Real-time defect detection method and device based on carbon fiber winding and medium

Abstract

The embodiment of the invention provides a real-time defect detection method and device based on carbon fiber winding and a medium, and relates to the technical field of carbon fiber laying layer detection.The method comprises the steps that real-time environment data of carbon fiber winding are obtained, and multi-mode cooperative detection is conducted on the real-time environment data of carbon fiber winding, obtaining a carbon fiber real-time defect data set; performing multi-mode space-time synchronous calibration on the carbon fiber real-time defect data set to determine carbon fiber defect synchronous data; performing strategy-level fusion on the synchronous data of the carbon fiber defects to obtain winding state preliminary judgment data; based on the winding state preliminary judgment data, multi-modal weight update data are determined through carbon fiber winding process stage analysis; and according to the modal weight updating data, determining the real-time winding defect of the carbon fiber through carbon fiber defect type analysis. According to the method, the technical problem of poor anti-interference capability in a single parameter analysis process in carbon fiber winding detection is solved.

Description

Technical Field

[0001] This application relates to the technical field of carbon fiber ply detection, and particularly to a real-time defect detection method, device and medium based on carbon fiber winding. Background Art

[0002] Carbon fiber composite materials are formed by combining high-strength carbon fibers with a resin matrix to form lightweight and high-strength structural materials, which are widely used in aerospace (such as aircraft wing beams, satellite brackets), new energy vehicles (battery boxes, body structures), wind power (blade main beams), etc. Its core manufacturing processes include prepreg laying, lamination curing, and forming processing, among which the ply quality directly determines the performance of the final product.

[0003] As a key automation module for carbon fiber plying, the real-time detection of the ply during the working process by a carbon fiber winding robot determines the quality of the carbon fiber ply. Since in the prior art, non-contact detection such as visual detection and eddy current detection is usually used to analyze the ply state, the analysis of a single parameter may be affected by other environmental interferences, resulting in abnormal detection results and insufficient robustness of the carbon fiber detection system. Summary of the Invention

[0004] The embodiments of this application provide a real-time defect detection method, device and medium based on carbon fiber winding, which solves the technical problem of poor anti-interference ability in the single-parameter analysis process of carbon fiber winding detection.

[0005] In a first aspect, the embodiments of this application provide a real-time defect detection method based on carbon fiber winding, which is characterized in that the method includes: obtaining real-time environmental data of carbon fiber winding, and performing multi-modal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time carbon fiber defect data set; performing multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data set to determine synchronous carbon fiber defect data; performing strategy-level fusion on the synchronous carbon fiber defect data to obtain preliminary determination data of the winding state; based on the preliminary determination data of the winding state, determining updated multi-modal weight data through process stage analysis of carbon fiber winding; and determining real-time carbon fiber winding defects through carbon fiber defect type analysis according to the updated modal weight data.

[0006] In an implementation of the present application, the real-time environmental data of carbon fiber winding further includes: carbon fiber layup, local mechanical vibration data, and broadband acoustic emission signal; performing multi-modal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time carbon fiber defect data set, specifically including: performing high-frequency eddy current array inversion analysis on the carbon fiber layup to obtain carbon fiber layup state data; wherein, the carbon fiber layup state data includes: fiber arrangement direction, deep defect data; performing surface vibration response analysis on the local mechanical vibration data to obtain local mechanical impedance; wherein, the surface vibration response analysis includes: swept-frequency mechanical vibration processing, surface laser vibration measurement analysis; based on the broadband acoustic emission signal, determining the layup defect type parameter through acoustic fingerprint feature recognition; obtaining the real-time carbon fiber defect data set according to the carbon fiber layup state data, local mechanical impedance, and layup defect type parameter.

[0007] In an implementation of the present application, performing multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data set to determine carbon fiber defect synchronization data, specifically including: performing multi-modal spatial calibration on the real-time carbon fiber defect data, and performing feature coordinate transformation on the feature coordinates obtained by the multi-modal spatial calibration to obtain defect point spatial coordinates; wherein, the feature coordinate transformation includes: electrical marker point coordinate transformation, stiffness mutation area coordinate transformation, acoustic fingerprint positioning coordinate transformation; obtaining the data processing delay of the preset sensor, and through reverse delay compensation based on the data processing delay, obtaining multi-modal time synchronization data; determining the carbon fiber defect synchronization data according to the defect point spatial coordinates and the multi-modal time synchronization data.

