindrical surface, and compare the dynamic cable fitting cylindrical surface with a standard cylindrical surface to obtain a roundness evaluation result, wherein the standard cylindrical surface is constructed according to a roundness constraint.

2. The system according to claim 1, characterized in that Determine the calibration positions of multiple image collectors, including: A calibration plate is arranged in the dynamic cable detection area, and a plurality of image collectors are controlled to capture images of the calibration plate to obtain a plurality of images of the calibration plate; The feature points of multiple calibration plate images are extracted, and the mapping relationship between the image and the three-dimensional space is established according to the pixel coordinates of the feature points to determine the calibration positions of multiple image collectors.

3. The system according to claim 2, characterized in that Extract the center point of the laser image, including: Binarizing the plurality of preprocessed surface laser images, extracting a plurality of surface laser profiles, calculating the geometric centers of the plurality of surface laser profiles, and obtaining a plurality of surface laser center lines; In the dynamic cable detection process, the pixel coordinates of the center points of the plurality of pre-processed surface laser images are extracted to obtain a plurality of pixel coordinates; The plurality of pixel coordinates are determined as the pixel coordinates of the center point of the laser image, and the laser image center points of the plurality of preprocessed surface laser images are determined.

4. The system according to claim 1, characterized in that The system further comprises: A roundness compensation area is provided after the dynamic cable detection area; When the roundness evaluation result does not satisfy the roundness constraint, extracting cabling parameters of the dynamic cable to be detected, wherein the cabling parameters include twisting parameters and filling parameters; Compensation extrusion parameters are generated according to the cabling parameters and the roundness constraint, and roundness compensation is performed on the dynamic cable to be detected.

5. The system according to claim 4, characterized in that Generating compensating extrusion parameters according to the cabling parameters and the roundness constraint includes: Collecting samples of the dynamic cable to be detected to obtain a cabling parameter sample set and a dynamic cable sample set; Performing a roundness mark on the dynamic cable sample set according to the roundness constraint, and constructing a compensation extrusion parameter sample set through the roundness mark and the cabling parameter sample set; Constructing a roundness compensator based on the cabling parameter sample set, the roundness constraint, and the compensation extrusion parameter sample set; The cabling parameters and the roundness constraint are input into a roundness compensator to generate compensation extrusion parameters.

6. The system according to claim 4, characterized in that The system further comprises: Determining a value range of the cabling parameter according to the roundness constraint, and constructing a parameter optimization space based on the value range; When the roundness evaluation result does not satisfy the roundness constraint, generating a roundness deviation according to the roundness evaluation result; Based on the cabling parameters of the dynamic cable to be detected, parameter optimization is performed in the parameter optimization space according to the roundness deviation to obtain optimized cabling parameters; The optimized cabling parameters are applied to a cabling control area to perform dynamic cabling roundness control, wherein the cabling control area is before the dynamic cabling detection area.

7. The system according to claim 6, characterized in that Obtain optimized cabling parameters, including: The parameter optimization is a segmented backtracking optimization of parameters; Determining the cabling parameters of the dynamic cable to be detected as initial cabling parameters, and determining the twisting parameters and the filling parameters as initial twisting parameters and initial filling parameters respectively; The initial twisting parameters are locked, and the initial filling parameters are optimized in the parameter optimization space to obtain the filling parameters to be tested; The filling parameters to be tested are locked, and the initial twisting parameters are optimized in the parameter optimization space to obtain the twisting parameters to be tested; Continue to iterate and optimize the twisting parameters to be tested and the filling parameters to be tested to obtain optimized cabling parameters.

8. The system according to claim 1, characterized in that The system further comprises: Setting a secondary detection area in the roundness compensation area based on the dynamic cable detection area; After the roundness compensation is performed on the dynamic cable to be tested, the dynamic cable to be tested is transferred to the secondary testing area, and a secondary roundness test is performed on the dynamic cable to obtain a compensated roundness evaluation result; The dynamic cable to be detected is marked according to the compensation roundness evaluation result.

9. A submarine cable, comprising: A wire core, a metal armor layer and an outer sheath, wherein the outer sheath is made by real-time monitoring of roundness using the dynamic cable shape recognition system as described in any one of claims 1 to 8.

10. The submarine cable according to claim 9, wherein the core comprises a copper conductor, a metal shielding layer and an inner sheath from inside to outside, characterized in that: The inner sheath is manufactured by real-time monitoring of roundness using the dynamic cable shape recognition system as described in any one of claims 1 to 8.

Images