A method and system for intelligent management of wire classification and hanging of a wire hanging rack

Abstract

The application discloses a kind of wire hanging frame wire classification hanging intelligent management method and system, it is related to wire hanging management field, comprising: to the wire image is three-dimensional reconstruction processing, judge the similarity between wire image and standard wire, analyze wire surface crack;Wire surface crack is combined with wire physical property parameter, simulate the nonlinear deformation path of wire under different current intensity, obtain the heating deformation process of wire;According to heating deformation process, establish linear influence judgment model, obtain the causal relationship between wire deformation and current intensity;Based on causal relationship, evaluate the crack propagation trend of wire material, predict the required duration of wire insulation pyrolysis due to deformation under the premise of target current;Based on the required duration prediction result, the wire is classified, and the matching of the wire and the hanging position is realized.The application can assist field operation personnel to optimize hanging according to risk level, improve hanging safety.

Description

Technical Field

[0001] This invention relates to the field of wire hanging management, and in particular to an intelligent management method and system for classifying and hanging wires on a wire hanging rack. Background Technology

[0002] A conductor is a long, strip-shaped conductor used to transmit electrical energy or signals. It is usually made of highly conductive metal materials (such as copper, aluminum, etc.) and is covered with an insulation layer, shielding layer, or protective layer depending on the usage environment and functional requirements. A conductor rack is a support and positioning device used for the orderly storage, classification management, and safe fixing of multiple conductors. It is usually made of metal or high-strength engineering plastics and has properties such as corrosion resistance, insulation, fire resistance, and aging resistance. Its structural design can adapt to conductors of different diameters and for different purposes.

[0003] The intelligent management of wire classification and placement using cable hangers transforms the traditional, extensive management model that relies on manual experience and memory into a refined management model that is digital, automated, and intelligent. It can endow cable hangers and wires with intelligence through Internet of Things technology, and combined with powerful back-end management capabilities, it can achieve precise binding of wire information, thereby improving efficiency, ensuring quality, reducing costs, and optimizing management.

[0004] However, existing intelligent management methods for classifying and hanging wires on cable trays only classify and hang them according to parameters such as the purpose, diameter, and length of the wires. They cannot identify wire defects during the management process or reflect wear marks based on defect information. This can lead to wires with serious pyrolysis risks being misjudged as low-risk and hung in a conventional manner. This results in mismatches between high-risk wires and critical wiring locations in the cable hanging layout, increasing operational risks and the probability of failure. Consequently, it limits the predictability and graded response capability of wire status.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] To address the aforementioned problems, this invention proposes an intelligent management method and system for classifying and hanging wires on a wire rack, thereby achieving the goal of rationally managing the placement of wires.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides an intelligent management method for classifying and hanging wires on a wire rack, comprising:

[0009] S1. Acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, determine the similarity between the conductor images and standard conductors, and analyze surface cracks on the conductors.

[0010] S2. Combine the surface cracks of the conductor with the physical properties of the conductor to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the heating deformation process of the conductor.

[0011] S3. Establish a linear influence judgment model based on the heating deformation process and the current intensity, analyze the influence of current intensity in combination with resistance characteristics, and obtain the causal relationship between conductor deformation and current intensity.

[0012] S4. Based on causal relationships and critical duration index, assess the crack propagation trend of conductor materials and predict the time required for insulation pyrolysis to occur in the conductor due to deformation under the premise of target current.

[0013] S5. Based on the predicted duration and time classification criteria, classify the conductors and communicate with the positioning sensor at the hanging frame to match the conductors with the hanging position.

[0014] Preferably, conductor images are acquired, and structured reconstruction techniques are used to perform three-dimensional reconstruction on the conductor images. Based on the processing results, the similarity between the conductor images and standard conductors is determined, and the analysis of surface cracks on the conductors includes:

[0015] S11. Based on the image acquisition device, the wire image of the wire to be classified and hung is acquired, and the noise and redundant information in the wire image is removed by the scale reconstruction technology to complete the structure-preserving denoising of the crack area and obtain the denoised wire image. At the same time, the physical size mapping relationship of each pixel in the denoised wire image is extracted as the scale information.

[0016] S12. The scale information is processed by taking the denoised traverse image as input and the structure is unified. The edge features of the denoised traverse image and the standard traverse image are extracted based on the clustering algorithm and the edge features are fused.

[0017] S13. Based on the edge feature fusion processing results, the weighted structural similarity algorithm is used to calculate the similarity between the denoised wire image and the standard wire image, and the defect status of the wire to be classified is judged according to the similarity.

[0018] S14. Based on the judgment result of the defect status, use contour segmentation technology to detect and statistically analyze the feature parameters of the wires to be classified and hung with defects, and obtain the surface crack status of the wires to be classified and hung.

[0019] Preferably, based on the edge feature fusion processing results, a weighted structural similarity algorithm is used to calculate the similarity between the denoised conductor image and the standard conductor image. The defect status of the conductor to be classified and placed is determined based on the similarity, including:

[0020] S131. Based on the edge feature fusion processing results, the gradient magnitude of the denoised guide image and the standard guide image is calculated using the edge detection operator, and two gradient thresholds with different values ​​are set according to the target requirements.

[0021] S132. Using the gradient threshold as the criterion, and setting the gradient magnitude that meets the gradient threshold requirement as the texture region of the denoised guide image and the standard guide image, solve for the structural similarity of the texture region.