[0008] In an implementation of the present application, performing strategy-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data, specifically including: performing basic confidence configuration analysis on the carbon fiber defect synchronization data to obtain the modal confidence corresponding to each mode; wherein, the modal confidence includes: eddy current mode confidence, mechanical impedance mode confidence, acoustic emission mode confidence; based on the modal confidence, determining the first evidence synthesis data through reliability weight correction; performing dynamic weight optimization on the first evidence synthesis data to obtain the second evidence synthesis data; obtaining the preliminary winding state determination data through modal conflict resolution analysis according to the second evidence synthesis data.

[0009] In an implementation of the present application, based on the preliminary winding state determination data, determining multi-modal weight update data through the process stage analysis of carbon fiber winding, specifically including: obtaining the current process stage, and based on the current process stage, obtaining multi-modal weight adjustment data through modal weight configuration; wherein, the current process stage includes: initial laying, lamination stage, emergency condition; determining the multi-modal weight update data through modal weight update according to the preliminary winding state determination data and the multi-modal weight adjustment data.

[0010] In one implementation of the present application, data is updated according to modal weights, and real-time winding defects of carbon fiber are determined through carbon fiber defect type analysis, specifically including: based on the updated data of modal weights, defect type weight data is determined through defect type-oriented weight assignment; dynamic weight adaptive processing is performed on the defect type weight data to determine multi-modal quality weight data; according to the multi-modal quality weight data, real-time winding defects of carbon fiber are determined through comprehensive confidence threshold judgment.

[0011] In one implementation of the present application, after real-time winding defects of carbon fiber are determined through carbon fiber defect type analysis according to the updated data of modal weights, the method further includes: dynamically adjusting the winding equipment parameters for the real-time winding defects of carbon fiber to determine equipment adjustment parameters; wherein, the equipment adjustment parameters include: pressure dynamic adjustment parameters, laying speed optimization parameters, and defect path avoidance path planning parameters; based on the equipment adjustment parameters, equipment closed-loop control is performed to complete the equipment adjustment corresponding to the real-time winding defects of carbon fiber.

[0012] In one implementation of the present application, dynamically adjusting the winding equipment parameters for the real-time winding defects of carbon fiber to determine equipment adjustment parameters specifically includes: constructing a pressure compensation relationship matrix, and based on the pressure compensation relationship matrix, through adaptive PID control analysis, obtaining pressure dynamic adjustment parameters; obtaining the density change data of the carbon fiber winding layer, and performing gradient descent optimization on the data coefficients of the density change data to obtain laying speed optimization parameters; obtaining the position of the defect area, and according to the position of the defect area, through B-spline curve optimization analysis, determining defect path avoidance path planning parameters.

[0013] In a second aspect, an embodiment of the present application further provides a real-time defect detection device based on carbon fiber winding, characterized in that the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: obtain real-time environmental data of carbon fiber winding, and perform multi-modal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time defect data group of carbon fiber; perform multi-modal spatio-temporal synchronization calibration on the real-time defect data group of carbon fiber to determine carbon fiber defect synchronization data; perform strategy-level fusion on the carbon fiber defect synchronization data to obtain preliminary judgment data of the winding state; based on the preliminary judgment data of the winding state, determine updated data of multi-modal weights through process stage analysis of carbon fiber winding; according to the updated data of modal weights, determine real-time winding defects of carbon fiber through carbon fiber defect type analysis.

[0014] In a third aspect, an embodiment of the present application further provides a non-volatile computer storage medium for real-time defect detection based on carbon fiber winding, storing computer-executable instructions, characterized in that the computer-executable instructions are set as follows: obtaining real-time environmental data of carbon fiber winding, and performing multi-modal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time carbon fiber defect data set; performing multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data set to determine synchronous carbon fiber defect data; performing policy-level fusion on the synchronous carbon fiber defect data to obtain preliminary determination data of the winding state; based on the preliminary determination data of the winding state, determining updated multi-modal weight data through analysis of the process stage of carbon fiber winding; and determining real-time carbon fiber winding defects through analysis of the carbon fiber defect type according to the updated modal weight data.

