An ultrasonic testing device and method for detecting cable aging

Abstract

The application relates to an ultrasonic detection device and method for detecting cable aging, which evaluates the aging condition of the outer surface of a cable by recording cable test data, identifies the fault conditions inside and outside the cable through a second neural network module to generate a cable image set analysis result, and obtains an evaluation score for the cable aging condition through a cable quality detection module based on the cable test data and the cable image set analysis result, thereby standardizing the evaluation standard of the cable aging, making the cable aging condition clearer and more explicit, and making the cable aging detection result more accurate and reasonable. The first neural network module is used for auditing the quality of the cable ultrasonic image photographed by the ultrasonic equipment, so as to avoid the problem that the identification effect of the second neural network module is affected by the defective image.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic testing technology, and in particular to an ultrasonic testing device and method for detecting cable aging. Background Technology

[0002] Cables are typically rope-like structures made of several or groups of conductors (at least two conductors per group) twisted together. Each group of conductors is insulated from the others and is often twisted around a central conductor, with the entire structure covered by a highly insulating outer layer. Cables are characterized by internal conductivity and external insulation. Types of cables include power cables, control cables, compensating cables, shielded cables, high-temperature cables, computer cables, signal cables, coaxial cables, fire-resistant cables, marine cables, mining cables, aluminum alloy cables, and so on. They are all composed of single or multiple strands of conductors and insulation layers, used to connect circuits and electrical appliances. Current cable testing methods rely on manual observation to assess cable aging, which cannot detect internal cable faults and lacks reasonable cable evaluation standards. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention provides an ultrasonic testing device and method for detecting cable aging, which is efficient and accurate in detecting cable aging.

[0004] The present invention adopts the following technical solution:

[0005] In a first aspect, an ultrasonic testing method for detecting cable aging includes the following steps:

[0006] Step S1: Inspect the cable using ultrasonic equipment; acquire ultrasonic images of the cable, perform noise reduction processing on the ultrasonic images of the cable, and obtain an initial cable image set;

[0007] Step S2: Determine the defects in the ultrasonic images of the cable in the initial cable image set and generate the target cable image set;

[0008] Step S3: Identify and analyze the features of the ultrasonic images of the cable in the target cable image set, and obtain the analysis results of the target cable image set;

[0009] Step S4: Perform quality inspection on the cable to generate cable test data. Based on the analysis results of the cable test data and the target cable image set, evaluate the aging condition of the cable.

[0010] Furthermore, the method for determining the defect status of the cable ultrasonic images in the initial cable image set and generating the target cable image set in step S2 is as follows:

[0011] The first neural network module is trained by using multiple sets of images with shooting defects, wherein the images with shooting defects include: shooting defect sample images and shooting defect category labels;

[0012] An initial set of cable images is input into a first neural network module; wherein the initial set of cable images includes multiple consecutive ultrasonic images of the cable.

[0013] S21. Input the initial cable image set into the first neural network module; wherein the initial cable image set includes multiple consecutive cable ultrasound images;

[0014] S22. Prepare to judge the first cable ultrasound image through the first neural network module;

[0015] S23. Determine if there is a defect in the current cable ultrasonic image; if so, delete the cable ultrasonic image; otherwise, save the cable ultrasonic image to the target cable image set.

[0016] S24. Determine whether all cable ultrasound images have been judged. If not, return to step S21 to continue judging the next cable ultrasound image; otherwise, proceed to step S25.

[0017] S25. The first neural network module determines whether the percentage of the number of cable ultrasound images in the target cable image set relative to the total number of cable ultrasound images input to the first neural network module reaches a set value. If it does not reach the set value, the output cable ultrasound images do not meet the requirements, and the process returns to step S1 to reacquire the initial cable image set; otherwise, the target cable image set is output.

[0018] Furthermore, step S3, which involves identifying and analyzing the features of the ultrasonic images of the target cable in the image set, further includes the following steps:

[0019] A second neural network module is trained using images of faulty ultrasonic cables, wherein the images of faulty ultrasonic cables include: cable fault sample images and cable fault category labels;

[0020] The ultrasonic images of the cable in the target cable image set are input into the second neural network module. For each ultrasonic image of the cable in the target cable image set, the type of cable fault and the probability of the fault are output, and the location of the cable fault is selected to generate the analysis results of the target cable image set.