[0022] S133. Perform weighted normalization on the structural similarity, and obtain the similarity detection result between the denoised wire image and the standard wire image based on the processing result. When the similarity detection result is less than 100%, it indicates that there are defects on the surface of the wire to be classified and hung.

[0023] Preferably, based on the defect status assessment result, contour segmentation technology is used to detect and statistically analyze the surface crack status of the wires to be classified and hung, which includes:

[0024] S141. Select a denoised conductor image with surface defects, and use the network combining the residual network and the semantic pyramid as the backbone network for feature extraction to process the denoised conductor image to obtain the conductor feature map.

[0025] S142. Input the feature map of the conductor into the region candidate network for binary classification, obtain the conductor and the defect background, select the defect background to generate the region of interest, and obtain a feature map of a fixed size.

[0026] S143. Input the feature map into the segmentation mask generation network to obtain a mask with the same shape and size as the defect. Use the region classifier to perform pixel-level coordinate alignment of the region of interest to accurately identify the defect.

[0027] S144. Based on the recognition results, obtain a binary image containing the defect extraction results and the same shape and size as the defect, and analyze the binary image to obtain the surface crack status of the wire to be classified and hung.

[0028] Preferably, the surface cracks of the conductor are combined with the conductor's physical properties to construct a physical field simulation model, simulating the nonlinear deformation path of the conductor under different current intensities, and obtaining the conductor's heating deformation process, including:

[0029] S21. High-resolution imaging of cracks on the conductor surface is performed and a continuous and differentiable crack phase field description is generated through morphology reconstruction. Deformation-dependent damage evolution parameters are calibrated based on Bayesian inversion.

[0030] S22. Based on the damage evolution parameters and crack phase field description, a physical field simulation model is constructed, and a parameter set for joint excitation of current amplitude and harmonic spectrum is designed to simulate the nonlinear deformation path of the conductor.

[0031] S23. Using insulation layer pyrolysis as a coupled model, and combining nonlinear deformation path to update conductor insulation thermal conductivity and interface bonding strength, analyze the scale displacement and curvature history from microcrack propagation to flexure.

[0032] S24. Based on the topological evolution of the damaged line of the output conductor over time according to the scale displacement and curvature history, identify the path bifurcation and critical point caused by the sudden change of current, and obtain the heating deformation process of the conductor.

[0033] Preferably, based on causality and critical duration index, the crack propagation trend of the conductor material is evaluated, and the time required for insulation pyrolysis due to deformation under the target current is predicted, including:

[0034] S41. Using neural networks and causal relationships, construct scale-based causal chains, and define the causal dependency graph between crack propagation, deformation accumulation and pyrolysis evolution with the three field variables of thermal crack as state nodes.

[0035] S42. Define the critical duration index based on the causal dependency graph, compare the critical duration index with the empirical threshold, and construct a lifetime prediction model using the critical integral method after the insulation pyrolysis triggering condition is reached.

[0036] S43. The stress response of the conductor under the target current and the interface degradation factor are linked and embedded into the life prediction model to reflect the multi-field feedback effect, and the required time when the conductor experiences insulation pyrolysis is obtained.

[0037] Secondly, the present invention also provides an intelligent management system for classifying and hanging wires on a wire rack, the system comprising:

[0038] The conductor surface crack analysis module is used to acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, the similarity between the conductor image and the standard conductor is determined, and the surface cracks of the conductor are analyzed.

[0039] The conductor heating deformation analysis module is used to combine surface cracks on the conductor with the conductor's physical properties to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the conductor's heating deformation process.

[0040] The conductor current causal analysis module is used to establish a linear influence judgment model based on the heating deformation process and the current intensity, and to analyze the influence of current intensity in combination with resistance characteristics to obtain the causal relationship between conductor deformation and current intensity.

[0041] The insulation pyrolysis time prediction module is used to assess the crack propagation trend of conductor materials based on causal relationships and critical time index, and predict the time required for insulation pyrolysis to occur in the conductor due to deformation under the premise of target current.

[0042] The conductor hanging position matching module is used to classify conductors based on the required duration prediction results and time classification criteria, and communicate the classification results with the positioning sensor at the hanging frame to achieve matching between the conductor and the hanging position.

[0043] The beneficial effects of this invention are as follows:

[0044] 1. This invention introduces image reconstruction, physical property simulation, multi-causal correlation and dynamic classification mechanisms to achieve full-link perception and intelligent assessment of crack evolution, thermal deformation and insulation failure in the early stage of conductor life cycle. It achieves high-precision identification of conductor status and crack feature extraction. By accurately binding conductor risk level with the position of the hanging frame, it not only realizes intelligent linkage between the physical state and spatial position of the line, but also assists on-site operation and maintenance personnel to optimize the hanging according to the risk level, thereby improving the hanging safety, maintenance efficiency and intelligence level.

[0045] 2. This invention improves the clarity and geometric accuracy of conductor images through denoising and scale extraction. By using edge feature clustering and fusion, it not only preserves the key geometric changes in the conductor outline, but also enhances the ability to identify details such as cracks and scratches in local abnormal structures. It achieves quantitative output of crack structural indicators, improves the credibility of defect identification, and provides accurate geometric input for subsequent physical simulation and deformation prediction.

[0046] 3. This invention couples surface cracks with material properties to construct a physical field simulation model that can perceive crack evolution and thermodynamic response, realizing dynamic simulation of the nonlinear deformation path of the conductor under different current intensities. This not only achieves a true reproduction of the conductor's thermo-mechanical coupling behavior, but also provides a dynamic basic model for further risk prediction and life assessment. Based on the life results, the risk level of the conductor is output and precisely bound to the position of the wire hanger, realizing the rationality of conductor hanging position management. Attached Figure Description

[0047] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0048] Figure 1 This is a flowchart of an intelligent management method for classifying and hanging wires on a wire rack according to an embodiment of the present invention;

[0049] Figure 2This is a schematic diagram of a wire classification and placement intelligent management system according to an embodiment of the present invention.