[0015] An embodiment of the present application provides a real-time defect detection method, device and medium based on carbon fiber winding. By means of multi-modal fusion perception of carbon fiber, intelligent decision-making of carbon fiber ply defects and dynamic closed-loop control, the technical problem of poor anti-interference ability in the single-parameter analysis process in carbon fiber winding detection is solved, multi-dimensional analysis of carbon fiber laying defects is realized, and the quality stability of carbon fiber laying and the system stability of carbon fiber winding robots are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings: Figure 1 It is a flowchart of a real-time defect detection method based on carbon fiber winding provided by an embodiment of the present application; Figure 2 It is a schematic internal structure diagram of a real-time defect detection device based on carbon fiber winding provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0018] The embodiments of the present application provide a real-time defect detection method, device and medium based on carbon fiber winding. Through the multi-modal fusion perception of carbon fiber, the intelligent decision-making of carbon fiber ply defects and the dynamic closed-loop control, the technical problem of poor anti-interference ability in the single-parameter analysis process of carbon fiber winding detection is solved, the multi-dimensional analysis of carbon fiber laying defects is realized, and the quality stability of carbon fiber laying and the system stability of carbon fiber winding robots are improved.

[0019] The technical solutions proposed in the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0020] Figure 1 It is a flowchart of a real-time defect detection method based on carbon fiber winding provided by the embodiments of the present application. As Figure 1 shown, a real-time defect detection method based on carbon fiber winding provided by the embodiments of the present application specifically includes the following steps: Step 101, obtain real-time carbon fiber winding environment data, and perform multi-modal collaborative detection on the real-time carbon fiber winding environment data to obtain a real-time carbon fiber defect data set.

[0021] Exemplarily, through the multi-modal collaborative detection of the real-time carbon fiber winding environment data, during the carbon fiber laying process, the influencing parameters that affect the laying quality of the carbon fiber ply are processed in their respective modalities, realizing the multi-modal detection of the environment data and improving the data breadth of the background data for ply quality detection.

[0022] Specifically, the real-time carbon fiber winding environment data further includes: carbon fiber ply, local mechanical vibration data, broadband acoustic emission signal; performing multi-modal collaborative detection on the real-time carbon fiber winding environment data to obtain a real-time carbon fiber defect data set, including: performing high-frequency eddy current array inversion analysis on the carbon fiber ply to obtain carbon fiber ply state data; wherein, the carbon fiber ply state data includes: fiber arrangement direction, deep defect data; performing surface vibration response analysis on the local mechanical vibration data to obtain local mechanical impedance; wherein, the surface vibration response analysis includes: swept-frequency mechanical vibration processing, surface laser vibrometry analysis; based on the broadband acoustic emission signal, determining the ply defect type parameter through acoustic fingerprint feature recognition; obtaining the real-time carbon fiber defect data set according to the carbon fiber ply state data, local mechanical impedance and ply defect type parameter.

[0023] In one embodiment, during the carbon fiber ply laying process, ply state data is collected in real time through a multi-modal sensor. The high-frequency eddy current array sensor operates in a swept-frequency mode of 100 kHz - 10 MHz, obtains the anisotropic conductivity data of fiber arrangement, the sampling rate of each channel is 2 kHz, generates a 64×64 impedance phase matrix, and further determines the carbon fiber ply state data.

[0024] Further, synchronously activate the piezoelectric actuator to generate a swept-frequency mechanical vibration of 10 - 1000 Hz, use a laser Doppler vibrometer (LDV) to measure the surface vibration velocity response with an accuracy of 0.1 μm, and calculate the dynamic stiffness value, and then determine the local mechanical impedance.

[0025] Further, synchronously activate the broadband acoustic emission sensor to capture the acoustic emission signals in the frequency band of 20 kHz - 2 MHz of the winding robot at a sampling rate of 50 MHz, extract the acoustic fingerprint features through wavelet packet transform, and then determine the parameters of the ply defect type.

[0026] Finally, a calibration plate can be fixed on the surface of the carbon fiber ply. There are common marking points for three modes on the calibration plate; establish a multi-modal space mapping relationship by scanning the conductive characteristics of the marking points with an eddy current array, scanning the stiffness characteristics of the marking points with mechanical impedance, and capturing the knocking acoustic fingerprints of the marking points with an acoustic sensor, so as to obtain the real-time carbon fiber defect data set.

[0027] Step 102: Perform multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data set to determine the carbon fiber defect synchronization data.

[0028] Exemplarily, all the real-time carbon fiber defect data are obtained through non-contact data acquisition methods. In order to achieve spatio-temporal alignment of multi-modal data, it is necessary to perform spatio-temporal synchronization calibration on the carbon fiber ply state data, local mechanical impedance, and ply defect type parameters through multi-modal spatio-temporal synchronization calibration.