[0021] Furthermore, in step S4, the method for quality inspection of the cable is as follows:

[0022] S41. Construct a cable database;

[0023] S42. Obtain the basic cable parameters of each cable in the area to be tested and enter them into the cable database;

[0024] S43. Test the cable on site, generate cable test data, and input the cable test data and cable image set analysis results into the quality inspection module to obtain the cable quality score;

[0025] S44. Assess the degree of aging of the cable based on the cable quality score.

[0026] Furthermore, the basic parameters of the cable include the cable type, allowable current carrying capacity, and insulation thickness.

[0027] Furthermore, the field test cable generates cable test data including:

[0028] Record the cable's usage time, surface gloss, and hardness, and then test the cable's insulation resistance.

[0029] The cable's usage time, surface gloss and hardness, and insulation resistance are sequentially entered into the cable database, which then generates cable test data.

[0030] Secondly, the present invention provides an ultrasonic testing device for detecting cable aging, characterized in that it includes an ultrasonic device, a first neural network module, a second neural network module, a quality testing module, and a cable database.

[0031] The ultrasonic device is used to acquire ultrasonic images of the cable;

[0032] The first neural network module is used to filter images with shooting defects from the initial cable image set;

[0033] The second neural network module is used to determine whether the cable has a fault;

[0034] The quality inspection module is used to evaluate cable quality based on cable quality inspection results and target cable image set analysis results.

[0035] The cable database is used to store basic cable data and cable test data.

[0036] Furthermore, it also includes a noise reduction module, which is used to make the ultrasonic images of the cable clearer, making it easier for the second neural network module to identify faults.

[0037] Thirdly, the present invention provides a terminal including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the method as described in any possible implementation of the first aspect.

[0038] Fourthly, another object of the present invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in any possible implementation of the first aspect.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] In view of the technical problems existing in the prior art and the difficulty of solving these problems, and closely combining the technical solution to be protected by this invention with the results and data during the research and development process, this paper analyzes in detail how the technical solution of this invention solves the technical problems, and the inventive technical effects brought about after solving the problems. The specific content is as follows:

[0041] 1. This invention assesses the aging of the cable's outer surface by recording cable test data, identifies internal and external faults in the cable through a second neural network module to generate target cable image set analysis results, and, based on the cable test data and target cable image set analysis results, uses a cable quality inspection module to derive an evaluation score for the cable aging condition. This standardizes the evaluation criteria for cable aging, making the cable aging condition clearer and more accurate and reasonable.

[0042] 2. The present invention includes a noise reduction module to process the ultrasonic images of the cable, making them clearer and increasing the accuracy of fault identification.

[0043] 3. The present invention uses a first neural network module to review the quality of the ultrasonic images of the cable captured by the ultrasonic equipment, thereby avoiding the impact of defects in the capture on the recognition effect of the second neural network module.

[0044] 4. This invention records basic cable parameters and cable test data. The basic cable parameters are used to make a preliminary assessment of the cable's durability, and the cable test data is used to make a preliminary assessment of the cable's aging degree. Finally, the cable aging condition is determined based on the cable evaluation score from the quality inspection module. By comparing the basic cable parameters with the cable evaluation score, it is possible to clearly determine which cable is more durable and provide data support for future cable selection. Attached Figure Description

[0045] Figure 1 This is a flowchart of an ultrasonic testing method for detecting cable aging according to the present invention.

[0046] Figure 2 This is a flowchart of the method for quality inspection of cables, generating cable test data, and evaluating cable aging in this invention.

[0047] Figure 3This is a schematic diagram of an ultrasonic testing device for detecting cable aging in an embodiment of the present invention.

[0048] Figure 4 This is a flowchart of the method for determining the defect status of cable ultrasonic images in the initial cable image set in this invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0050] Example 1:

[0051] An ultrasonic testing method for detecting cable aging, characterized in that the method includes the following steps:

[0052] Step S1: Inspect the cable using ultrasonic equipment; acquire ultrasonic images of the cable, perform noise reduction processing on the ultrasonic images of the cable, and obtain an initial cable image set;

[0053] Step S2: Determine the defects in the ultrasonic images of the cable in the initial cable image set and generate the target cable image set.

[0054] Step S3: Identify and analyze the features of the ultrasonic images of the cable in the target cable image set, and obtain the analysis results of the target cable image set;

[0055] Step S4: Perform quality inspection on the cable to generate cable test data. Based on the analysis results of the cable test data and the target cable image set, evaluate the aging condition of the cable.