[0050] In the picture:

[0051] 1. Conductor surface crack analysis module; 2. Conductor heating deformation analysis module; 3. Conductor current causal analysis module; 4. Insulation pyrolysis duration prediction module; 5. Conductor hanging position matching module. Detailed Implementation

[0052] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0053] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0054] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0055] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0056] Please see Figure 1 This invention provides an intelligent management method for classifying and hanging wires on a wire rack, comprising:

[0057] S1. Acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, determine the similarity between the conductor images and standard conductors, and analyze surface cracks on the conductors.

[0058] In this embodiment, conductor images are acquired, and structured processing techniques are used to perform three-dimensional reconstruction of the conductor images. Based on the processing results, the similarity between the conductor images and standard conductors is determined, and the analysis of surface cracks on the conductors includes:

[0059] S11. Based on the image acquisition device, the wire image of the wire to be classified and hung is acquired, and the noise and redundant information in the wire image is removed by the scale reconstruction technology to complete the structure-preserving denoising of the crack area and obtain the denoised wire image. At the same time, the physical size mapping relationship of each pixel in the denoised wire image is extracted as the scale information.

[0060] S12. The scale information is processed by taking the denoised traverse image as input and the structure is unified. The edge features of the denoised traverse image and the standard traverse image are extracted based on the clustering algorithm and the edge features are fused.

[0061] S13. Based on the edge feature fusion processing results, the weighted structural similarity algorithm is used to calculate the similarity between the denoised wire image and the standard wire image, and the defect status of the wire to be classified is judged according to the similarity.

[0062] S14. Based on the judgment result of the defect status, use contour segmentation technology to detect and statistically analyze the feature parameters of the wires to be classified and hung with defects, and obtain the surface crack status of the wires to be classified and hung.

[0063] It should be explained that high-resolution image acquisition equipment can be used in the process of fusing edge features, and scale reconstruction technology can be used for preprocessing. This technology is based on multi-scale analysis (such as wavelet transform or pyramid filtering) to remove noise and redundant information in the image, achieving a balance between image smoothing and boundary sharpening. This completes the structure-preserving denoising of local areas (especially crack edges), which can preserve the details of small cracks and avoid information loss caused by excessive smoothing. At the same time, the scale information of the conductor image is extracted, that is, the physical size mapping relationship corresponding to each pixel in the conductor image. For example, the conversion coefficient between image pixels and millimeters is obtained by calibrating a reference ruler (such as 1 pixel corresponding to 0.02 mm), which provides a basis for the accurate calculation of subsequent structural parameters such as crack length and width.

[0064] The denoised guideline image is used as input. Spatial scale information is preserved through scale reconstruction for structural unification. Edge pixels are then classified and extracted using clustering algorithms (such as K-means or DBSCAN) to distinguish key regions such as guideline and background, crack edges and normal edges. Complete closure of the guideline contour is achieved through edge aggregation and difference reconstruction. Simultaneously, edge features (such as contour curvature, edge gradient distribution, and texture changes) extracted from the denoised guideline image and the standard guideline image are fused. This fusion achieves feature consistency through graph structure matching or edge descriptor alignment. Sexual calibration forms a unified edge representation for structural similarity calculation, providing a complete foundation for subsequent use of weighted structural similarity algorithms (such as SSIM, GSSIM, or deep feature structure embedding algorithms). Its advantage lies in eliminating structural deviations caused by imaging angle, illumination, and slight deformation, making similarity assessment more robust and generalizable. This not only improves the clarity and geometric fidelity of image input data, but also opens up a channel for conductor geometric structure analysis and defect pattern recognition at the edge feature extraction and fusion processing level, providing an accurate and reliable data foundation for subsequent judgment of defect status and statistical modeling of cracks.

[0065] Among them, based on the edge feature fusion processing results, a weighted structural similarity algorithm is used to calculate the similarity between the denoised guide wire image and the standard guide wire image. The defect status of the guide wire to be classified is determined according to the similarity, including:

[0066] S131. Based on the edge feature fusion processing results, the gradient magnitude of the denoised guide image and the standard guide image is calculated using the edge detection operator, and two gradient thresholds with different values ​​are set according to the target requirements.

[0067] S132. Using the gradient threshold as the criterion, and setting the gradient magnitude that meets the gradient threshold requirement as the texture region of the denoised guide image and the standard guide image, solve for the structural similarity of the texture region.

[0068] S133. Perform weighted normalization on the structural similarity, and obtain the similarity detection result between the denoised wire image and the standard wire image based on the processing result. When the similarity detection result is less than 100%, it indicates that there are defects on the surface of the wire to be classified and hung.

[0069] Specifically, based on the defect status assessment results, contour segmentation technology is used to detect and statistically analyze the surface crack status of the wires to be classified and hung, which includes:

[0070] S141. Select a denoised conductor image with surface defects, and use the network combining the residual network and the semantic pyramid as the backbone network for feature extraction to process the denoised conductor image to obtain the conductor feature map.

[0071] S142. Input the feature map of the conductor into the region candidate network for binary classification, obtain the conductor and the defect background, select the defect background to generate the region of interest, and obtain a feature map of a fixed size.