[0029] Specifically, performing multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data set to determine the carbon fiber defect synchronization data includes: performing multi-modal space calibration on the real-time carbon fiber defect data, and performing feature coordinate transformation on the feature coordinates obtained by multi-modal space calibration to obtain the spatial coordinates of the defect points; among them, the feature coordinate transformation includes: electrical marking point coordinate transformation, stiffness mutation area coordinate transformation, and acoustic fingerprint positioning coordinate transformation; obtaining the data processing delay of the preset sensor, and through reverse delay compensation based on the data processing delay, obtaining multi-modal time synchronization data; determining the carbon fiber defect synchronization data according to the spatial coordinates of the defect points and the multi-modal time synchronization data.

[0030] In one embodiment, to achieve spatio-temporal alignment of multi-modal data, the IEEE 1588 Precision Time Protocol (PTP) is used for hardware-level clock synchronization to ensure that the time error ≤ 1 μs; a space coordinate transformation matrix is established through a multi-modal calibration plate, and the registration error is limited to ≤ 0.1 mm.

[0031] First, fix a multi-modal calibration plate on the surface of the carbon fiber ply. There are conductive marking points (copper), stiffness mutation areas (silicone embedded), and acoustic reflection protrusions (ceramic material) on the calibration plate to meet the coordinate calibration in three modes.

[0032] Since the coordinate systems of the three modalities are not the same, it is necessary to perform a unified transformation of the coordinate system to establish a unified spatial coordinate system transformation matrix to achieve spatial synchronization.

[0033] It should be noted that coordinate transformations such as those for electrical marker points, stiffness mutation regions, and acoustic fingerprint localization can all be used. The purpose of the coordinate transformation is to construct a unified coordinate system, and users can perform the transformation according to their own needs.

[0034] Finally, the data processing delays of each sensor are monitored in real time, and reverse delay compensation is applied to the subsequent fusion module to meet time synchronization.

[0035] Step 103: Perform strategy-level fusion on the carbon fiber defect synchronization data to obtain preliminary determination data of the winding state.

[0036] Exemplarily, by integrating the detection results of multi-modal sensors (eddy current, mechanical impedance, acoustic emission), and combining physical constraints and dynamic weight optimization, the trustworthiness of the preliminary determination data of the winding state is improved, and the influence of external factors on the analysis of the carbon fiber layup state is reduced.

[0037] Specifically, performing strategy-level fusion on the carbon fiber defect synchronization data to obtain preliminary determination data of the winding state specifically includes: performing basic confidence configuration analysis on the carbon fiber defect synchronization data to obtain the modal confidence corresponding to each modality; wherein, the modal confidence includes: eddy current modal confidence, mechanical impedance modal confidence, and acoustic emission modal confidence; based on the modal confidence, through reliability weight correction, determining the first evidence synthesis data; performing dynamic weight optimization on the first evidence synthesis data to obtain the second evidence synthesis data; and according to the second evidence synthesis data, through modal conflict resolution analysis, obtaining the preliminary determination data of the winding state.

[0038] In one embodiment, first, basic confidence assignment functions are respectively defined for the three modalities of high-frequency eddy current, mechanical impedance, and acoustic emission. The eddy current modal confidence is calculated based on impedance phase consistency, the mechanical impedance modal confidence is calculated based on the stiffness deviation rate, and the acoustic emission modal confidence can be output based on the CNN classification probability.

[0039] Then, construct the first D-S evidence synthesis data, correct the D-S evidence synthesis data through the modal reliability weight, determine the second evidence synthesis data, and perform conflict resolution dominated by the acoustic modality on the second evidence synthesis data to obtain the preliminary determination data of the winding state.

[0040] It should be noted that the conflict resolution dominated by the acoustic modality is determined through a preset threshold, and is used to resolve the conflicts of the remaining two modalities in the case where the acoustic emission modality has an adverse effect on the preliminary determination of the winding state.

[0041] Step 104: Based on the preliminary determination data of the winding state, through the process stage analysis of carbon fiber winding, determine the multi-modal weight update data.

[0042] Exemplarily, in order to meet the states of the carbon fiber winding robot at different laying stages, step the multi-modal weights for different process stages so that the defects of the carbon fiber layup under different process stages are all close to the actual state, in order to improve the accuracy of carbon fiber defect analysis.