[0056] More specifically, denoising is a crucial step in image processing. Specific methods include one or more of the following: mean filtering, Gaussian filtering, median filtering, and bilateral filtering. In this embodiment, the denoising method employs median filtering, a technique already known in the art. While linear filtering can typically be used to process noisy images, many linear filters suffer from low-pass properties, blurring edge information while removing noise. Therefore, in certain cases, median filtering can remove noise while preserving image edges, making it a non-linear noise removal method. The principle behind median filtering is to replace the value of a point in the digital image with the median value of all points within the region containing that point. Compared to mean filtering, median filtering preserves more image detail while removing noise. Median filtering is particularly effective at filtering impulse noise. Impulse noise, also known as salt-and-pepper noise, is a common type of noise in images. It consists of randomly appearing white or black dots, possibly black pixels in bright areas or white pixels in dark areas (or both).

[0057] In this embodiment, ultrasound refers to sound waves with a vibration frequency greater than 20 kHz. It propagates in a medium and, upon encountering heterogeneous interfaces with different acoustic impedances (such as defects or the object being probed), will produce reflection and transmission. Scattering and attenuation occur during propagation. By receiving the reflected and transmitted waves, useful information is extracted for ultrasonic detection and imaging. Commonly used information includes amplitude and frequency. Black and white ultrasound is based on the principle of ultrasonic imaging. Ultrasound is actually formed by computer simulation processing based on the amount of ultrasonic energy reflected by different substances. When the medium density is high, more ultrasonic energy is reflected, and the image appears white. When the medium density is low, less energy is reflected, and the image appears black—this is what we commonly refer to as black and white ultrasound imaging.

[0058] The method for determining the defect status of the cable ultrasonic images in the initial cable image set and generating the target cable image set in step S2 is as follows:

[0059] The first neural network module is trained by using multiple sets of images with shooting defects; the images with shooting defects include: shooting defect sample images and shooting defect category labels.

[0060] More specifically, due to the excessive length of the cable, it is necessary to divide the shooting area according to the cable length and shoot separately. When shooting the cable in each area, it is necessary to capture the full view of the cable in that area. If only a part is captured, the photo will not be representative and will affect the subsequent analysis of cable faults. Therefore, in this embodiment, the image of the defect refers to the capture of a part of the cable, or the capture is unclear due to foreign objects obstructing the view or shaking during shooting.

[0061] More specifically, shooting defect labels include: unclear image, obstruction by foreign objects, and incomplete shooting.

[0062] More specifically, the algorithm of the first neural network module in this embodiment adopts the Faster-RCNN algorithm in the prior art.

[0063] S21. Input the initial cable image set into the first neural network module; wherein the initial cable image set includes multiple consecutive cable ultrasound images;

[0064] S22. Prepare to judge the first cable ultrasound image through the first neural network module;

[0065] S23. Determine if there is a defect in the current cable ultrasonic image; if so, delete the cable ultrasonic image; otherwise, save the cable ultrasonic image to the target cable image set.

[0066] S24. Determine whether all cable ultrasound images have been judged. If not, return to step S21 to continue judging the next cable ultrasound image; otherwise, proceed to step S25.

[0067] S25. The first neural network module determines whether the percentage of the number of cable ultrasound images in the target cable image set relative to the total number of cable ultrasound images input to the first neural network module reaches a set value. If it does not reach the set value, the output cable ultrasound images do not meet the requirements, and the process returns to step S1 to reacquire the initial cable image set; otherwise, the target cable image set is output.

[0068] Step S3 involves identifying and analyzing the features of the ultrasonic images of the target cable in the image set, and further includes the following steps:

[0069] A second neural network module is trained using images of faulty ultrasonic cables, wherein the images of faulty ultrasonic cables include: cable fault sample images and cable fault category labels;

[0070] More specifically, the algorithm of the second neural network module in this embodiment also adopts the existing Faster-RCNN algorithm.

[0071] The ultrasonic images of the cable in the target cable image set are input into the second neural network module;

[0072] For each ultrasonic image of the target cable image set, the type of cable fault and the probability of the fault are output, and the location of the cable fault is selected to generate the analysis results of the target cable image set.

[0073] More specifically, cable fault types include: damage from external forces, insulation dampness, chemical corrosion, and long-term overload operation;

[0074] Damage caused by external forces: This is caused by damage to mechanical equipment. For example, improper construction during cable laying and installation can easily lead to damage to mechanical equipment; civil engineering work on directly buried cables can also easily damage the operating cables. Sometimes, if the damage is not severe, it may take several months or even years for the damaged area to completely break down and cause a fault. Sometimes, more severe damage may cause a short circuit, seriously damaging the equipment.