[0072] S143. Input the feature map into the segmentation mask generation network to obtain a mask with the same shape and size as the defect. Use the region classifier to perform pixel-level coordinate alignment of the region of interest to accurately identify the defect.

[0073] S144. Based on the recognition results, obtain a binary image containing the defect extraction results and the same shape and size as the defect, and analyze the binary image to obtain the surface crack status of the wire to be classified and hung.

[0074] Specifically, based on the recognition results, a binary image containing the defect extraction results and the same shape and size as the defect is obtained. The binary image is then analyzed to obtain the surface crack state of the wire to be classified, including:

[0075] S1441. Based on the recognition results, obtain the defect extraction results and a binary image with the same shape and size as the defect corresponding to the denoised wire image with surface defects, and perform erosion and dilation processing on the binary image.

[0076] S1442. Based on the erosion and dilation processing results, perform edge detection on the binary image to extract the edge features of defects, and use the ellipse fitting algorithm to calculate the number of pixels corresponding to the major and minor axes of defects.

[0077] S1443. Calculate the number of pixels and obtain the major axis length, minor axis length, perimeter and area of ​​each defect area as shape feature parameters of the surface cracks of the wire to be classified.

[0078] It needs to be explained that the core of the wire crack recognition process includes two stages: structural similarity judgment and semantic segmentation recognition. It integrates technologies such as image gradient processing, deep neural networks, region candidate recognition, and shape feature extraction. Edge detection operators such as Sobel or Canny are used to calculate the gradient magnitude map of the corresponding positions of the denoised wire image and the standard wire image. Then, two different gradient thresholds are set according to the specific recognition task requirements (such as a low threshold of 20 and a high threshold of 80) to accommodate edge responses of different intensities in the image. This ensures that the gradient response area of ​​the image can cover potential crack features while avoiding high-frequency noise interference. The gradient magnitude area that meets the threshold condition is defined as the texture area of ​​the image, and structural similarity is calculated for these areas (such as using the SSIM formula). At the same time, the similarity matrix is ​​weighted and normalized (such as by weighting based on the area and gradient energy of each texture block) to obtain a similarity score under a unified standard. If the score is less than 100% (such as 92.4% in actual measurement), the wire is determined to have a defect and enters the next round of recognition process. If it is equal to 100%, it is directly hung up.

[0079] For denoised conductor images with defects, fine crack extraction is performed. A backbone feature extraction network is formed by integrating a residual network (such as ResNet50) and a semantic pyramid network (such as FPN or PSPNet) to extract features at the defective image level. The output is a conductor feature map that preserves both resolution and semantic information. This feature map is then input into a region candidate network (such as RPN or Selective Search) for binary classification of foreground and background, extracting candidate regions between the conductor body and the crack background, and generating a fixed-size region of interest (e.g., 128×128 pixels). This provides standard input for subsequent fine mask generation. These feature maps are then input into the Mask. Pixel-level segmentation masks are generated in R-CNN or FCN structures, and fine localization is achieved through region classifiers to solve the positional error problem caused by edge blurring, thus realizing accurate mask localization of defects. The recognition results are output as a binary image and a defect shape extraction image, where 1 represents defect pixels and 0 represents background. The binary image is subjected to erosion and dilation processing to remove isolated points and connect crack edges. Edge detection (such as using Canny) is performed on the processed image to extract crack contours. Ellipse fitting algorithms (such as least squares ellipse fitting) are used to extract the number of pixels along the major and minor axes of each crack. Finally, the true size of the crack is calculated through pixel conversion scale parameters, including the length of the major axis, the length of the minor axis, the perimeter of the contour, and the crack area. These parameters can be used as shape feature vectors to participate in the input of subsequent wire classification, risk assessment, and pyrolysis prediction models, thereby realizing multi-level processing from structural similarity judgment to pixel-level crack quantification and improving the accuracy of crack recognition.

[0080] S2. Combine the surface cracks of the conductor with the physical parameters of the conductor to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the heating deformation process of the conductor.

[0081] In one embodiment, surface cracks in the conductor are combined with the conductor's physical properties to construct a physical field simulation model. This model simulates the nonlinear deformation path of the conductor under different current intensities, and obtains the conductor's heating deformation process, including:

[0082] S21. High-resolution imaging of cracks on the conductor surface is performed and a continuous and differentiable crack phase field description is generated through morphology reconstruction. Deformation-dependent damage evolution parameters are calibrated based on Bayesian inversion.

[0083] S22. Based on the damage evolution parameters and crack phase field description, a physical field simulation model is constructed, and a parameter set for joint excitation of current amplitude and harmonic spectrum is designed to simulate the nonlinear deformation path of the conductor.

[0084] S23. Using insulation layer pyrolysis as a coupled model, and combining nonlinear deformation path to update conductor insulation thermal conductivity and interface bonding strength, analyze the scale displacement and curvature history from microcrack propagation to flexure.

[0085] S24. Based on the topological evolution of the damaged line of the output conductor over time according to the scale displacement and curvature history, identify the path bifurcation and critical point caused by the sudden change of current, and obtain the heating deformation process of the conductor.

[0086] It needs to be explained that the process of simulating the nonlinear deformation path of the conductor is achieved by coupling the surface crack with physical property parameters to model, thus realizing a precise correlation between the material's microstructure and macroscopic electrothermal response. High-resolution imaging of the crack region on the conductor surface is then performed, for example, using laser confocal microscopy or white light interferometry with a resolution of 10 μm, to obtain the fine three-dimensional morphology of the crack. Then, phase-field modeling methods are used to reconstruct it, converting the original image into a continuously differentiable crack phase-field variable. φ ,in φ =1 indicates a crack-free region. φ =0 indicates a completely broken region.