[0043] Specifically, based on the preliminary determination data of the winding state, through the process stage analysis of carbon fiber winding, determine the multi-modal weight update data, including: obtaining the current process stage, and based on the current process stage, through modal weight configuration, obtain the multi-modal weight adjustment data; wherein, the current process stage includes: initial laying, lamination stage, emergency condition; according to the preliminary determination data of the winding state and the multi-modal weight adjustment data, through modal weight update, determine the multi-modal weight update data.

[0044] In one embodiment, set the modal weight coefficient according to the process stage, and through modal weight configuration, obtain the multi-modal weight adjustment data, which is explained by the following formula: (1) Wherein, is the data quality score, used to quantify the reliability of the current data of each modality; is the modal default weight ( = 3, representing three weights; similarly, is also 3), the initial weight preset according to the prior knowledge of different carbon fiber laying process stages, reflecting the inherent importance of each modality at a specific stage; is the weight distribution ratio for controlling the default weight and the real-time data quality. In this application, the value is controlled between 0.6 - 0.7; is the th modal real-time data quality score ( takes the same value as , which is 1 - 3); represents the sum of the real-time data quality scores of all modalities; is the adjusted weight of the th modality ( takes values from 1 to 3).

[0045] Step 105: According to the modal weight update data, through carbon fiber defect type analysis, determine the real-time winding defects of carbon fiber.

[0046] Exemplarily, after determining the modal weight update data, the analysis status under each laying process of the Taixuanwei winding robot can be configured. Through the analysis of carbon fiber defect types, the defect situation of the carbon fiber ply at a certain process stage is determined, realizing non-contact analysis of carbon fiber ply defects.

[0047] Specifically, according to the modal weight update data, through the analysis of carbon fiber defect types, the real-time winding defects of carbon fiber are determined, specifically including: based on the modal weight update data, through defect type-oriented weight assignment, the defect type weight data is determined; the defect type weight data is subjected to dynamic weight adaptive processing to determine the multi-modal quality weight data; according to the multi-modal quality weight data, through comprehensive confidence threshold determination, the real-time winding defects of carbon fiber are determined.

[0048] Further, after determining the real-time winding defects of carbon fiber through the analysis of carbon fiber defect types according to the modal weight update data, the method further includes: dynamically adjusting the winding equipment parameters for the real-time winding defects of carbon fiber to determine the equipment adjustment parameters; wherein, the equipment adjustment parameters include: pressure dynamic adjustment parameters, laying speed optimization parameters, and defect path avoidance path planning parameters; based on the equipment adjustment parameters, through equipment closed-loop control, the equipment adjustment corresponding to the real-time winding defects of carbon fiber is completed.

[0049] Further, dynamically adjusting the winding equipment parameters for the real-time winding defects of carbon fiber to determine the equipment adjustment parameters specifically includes: constructing a pressure compensation relationship matrix, and based on the pressure compensation relationship matrix, through adaptive PID control analysis, the pressure dynamic adjustment parameters are obtained; obtaining the density change data of the carbon fiber winding layer, and performing gradient descent optimization of the data coefficient on the density change data to obtain the laying speed optimization parameters; obtaining the position of the defect area, and according to the position of the defect area, through B-spline curve optimization analysis, the defect path avoidance path planning parameters are determined.

[0050] In one embodiment, the laying process parameters are adjusted in real time based on the fusion result, and a pressure gradient compensation strategy is adopted for fiber wrinkle defects.

[0051] First, establish a relationship matrix between the defect type and the pressure compensation amount, and determine the dynamic tuning parameters through adaptive PID control.

[0052] Then, in order to meet the stability of carbon fiber laying when defects occur, gradient descent optimization of the data coefficient is performed on the density change data. When the pore density exceeds the preset threshold during the lamination stage, the moving speed of the carbon fiber ply is reduced to further reduce the porosity.

[0053] Finally, due to the layup process limitations of the defective part, taking the defective area as an obstacle, a smooth avoidance path is generated to reduce the influence of uneven pressure on the surrounding carbon fiber layup caused by the defect.

[0054] Furthermore, through B-spline curve optimization analysis, the original path is interpolated and corrected to determine the defect path avoidance path planning parameters.

[0055] The above is the method embodiment proposed in this application. Based on the same inventive concept, the embodiments of this application also provide a real-time defect detection device based on carbon fiber winding, and its structure is as Figure 2 shown.