[0075] Insulation dampness: This is also a common problem, usually occurring at cable joints in direct burial or conduit installations. For example, improper cable joint construction or making connections in humid and cold conditions can cause leaks or water vapor to seep into the joint, gradually damaging the cable's insulation strength and leading to failure.

[0076] Chemical corrosion: When cables are buried directly in areas with strong acids and alkalis, the cable's armor, lead sheath, or outer sheath will be corroded. Due to long-term chemical or electrolytic corrosion, the protective layer becomes ineffective, insulation is reduced, and cable failure will occur.

[0077] Long-term overload operation: During long-term overload operation, the thermal effect of the current causes the conductor to heat up when the load current passes through the cable. Additionally, the skin effect of the charge, eddy current losses in the armor, and dielectric losses in the insulation also generate additional heat, further increasing the cable temperature. Under long-term overload operation, excessively high temperatures accelerate insulation aging, and in severe cases, the insulation layer may deform significantly, leading to insulation breakdown. Especially in the hot summer, the temperature rise in cables often causes weak points in the insulation to break down first, resulting in a high incidence of cable faults during the summer.

[0078] This invention incorporates a noise reduction module to process ultrasonic images of cables, making them clearer and increasing the accuracy of fault identification.

[0079] This invention uses a first neural network module to review the quality of the ultrasonic images of cables captured by the ultrasonic equipment, thus avoiding the impact of defects in the imaging on the recognition effect of the second neural network module.

[0080] In step S4, the method for quality inspection of the cable is as follows:

[0081] S41. Construct a cable database;

[0082] S42. Obtain the basic cable parameters of each cable in the area to be tested and enter them into the cable database;

[0083] S43. Test the cable on site, generate cable test data, and input the cable test data and cable image set analysis results into the quality inspection module to obtain the cable quality score;

[0084] S44. Assess the degree of aging of the cable based on the cable quality score.

[0085] Basic cable parameters include cable type, allowable current carrying capacity, insulation thickness, and other parameters affecting cable quality. The cable type primarily includes the application code, insulation code, conductor material code, inner sheath code, derived code, outer sheath code, core cross-sectional area, and number of cores. These are used to identify the cable's basic data. Among them:

[0086] The application codes are: not identified as power cable, K for control cable, and P for signal cable.

[0087] The insulation codes are: oil-impregnated paper, X rubber, V polyvinyl chloride, YJ cross-linked polyethylene, and Y polyethylene.

[0088] The conductor material codes are: copper if not marked, and aluminum if L.

[0089] The inner sheath codes are: Q (lead sheath), L (aluminum sheath), H (rubber sheath), and V (polyvinyl chloride sheath). The inner sheath is generally not marked.

[0090] The derived code is: D non-dripping, P dry insulation.

[0091] The outer sheath code is: V for polyvinyl chloride, Y for polyethylene. Power cables and control cables have an outer sheath, while rubber-sheathed cables generally do not have an outer sheath.

[0092] Special product codes are: TH (humid tropical), TA (dry tropical), ZR (flame retardant), NH (fire resistant), WDZ (low smoke halogen-free, enterprise standard).

[0093] Rated voltage is in kV.

[0094] Field testing of cables generates cable test data including:

[0095] Record the cable's service life, the gloss level of the cable's outer surface, and its hardness; test the cable's insulation resistance.

[0096] The cable's usage time, surface gloss, hardness, and insulation resistance are sequentially entered into the cable database, which then generates cable test data.

[0097] An ultrasonic testing device for detecting cable aging includes an ultrasonic device, a first neural network module, a second neural network module, a quality testing module, and a cable database.

[0098] Ultrasonic equipment is used to acquire ultrasonic images of cables;

[0099] The first neural network module is used to filter images with shooting defects from the initial cable image set;

[0100] The second neural network module is used to determine whether the cable has a fault.

[0101] The quality inspection module is used to evaluate cable quality based on cable quality inspection results and target cable image set analysis results; the cable database is used to store basic cable data and cable test data.

[0102] It also includes a noise reduction module, which is used to make the ultrasonic images of the cable clearer, making it easier for the second neural network module to identify faults.

[0103] Example 2:

[0104] The specific implementation process of this invention is as follows:

[0105] The imaging area is divided according to the cable length, and the cable in each area is imaged using an ultrasonic device. In this embodiment, the cable to be tested is a complete cable. Ultrasonic images of the cable are acquired, and the cable images are denoised using a median filtering algorithm to generate an initial cable image set.