[0087] To accurately reflect the crack growth behavior of a conductor during deformation evolution under current, damage evolution parameters are precisely calibrated. A Bayesian inversion method is introduced, using a pre-defined prior distribution and posterior inference based on finite experimental samples to determine the reliability interval of the parameter distribution. This not only improves the model's reliability but also enhances its adaptability to sample variations. Simultaneously, the constructed physical field simulation model incorporates phase field variables... φ With stress field σ Current density J and temperature field TCoupling is performed to form a nonlinear response model involving multiple physics fields. A set of multiple excitation parameters is designed, consisting of current amplitude (e.g., 0.5A~10A) and harmonic spectrum (e.g., 50Hz fundamental frequency superimposed with the 3rd and 5th harmonics). This simulates the combined effect of electrothermal stress on the conductor during actual operation. Such joint excitation inputs can reveal the role of frequent fluctuations in low-frequency current on the turning point of the microcrack-induced deformation path and the initiation of interface pyrolysis. During the simulation, the conductor is solved step by step by subdividing the mesh (e.g., 0.01mm mesh scale), and the deformation path is output. u ( x , t ) and temperature rise trajectory T ( x , t This allows for the reproduction of complex phenomena such as nonlinear bending, stress concentration, and thermal softening of conductors under real energized conditions, thereby achieving a high degree of coupling between crack evolution parameters and thermoelectric loads, enabling the simulation results to have sensitivity to crack distribution and response to current changes.

[0088] Specifically, using insulation pyrolysis as a coupled model, and combining nonlinear deformation paths to update the thermal conductivity and interfacial bonding strength of the conductor insulation, the analysis of the scale displacement and curvature history from microcrack propagation to flexure includes:

[0089] S231. Introduce state variables and volume fractions with insulation layer pyrolysis as the main coupling field, establish a bulk mapping between insulation layer pyrolysis and interface adhesion parameters, and generate a coupling model based on the bulk mapping results.

[0090] S232. Couple the coupled model with the nonlinear deformation path isotopic and isomorphic to establish a meshless node model of the conductor crack, and divide the nodes into regular nodes, step propagation nodes and crack tip propagation nodes.

[0091] S233. Based on the node division results, the approximate function is calculated using the moving least squares method to obtain the displacement field and stress field of the conductor insulation and heat conduction, and the interface bonding strength is solved using the interaction integral method.

[0092] S234. Based on the displacement field, stress field and interface bonding strength, analyze the insulation stiffness of the conductor, generate the size displacement path of the conductor crack, calculate the curvature corresponding to the size displacement path, and obtain the curvature history.

[0093] It needs to be explained that the key to the high-fidelity physical simulation modeling mechanism that finely couples the pyrolysis behavior of the insulation layer with the nonlinear deformation process of the conductor lies not only in considering the thermodynamic behavior of the conductor material itself, but also in deeply introducing the dynamic evolution path of interfacial pyrolysis and insulation performance degradation. The insulation layer pyrolysis process is the main coupling field, and state variables (such as the degree of pyrolysis) are introduced. θ ) and volume fraction variables (such as insulation residual rate) αDescribe the chemical change behavior of insulating materials under temperature rise conditions and correlate it with interfacial bond strength. σ b Thermal conductivity k A volumetric mapping relationship is established between isophysical parameters to form a structured coupled model, clarifying how the pyrolysis process substantially weakens bonding and thermal conductivity. This coupled model is homotopically and isomorphically treated with the nonlinear deformation path of the conductor, maintaining the correspondence between the time axis and spatial topology, ensuring synchronous evolution of pyrolysis and deformation. Simultaneously, a meshless node model is constructed to avoid the mesh distortion problem in the crack tip region of traditional finite element methods. The conductor nodes are divided into regular nodes (for continuous regions), step propagation nodes (to capture crack abrupt changes), and crack tip propagation nodes (for analytical stress singularities). This classification helps improve the crack propagation path... Analytical accuracy: A local approximation function is constructed for each type of node using the moving least squares method to form a smooth displacement and stress field. Then, the change in bond strength at the crack interface is solved using the interaction integral method (such as J integral or M line integral). Based on the solved displacement and stress field data, the equivalent stiffness of the conductor insulation layer is calculated in reverse. The crack size change and flexural deformation at each time point during the crack propagation along the path are tracked to generate the size displacement path of the conductor crack (i.e., the spatial trajectory of the crack over time). Then, the corresponding curvature history is solved using the curvature calculation formula to obtain the deformation curve of the entire process from microscopic crack initiation to macroscopic flexural deformation.

[0094] Furthermore, it can accurately characterize the material softening, interface debonding, and crack deflection phenomena caused by pyrolysis. Its key advantage lies in the high integration of structure, thermal field, and interface mechanics. It can not only predict how cracks evolve to the material's ultimate state, but also identify thermally induced path abrupt changes and potential failure inflection points, providing strong simulation support and quantitative indicators for high-risk zone identification, remaining lifetime estimation, and intelligent classification and placement of conductors.

[0095] S3. Establish a linear influence judgment model based on the heating deformation process and the current intensity, and analyze the influence of current intensity in combination with resistance characteristics to obtain the causal relationship between conductor deformation and current intensity.