[0056] Figure 2 It is a schematic diagram of the internal structure of a real-time defect detection device based on carbon fiber winding provided by the embodiment of this application. As Figure 2 shown, the device includes: At least one processor 201; And a memory 202 communicatively connected to the at least one processor; Wherein, the memory 202 stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor 201 so that the at least one processor 201 can: Obtain the real-time carbon fiber winding environment data, and perform multi-modal collaborative detection on the real-time carbon fiber winding environment data to obtain a real-time carbon fiber defect data group; perform multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data group to determine carbon fiber defect synchronization data; perform policy-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data; based on the preliminary winding state determination data, determine multi-modal weight update data through the process stage analysis of carbon fiber winding; according to the modal weight update data, determine the real-time carbon fiber winding defect through the carbon fiber defect type analysis.

[0057] Some embodiments of this application provide a Figure 1 non-volatile computer storage medium for real-time defect detection based on carbon fiber winding, storing computer-executable instructions, and the computer-executable instructions are set to: Obtain the real-time carbon fiber winding environment data, and perform multi-modal collaborative detection on the real-time carbon fiber winding environment data to obtain a real-time carbon fiber defect data group; perform multi-modal spatio-temporal synchronization calibration on the real-time carbon fiber defect data group to determine carbon fiber defect synchronization data; perform policy-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data; based on the preliminary winding state determination data, determine multi-modal weight update data through the process stage analysis of carbon fiber winding; according to the modal weight update data, determine the real-time carbon fiber winding defect through the carbon fiber defect type analysis.

[0058] Each embodiment in this application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the Internet of Things devices and media, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the related content.

[0059] The systems and media provided by the embodiments of this application correspond one-to-one with the methods. Therefore, the systems and media also have beneficial technical effects similar to those of the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be elaborated here.

[0060] Those skilled in the art should understand that the embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0061] This application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0062] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.

[0064] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0065] The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

[0066] Computer-readable media includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0067] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, commodity or device including the element.

[0068] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A real-time defect detection method based on carbon fiber winding, characterized in that: The method comprises: Acquire real-time environmental data of carbon fiber winding, and perform multimodal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time defect data group of carbon fiber; Performing multi-modal spatiotemporal synchronization calibration on the carbon fiber real-time defect data group to determine carbon fiber defect synchronization data; Performing strategic-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data; Based on the preliminary determination data of the winding state, the multimodal weight update data is determined by analyzing the process stage of carbon fiber winding; According to the modal weight update data, the real-time winding defects of the carbon fiber are determined by analyzing the carbon fiber defect types.

2. A real-time defect detection method based on carbon fiber winding according to claim 1, characterized in that: The carbon fiber winding real-time environmental data also includes: carbon fiber layup, local mechanical vibration data, and broadband acoustic emission signals; The carbon fiber winding real-time environmental data is subjected to multimodal collaborative detection to obtain a carbon fiber real-time defect data group, specifically including: Performing high-frequency eddy current array inversion analysis on the carbon fiber ply to obtain carbon fiber ply state data; wherein the carbon fiber ply state data includes: fiber arrangement direction and deep defect data; Performing surface vibration response analysis on the local mechanical vibration data to obtain local mechanical impedance; wherein the surface vibration response analysis includes: frequency sweep mechanical vibration processing and surface laser vibration measurement analysis; Based on the broadband acoustic emission signal, determining ply defect type parameters through acoustic print feature recognition; The carbon fiber real-time defect data group is obtained according to the carbon fiber ply state data, the local mechanical impedance and the ply defect type parameter.

3. The real-time defect detection method based on carbon fiber winding according to claim 1 is characterized in that: The multimodal time-space synchronization calibration is performed on the carbon fiber real-time defect data group to determine the carbon fiber defect synchronization data, specifically including: The real-time defect data of the carbon fiber is calibrated in a multimodal space, and the characteristic coordinates obtained by the multimodal space calibration are converted into characteristic coordinates to obtain the spatial coordinates of the defect point; wherein the characteristic coordinate conversion includes: electrical marking point coordinate conversion, stiffness mutation zone coordinate conversion, and voiceprint positioning coordinate conversion; Acquire a data processing delay of a preset sensor, and obtain multimodal time synchronization data by performing reverse delay compensation based on the data processing delay; The carbon fiber defect synchronization data is determined according to the defect point spatial coordinates and the multimodal time synchronization data.