[0106] The initial cable image set is input into the first neural network module, and the first cable ultrasound image is prepared for judgment. The first neural network module then judges whether there is a shooting defect in the current cable ultrasound image. If there is, the cable ultrasound image is deleted; otherwise, the cable ultrasound image is saved to the target cable image set.

[0107] The algorithm determines whether all cable ultrasound images have been evaluated. If not, it continues to evaluate the next cable ultrasound image. Otherwise, it determines whether the percentage of cable ultrasound images in the target cable image set relative to the total number of cable ultrasound images input to the first neural network module reaches a set value. If not, the output cable ultrasound image does not meet the requirements and the cable needs to be re-inspected by the ultrasound equipment. The algorithm acquires cable ultrasound images, performs noise reduction processing on the cable ultrasound images, and obtains the initial cable image set. Otherwise, it outputs the target cable image set.

[0108] The set value in the first neural network module is set to 90%. The first neural network module determines whether the percentage of the number of cable ultrasound images in the target cable image set relative to the total number of cable ultrasound images input to the first neural network module reaches the set value. If it does not reach the set value, the output cable ultrasound images do not meet the requirements, and the process returns to step S1 to reacquire cable ultrasound images; otherwise, the target cable image set is output.

[0109] The ultrasonic images of the target cable image set are input into the second neural network module to generate the target cable image set analysis results, which include:

[0110] 1) Fault identification results for each ultrasonic image of the target cable in the image set, including a square box marking the fault location, the fault type, and the probability of the corresponding fault.

[0111] 2) A fault summary table of the target cable image set, including all fault categories and the maximum probability of each fault.

[0112] Build a cable database, obtain the basic cable parameters of each cable in the area to be tested, and enter them into the cable database;

[0113] On-site testing of the cable under test, recording the cable's usage time, surface gloss, hardness, and insulation resistance;

[0114] The cable's usage time, surface gloss, hardness, and insulation resistance are sequentially entered into the cable database, which then generates cable test data.

[0115] Input the cable test data and cable image set analysis results into the quality inspection module to obtain the cable quality score;

[0116] The method for evaluating cable test data scores is as follows:

[0117] The evaluation score for cable service life is equal to the number of years the cable has been used. When the cable has been used for more than 20 years, the evaluation score is 20 points; that is, if the cable has been used for 5 years, the evaluation score is 5 points; if the cable has been used for 15 years, the evaluation score is 15 points; and if the cable has been used for 23 years, the evaluation score is 20 points.

[0118] The maximum score for evaluating the glossiness of the cable surface is 5 points; the lower the glossiness of the cable surface, the higher the score.

[0119] The maximum score for cable surface hardness assessment is 5 points; the greater the stiffness of the cable surface, the higher the score.

[0120] The maximum score for cable insulation resistance evaluation is 5 points; the higher the cable insulation resistance, the higher the score.

[0121] The scoring method for cable image set analysis results is as follows: fault type and the probability of the corresponding fault.

[0122] There are four types of faults: damage from external forces, insulation dampness, chemical corrosion, and long-term overload operation. The maximum score for each fault is 15 points.

[0123] Q zl = Q z - Q cs - Q fx

[0124] Where Qzl is the cable quality score, Qz is the total cable quality score, Qcs is the sum of the evaluation scores of the cable test data, and Qfx is the sum of the evaluation scores of the cable image set analysis results. In this real-time example, the value of Qz is 100. The smaller the value of Qzl, the higher the degree of cable aging.

[0125] Where Qcs = Y + L + D + J

[0126] Y represents the cable service life assessment score, L represents the cable surface gloss assessment score, D represents the cable surface hardness assessment score, and J represents the cable insulation resistance assessment score.

[0127] Q fx =15×(G wl +G sc +G fs +G gz )

[0128] Where Gwl is the probability of failure caused by external force damage, Gsc is the probability of failure caused by insulation moisture, Gfs is the probability of failure caused by chemical corrosion, and Ggz is the probability of failure caused by long-term overload operation. The values ​​of Gwl, Gsc, Gfs, and Ggz are all between 0% and 100%.

[0129] This invention records basic cable parameters and cable test data. The basic cable parameters are used to make a preliminary assessment of the cable's durability, and the cable test data is used to make a preliminary assessment of the cable's aging degree. Finally, the cable aging condition is determined based on the cable evaluation score from the quality inspection module. By comparing the basic cable parameters with the cable evaluation score, it is possible to clearly determine which cable is more durable, providing data support for future cable selection.