[0096] In one embodiment, a linear influence judgment model is established based on the heating deformation process and the current intensity. The influence of current intensity is analyzed in conjunction with resistance characteristics to obtain the causal relationship between conductor deformation and current intensity. This includes: extracting simulation results of conductor heating deformation; based on the acquired nonlinear deformation path data, extracting characteristic parameters such as maximum flexural displacement per unit time, equivalent thermally induced deformation, and the rate of change of insulation modulus induced by pyrolysis (dE / dt); combining the simulation input current intensity I (e.g., results at multiple levels such as 0.5A, 2A, 5A, 10A), constructing a mapping sample between deformation and current intensity in the data dimension; using material resistance characteristics to correlate the conductor's resistance change rate with temperature rise, and substituting it into the Joule heating equation to establish the energy flow path between thermally induced stress and current intensity. This not only captures the influence of current on the conductor's heating rate but also considers the nonlinear growth trend of heat and work after bidirectional feedback with material resistance characteristics and temperature. Based on this, a function of the form Δ is constructed through least squares regression or principal component linear modeling. u = a · I + b · R + c ,in a , b These are the weighting coefficients. c For constant terms, I Indicates the current intensity. R Representing the material's resistance characteristics, this model can infer and predict the deformation trend of a conductor after inputting any current value. At the same time, the Granger causal analysis method is introduced for causal verification. That is, by judging whether adding the current current historical value sequence significantly improves the prediction accuracy of the conductor's future deformation, if the result is significant (e.g., <0.05), it can be determined that "current → deformation" is a strong causal path, providing a reliable mechanism for subsequent dynamic adjustment of the conductor's hanging safety area and the formulation of load balancing strategies based on current.

[0097] S4. Based on causal relationships and critical duration index, assess the crack propagation trend of conductor materials and predict the time required for insulation pyrolysis due to deformation under the target current.

[0098] In one embodiment, based on causality and a critical duration index, the crack propagation trend of the conductor material is assessed, and the time required for insulation pyrolysis due to deformation under a target current is predicted, including:

[0099] S41. Using neural networks and causal relationships, construct scale-based causal chains, and define the causal dependency graph between crack propagation, deformation accumulation and pyrolysis evolution with the three field variables of thermal crack as state nodes.

[0100] S42. Define the critical duration index based on the causal dependency graph, compare the critical duration index with the empirical threshold, and construct a lifetime prediction model using the critical integral method after the insulation pyrolysis triggering condition is reached.

[0101] S43. The stress response of the conductor under the target current and the interface degradation factor are linked and embedded into the life prediction model to reflect the multi-field feedback effect, and the required time when the conductor experiences insulation pyrolysis is obtained.

[0102] It needs to be explained that in predicting the time required for insulation pyrolysis in the conductor, the three states (temperature field, temperature field, and lifespan) are considered using a "causal + threshold lifetime" approach. x Mechanical field y ,crack z Closed-loop modeling on the time axis: Constructing a structural causal model (SCM) constrained by a neural network, with current as the exogenous driver, entering the thermal field nodes via Joule heating, and acting on the damage rate and phase field evolution through thermo-elastic coupling. The neural network part adopts a "physical prior + causal masking" approach: using a graph neural network / temporal Transformer with sparse gating to model { x , y , z The evolutionary operator} is used for regression, but only with causality is allowed. Figure 1 The edge weights are non-zero, and the loss function is constrained by Arrhenius-type pyrolysis prior and thermoelastic constitutive prior. A "critical duration index" is defined to measure the time margin from the current state to triggering insulation pyrolysis. The risk rate of pyrolysis triggering is set, and a critical duration index is generated. When the critical duration index or any hard threshold is broken, it is considered to enter the pyrolysis triggering domain. In order to obtain the remaining duration, the critical integral lifetime model is used to accumulate damage to the multi-source load. The stress response, temperature trajectory, interface degradation factor dynamics and accumulated damage are combined under the target current scenario. The duration that satisfies the integral equation is the duration required for insulation pyrolysis induced by deformation under current.

[0103] S5. Based on the predicted duration and time classification criteria, classify the conductors and communicate with the positioning sensor at the hanging frame to match the conductors with the hanging position.

[0104] In one embodiment, the conductor is classified based on the required duration prediction result and the time classification standard, and the classification result is communicated with the positioning sensor at the hanging frame to achieve the matching of the conductor and the hanging position. This includes matching the required duration of insulation pyrolysis with the preset time classification standard, which can be set to three segments: for example, insulation pyrolysis time < 600 seconds is high risk, 600 ≤ insulation pyrolysis time < 1800 seconds is medium risk, and insulation pyrolysis time ≥ 1800 seconds is low risk. Alternatively, it can be adaptively divided according to the actual operating environment (such as external temperature and current fluctuations). The key function of this classification is to convert the continuous life prediction result into a discrete risk level, thereby providing a basis for the rule-based execution of subsequent hanging operations.

[0105] Based on the risk level of the conductors, communication access is established with the positioning sensors already deployed on the cable trays. These sensors use communication methods such as RFID, UWB, or NFC, and are pre-bound with a number and attribute tag for each hanging location (such as "edge cooling zone," "central high load zone," "redundant hanging point zone," etc.), and periodically upload the current environmental status to the scheduling system. After communication is completed, each conductor and its hanging location is mapped one-to-one using rule matching algorithms or optimized scheduling strategies (such as genetic algorithms, local search, or Boolean satisfiability algorithms like SAT). This ensures that high-risk conductors are assigned to areas with lower temperatures or lower current loads, while low-risk conductors can be used to preferentially occupy main hanging points, thereby achieving optimal matching between conductor risk levels and the physical characteristics of hanging locations. After matching is completed, the classification results are synchronized to the execution layer, instructing operators or robots to complete precise hanging, while recording the conductor ID and hanging point ID for full lifecycle tracking and management. This enables conductor maintenance to shift from static planning to dynamic risk control, significantly improving the overall cabling system's operational safety and management efficiency.