4. The real-time defect detection method based on carbon fiber winding according to claim 1 is characterized in that: The carbon fiber defect synchronization data is strategically integrated to obtain preliminary winding state determination data, specifically including: Performing basic confidence configuration analysis on the carbon fiber defect synchronization data to obtain modal confidence corresponding to each mode; wherein the modal confidence includes: eddy current modal confidence, mechanical impedance modal confidence, and acoustic emission modal confidence; Based on the modal confidence, determining first evidence synthesis data through reliability weight correction; Dynamically weighting the first evidence synthesis data to obtain second evidence synthesis data; The preliminary determination data of the entanglement state is obtained by synthesizing the data based on the second evidence through modal conflict resolution analysis.

5. The real-time defect detection method based on carbon fiber winding according to claim 1 is characterized in that: Based on the preliminary determination data of the winding state, the multimodal weight update data is determined through the process stage analysis of the carbon fiber winding, specifically including: Acquire the current process stage, and based on the current process stage, obtain multi-modal weight adjustment data through modal weight configuration; wherein the current process stage includes: initial laying, lamination stage, and emergency working condition; The multimodal weight update data is determined by updating the modal weight according to the preliminary determination data of the entanglement state and the multimodal weight adjustment data.

6. The real-time defect detection method based on carbon fiber winding according to claim 1 is characterized in that: According to the modal weight update data, the real-time winding defects of carbon fiber are determined by analyzing the carbon fiber defect types, specifically including: Based on the modal weight update data, defect type weight data is determined by defect type guided weight allocation; Performing dynamic weight adaptive processing on the defect type weight data to determine multimodal quality weight data; The real-time winding defect of the carbon fiber is determined based on the multimodal quality weight data through a comprehensive confidence threshold judgment.

7. The real-time defect detection method based on carbon fiber winding according to claim 1 is characterized in that: After updating the data according to the modal weight and determining the real-time winding defect of the carbon fiber by analyzing the carbon fiber defect type, the method further includes: Dynamically adjusting the winding equipment parameters for the real-time winding defects of the carbon fiber to determine the equipment adjustment parameters; wherein the equipment adjustment parameters include: dynamic pressure adjustment parameters, laying speed optimization parameters, and defect path avoidance path planning parameters; Based on the equipment adjustment parameters, equipment closed-loop control is performed to complete equipment adjustment corresponding to the real-time winding defects of the carbon fiber.

8. The real-time defect detection method based on carbon fiber winding according to claim 7 is characterized in that: Dynamically adjusting the winding equipment parameters for the real-time winding defects of the carbon fiber to determine the equipment adjustment parameters specifically includes: Constructing a pressure compensation relationship matrix, and based on the pressure compensation relationship matrix, obtaining the pressure dynamic adjustment parameter through adaptive PID control analysis; Acquiring density change data of the carbon fiber winding layer, and performing gradient descent optimization of data coefficients on the density change data to obtain the laying speed optimization parameter; The defect area position is obtained, and according to the defect area position, the defect path avoidance path planning parameters are determined through B-spline curve optimization analysis.

9. A real-time defect detection device based on carbon fiber winding, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Acquire real-time environmental data of carbon fiber winding, and perform multimodal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time defect data group of carbon fiber; Performing multi-modal spatiotemporal synchronization calibration on the carbon fiber real-time defect data group to determine carbon fiber defect synchronization data; Performing strategic-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data; Based on the preliminary determination data of the winding state, the multimodal weight update data is determined by analyzing the process stage of carbon fiber winding; According to the modal weight update data, the real-time winding defects of the carbon fiber are determined by analyzing the carbon fiber defect types.

10. A non-volatile computer storage medium for real-time defect detection based on carbon fiber winding, storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Acquire real-time environmental data of carbon fiber winding, and perform multimodal collaborative detection on the real-time environmental data of carbon fiber winding to obtain a real-time defect data group of carbon fiber; Performing multi-modal spatiotemporal synchronization calibration on the carbon fiber real-time defect data group to determine carbon fiber defect synchronization data; Performing strategic-level fusion on the carbon fiber defect synchronization data to obtain preliminary winding state determination data; Based on the preliminary determination data of the winding state, the multimodal weight update data is determined by analyzing the process stage of carbon fiber winding; According to the modal weight update data, the real-time winding defects of the carbon fiber are determined by analyzing the carbon fiber defect types.

Images