[0130] This invention assesses the aging of the cable's outer surface by recording cable test data, identifies internal and external faults in the cable through a second neural network module to generate cable image set analysis results, and, based on the cable test data and cable image set analysis results, uses a cable quality inspection module to derive an evaluation score for the cable aging condition. This standardizes the evaluation criteria for cable aging, making the cable aging condition clearer and more accurate and reasonable.

[0131] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0132] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An ultrasonic testing method for detecting cable aging, characterized in that, The method includes the following steps: Step S1: Inspect the cable using ultrasonic equipment; acquire ultrasonic images of the cable, perform noise reduction processing on the ultrasonic images of the cable, and obtain an initial cable image set; Step S2: Determine the defects in the ultrasonic images of the cable in the initial cable image set and generate the target cable image set; Step S3: Identify and analyze the features of the ultrasonic images of the cable in the target cable image set, and obtain the analysis results of the target cable image set; Step S4 involves performing quality inspection on the cable to generate cable test data. Based on the analysis results of the cable test data and the target cable image set, the aging condition of the cable is evaluated. The method for performing quality inspection on the cable in step S4 is as follows: Step S41: Build a cable database; Step S42: Obtain the basic cable parameters of each cable in the area to be tested and enter them into the cable database; Step S43: Test the cable on site, generate cable test data, and input the cable test data and cable image set analysis results into the quality inspection module to obtain the cable quality score; In step S43, the cable test data generated during on-site testing includes: Record the cable's usage time, surface gloss, and hardness, and then test the cable's insulation resistance. The cable's service time, surface gloss and hardness, and insulation resistance are sequentially entered into the cable database, which then generates cable test data. S44. Assess the degree of aging of the cable based on the cable quality score.

2. The ultrasonic testing method for detecting cable aging according to claim 1, characterized in that: The method for determining the defect status of the cable ultrasonic images in the initial cable image set and generating the target cable image set in step S2 is as follows: This method is applied to the first neural network module. The first neural network module is trained and obtained by using multiple sets of images with shooting defects. The images with shooting defects include: shooting defect sample images and shooting defect category labels. S21. Input the initial cable image set into the first neural network module; wherein the initial cable image set includes multiple consecutive cable ultrasound images; S22. Prepare to judge the first cable ultrasound image; S23. Determine if there is a defect in the current cable ultrasonic image; if so, delete the cable ultrasonic image; otherwise, save the cable ultrasonic image to the target cable image set. S24. Determine whether all cable ultrasound images have been judged. If not, return to step S21 to continue judging the next cable ultrasound image; otherwise, proceed to step S25. S25. Determine whether the percentage of the number of cable ultrasound images in the target cable image set relative to the total number of cable ultrasound images input to the first neural network module reaches the set value. If it does not reach the set value, the output cable ultrasound images do not meet the requirements, and return to step S1 to reacquire the initial cable image set; otherwise, output the target cable image set.

3. The ultrasonic testing method for detecting cable aging according to claim 1, characterized in that: Step S3, which involves identifying and analyzing the features of the ultrasonic images of the target cable in the target cable image set, further includes the following steps: A second neural network module is obtained by training with images of faulty ultrasonic cables, wherein the images of faulty ultrasonic cables include cable fault sample images and cable fault category labels. The ultrasonic images of the cable in the target cable image set are input into the second neural network module. For each ultrasonic image of the cable in the target cable image set, the type of cable fault and the probability of the fault are output, and the location of the cable fault is selected to generate the analysis results of the target cable image set.

4. The ultrasonic testing method for detecting cable aging according to claim 1, characterized in that: The basic parameters of the cable include the cable type, allowable current carrying capacity, and insulation thickness.

5. An ultrasonic testing device for detecting cable aging, based on the ultrasonic testing method for detecting cable aging according to any one of claims 1-4, characterized in that, Includes ultrasonic equipment, a first neural network module, a second neural network module, a quality inspection module, and a cable database; The ultrasonic device is used to acquire ultrasonic images of the cable; The first neural network module is used to filter images with shooting defects from the initial cable image set; The second neural network module is used to determine whether the cable has a fault; The quality inspection module is used to evaluate cable quality based on cable quality inspection results and target cable image set analysis results. The cable database is used to store basic cable data and cable test data.

6. The ultrasonic testing device for detecting cable aging according to claim 5, characterized in that: It also includes a noise reduction module, which is used to make the ultrasonic images of the cable clearer, making it easier for the second neural network module to identify faults.

7. A terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 4 above.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 4 above.

Images