[0106] Please see Figure 2 The present invention also provides an intelligent management system for classifying and hanging wires on a wire rack, the system comprising:

[0107] The conductor surface crack analysis module 1 is used to acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, the similarity between the conductor images and standard conductors is determined, and surface cracks of the conductors are analyzed.

[0108] The conductor heating deformation analysis module 2 is used to combine the surface cracks of the conductor with the conductor's physical property parameters to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the heating deformation process of the conductor.

[0109] The conductor current causal analysis module 3 is used to establish a linear influence judgment model based on the heating deformation process and the current intensity, and to analyze the influence of current intensity in combination with resistance characteristics to obtain the causal relationship between conductor deformation and current intensity.

[0110] The insulation pyrolysis time prediction module 4 is used to evaluate the crack propagation trend of the conductor material based on causal relationship and critical time index, and predict the time required for insulation pyrolysis of the conductor due to deformation under the premise of target current.

[0111] The conductor hanging position matching module 5 is used to classify the conductors based on the required duration prediction results and time classification criteria, and communicate the classification results with the positioning sensor at the hanging frame to achieve matching between the conductors and the hanging position.

[0112] The conductor surface crack analysis module 1, conductor heating deformation analysis module 2, conductor current causal analysis module 3, insulation pyrolysis duration prediction module 4, and conductor hanging position matching module 5 are connected in sequence.

[0113] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0114] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for intelligent management of wire classification and placement using a wire hanging rack, characterized in that, include: S1. Acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, determine the similarity between the conductor images and standard conductors, and analyze surface cracks on the conductors. S2. Combine the surface cracks of the conductor with the physical properties of the conductor to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the heating deformation process of the conductor. S3. Establish a linear influence judgment model based on the heating deformation process and the current intensity, analyze the influence of current intensity in combination with resistance characteristics, and obtain the causal relationship between conductor deformation and current intensity. S4. Based on causal relationships and critical duration index, assess the crack propagation trend of conductor materials and predict the time required for insulation pyrolysis to occur in the conductor due to deformation under the premise of target current. S5. Based on the predicted duration and time classification criteria, classify the conductors and communicate with the positioning sensor at the hanging frame to match the conductors with the hanging position.

2. The intelligent management method for classifying and hanging wires on a wire rack according to claim 1, characterized in that, The process involves acquiring conductor images and performing three-dimensional reconstruction using structured techniques. Based on the processing results, the similarity between the conductor images and standard conductors is determined. Analysis of surface cracks on the conductor includes: S11. Based on the image acquisition device, the wire image of the wire to be classified and hung is acquired, and the noise and redundant information in the wire image is removed by the scale reconstruction technology to complete the structure-preserving denoising of the crack area and obtain the denoised wire image. At the same time, the physical size mapping relationship of each pixel in the denoised wire image is extracted as the scale information. S12. The scale information is processed by taking the denoised traverse image as input and the structure is unified. The edge features of the denoised traverse image and the standard traverse image are extracted based on the clustering algorithm and the edge features are fused. S13. Based on the edge feature fusion processing results, the weighted structural similarity algorithm is used to calculate the similarity between the denoised wire image and the standard wire image, and the defect status of the wire to be classified is judged according to the similarity. S14. Based on the judgment result of the defect status, use contour segmentation technology to detect and statistically analyze the feature parameters of the wires to be classified and hung with defects, and obtain the surface crack status of the wires to be classified and hung.

3. The intelligent management method for classifying and hanging wires on a wire rack according to claim 2, characterized in that, The edge feature fusion processing result is used to calculate the similarity between the denoised guide wire image and the standard guide wire image using a weighted structural similarity algorithm. The defect status of the guide wire to be classified is determined based on the similarity, including: S131. Based on the edge feature fusion processing results, the gradient magnitude of the denoised guide image and the standard guide image is calculated using the edge detection operator, and two gradient thresholds with different values ​​are set according to the target requirements. S132. Using the gradient threshold as the criterion, and setting the gradient magnitude that meets the gradient threshold requirement as the texture region of the denoised guide image and the standard guide image, solve for the structural similarity of the texture region. S133. Perform weighted normalization on the structural similarity, and obtain the similarity detection result between the denoised wire image and the standard wire image based on the processing result. When the similarity detection result is less than 100%, it indicates that there are defects on the surface of the wire to be classified and hung.

4. The intelligent management method for classifying and hanging wires on a wire rack according to claim 3, characterized in that, Based on the defect status judgment result, the contour segmentation technology is used to detect and statistically analyze the surface crack status of the wires to be classified and hung based on the defect status, and obtain the surface crack status of the wires to be classified and hung based on the defect status judgment result. S141. Select a denoised conductor image with surface defects, and use the network combining the residual network and the semantic pyramid as the backbone network for feature extraction to process the denoised conductor image to obtain the conductor feature map. S142. Input the feature map of the conductor into the region candidate network for binary classification, obtain the conductor and the defect background, select the defect background to generate the region of interest, and obtain a feature map of a fixed size. S143. Input the feature map into the segmentation mask generation network to obtain a mask with the same shape and size as the defect. Use the region classifier to perform pixel-level coordinate alignment of the region of interest to accurately identify the defect. S144. Based on the recognition results, obtain a binary image containing the defect extraction results and the same shape and size as the defect, and analyze the binary image to obtain the surface crack status of the wire to be classified and hung.

5. The intelligent management method for classifying and hanging wires on a wire rack according to claim 4, characterized in that, The process of obtaining a binary image containing defect extraction results and the same shape and size as the defect based on the recognition results, and analyzing the binary image to obtain the surface crack status of the wire to be classified includes: S1441. Based on the recognition results, obtain the defect extraction results and a binary image with the same shape and size as the defect corresponding to the denoised wire image with surface defects, and perform erosion and dilation processing on the binary image. S1442. Based on the erosion and dilation processing results, perform edge detection on the binary image to extract the edge features of defects, and use the ellipse fitting algorithm to calculate the number of pixels corresponding to the major and minor axes of defects. S1443. Calculate the number of pixels and obtain the major axis length, minor axis length, perimeter and area of ​​each defect area as shape feature parameters of the surface cracks of the wire to be classified.

6. The intelligent management method for classifying and hanging wires on a wire rack according to claim 1, characterized in that, The process of combining surface cracks in the conductor with its physical properties to construct a physical field simulation model, simulating the nonlinear deformation path of the conductor under different current intensities, and obtaining the heating deformation process of the conductor includes: S21. High-resolution imaging of cracks on the conductor surface is performed and a continuous and differentiable crack phase field description is generated through morphology reconstruction. Deformation-dependent damage evolution parameters are calibrated based on Bayesian inversion. S22. Based on the damage evolution parameters and crack phase field description, a physical field simulation model is constructed, and a parameter set for joint excitation of current amplitude and harmonic spectrum is designed to simulate the nonlinear deformation path of the conductor. S23. Using insulation layer pyrolysis as a coupled model, and combining nonlinear deformation path to update conductor insulation thermal conductivity and interface bonding strength, analyze the scale displacement and curvature history from microcrack propagation to flexure. S24. Based on the topological evolution of the damaged line of the output conductor over time according to the scale displacement and curvature history, identify the path bifurcation and critical point caused by the sudden change of current, and obtain the heating deformation process of the conductor.

7. The intelligent management method for classifying and hanging wires on a wire rack according to claim 6, characterized in that, The method of using insulation pyrolysis as a coupled model, combining nonlinear deformation paths to update conductor insulation thermal conductivity and interfacial bonding strength, and analyzing the scale displacement and curvature history from microcrack propagation to flexure includes: S231. Introduce state variables and volume fractions with insulation layer pyrolysis as the main coupling field, establish a bulk mapping between insulation layer pyrolysis and interface adhesion parameters, and generate a coupling model based on the bulk mapping results. S232. Couple the coupled model with the nonlinear deformation path isotopic and isomorphic to establish a meshless node model of the conductor crack, and divide the nodes into regular nodes, step propagation nodes and crack tip propagation nodes. S233. Based on the node division results, the approximate function is calculated using the moving least squares method to obtain the displacement field and stress field of the conductor insulation and heat conduction, and the interface bonding strength is solved using the interaction integral method. S234. Based on the displacement field, stress field and interface bonding strength, analyze the insulation stiffness of the conductor, generate the size displacement path of the conductor crack, calculate the curvature corresponding to the size displacement path, and obtain the curvature history.

8. The intelligent management method for classifying and hanging wires on a wire rack according to claim 1, characterized in that, The method for assessing the crack propagation trend of conductor materials based on causality and critical duration index, and predicting the time required for insulation pyrolysis due to deformation under the target current, includes: S41. Using neural networks and causal relationships, construct scale-based causal chains, and define the causal dependency graph between crack propagation, deformation accumulation and pyrolysis evolution with the three field variables of thermal crack as state nodes. S42. Define the critical duration index based on the causal dependency graph, compare the critical duration index with the empirical threshold, and construct a lifetime prediction model using the critical integral method after the insulation pyrolysis triggering condition is reached. S43. The stress response of the conductor under the target current and the interface degradation factor are linked and embedded into the life prediction model to reflect the multi-field feedback effect, and the required time when the conductor experiences insulation pyrolysis is obtained.

9. A smart management system for classifying and hanging wires on a cable tray, used to implement the smart management method for classifying and hanging wires on a cable tray as described in any one of claims 1-8, characterized in that, The system includes: The conductor surface crack analysis module is used to acquire conductor images and perform three-dimensional reconstruction processing on the conductor images using structured technology. Based on the processing results, the similarity between the conductor image and the standard conductor is determined, and the surface cracks of the conductor are analyzed. The conductor heating deformation analysis module is used to combine surface cracks on the conductor with the conductor's physical properties to construct a physical field simulation model, simulate the nonlinear deformation path of the conductor under different current intensities, and obtain the conductor's heating deformation process. The conductor current causal analysis module is used to establish a linear influence judgment model based on the heating deformation process and the current intensity, and to analyze the influence of current intensity in combination with resistance characteristics to obtain the causal relationship between conductor deformation and current intensity. The insulation pyrolysis time prediction module is used to assess the crack propagation trend of conductor materials based on causal relationships and critical time index, and predict the time required for insulation pyrolysis to occur in the conductor due to deformation under the premise of target current. The conductor hanging position matching module is used to classify conductors based on the required duration prediction results and time classification criteria, and communicate the classification results with the positioning sensor at the hanging frame to achieve matching between the conductor and the hanging position.

Images