Method and system for detecting cable seal lead structure

Abstract

This invention relates to the field of cable inspection technology, and proposes a method and system for inspecting cable lead-sealed structures. The method involves applying an excitation signal to a sensor array deployed on the lead-sealed structure of the cable to be inspected, acquiring the differential induced signal generated by the cable lead-sealed structure under the excitation magnetic field, preprocessing it, and constructing an image of the conductivity distribution of the lead-sealed structure cross-section for defect location, quantitative identification, and condition assessment. Based on magnetic induction tomography, the internal conductivity characteristics of the cable lead-sealed structure are visualized, thereby completing defect location, quantitative identification, and overall condition assessment. Furthermore, the sensor array is configured as multiple differential coil units, with symmetrical anti-phase series induction coils on both sides of the excitation coil of each differential coil unit, effectively suppressing environmental common-mode interference and retaining only useful signals from defects, thus improving the ability to identify weak defect signals. This effectively improves the accuracy, reliability, and early warning capability of cable lead-sealed structure inspection.

Description

Technical Field

[0001] This invention relates to the field of cable testing technology, and in particular to a method and system for testing the lead-sealed structure of cables. Background Technology

[0002] High-voltage cables are core equipment in power transmission systems. The lead-sealed structure of the cable joint is used to achieve the electrical connection between the metal sheath and the grounding wire, and is a key component to ensure the reliability of the grounding system. During long-term operation, the lead-sealed structure is prone to microscopic defects such as cracks, air gaps, and poor contact due to factors such as differences in thermal expansion coefficients, poor process consistency, electromagnetic vibration, thermal cycling, and electrochemical corrosion. These defects are highly concealed and difficult to identify in the early stages using conventional detection methods. They can easily lead to increased contact resistance, localized overheating, grounding failure, and even cable joint failure and explosion, seriously threatening the safety of the power grid.

[0003] Existing methods for inspecting lead-sealed structures mainly include infrared thermography and grounding current detection. However, infrared thermography can only reflect surface temperature and is not sensitive to internal defects, while grounding current detection cannot accurately locate and determine the extent of defects. In other words, existing methods for inspecting lead-sealed structures generally suffer from weak internal defect identification capabilities and insufficient location accuracy, failing to meet the practical needs for efficient, accurate, and non-destructive testing of cable lead-sealed structures. Summary of the Invention

[0004] This invention aims to at least solve the technical problems of weak internal defect identification capability and insufficient positioning accuracy in existing lead-sealed structure inspection methods. Therefore, the purpose of this invention is to propose a method and system for inspecting cable lead-sealed structures.

[0005] In a first aspect, the present invention proposes a method for detecting the lead-sealed structure of a cable, comprising the following steps: applying an excitation signal to a sensor array deployed on the lead-sealed structure of the cable to be tested, wherein the sensor array includes multiple differential coil units, each of the differential coil units including an excitation coil and two induction coils symmetrically distributed on both sides of the excitation coil and connected in antiphase; acquiring the induced differential signal generated by the lead-sealed structure of the cable under the excitation magnetic field, and preprocessing the induced differential signal; constructing a conductivity distribution image of the cross-section of the lead-sealed structure based on the preprocessed induced differential signal, and performing defect location, quantitative identification, and condition assessment based on the conductivity distribution image.

[0006] According to the cable lead seal structure detection method of the present invention, an excitation signal is first applied to a sensor array deployed on the cable lead seal structure to be tested. Then, the induced differential signal generated by the cable lead seal structure under the excitation magnetic field is acquired and preprocessed. Finally, based on the preprocessed induced differential signal, a conductivity distribution image of the lead seal structure cross section is constructed. Defect location, quantitative identification, and status assessment are performed based on the conductivity distribution image. In this way, by applying an excitation signal, acquiring and preprocessing the induced differential signal, constructing a conductivity distribution image, and performing defect identification and status assessment, the internal conductivity characteristics of the cable lead seal structure are visualized based on magnetic induction tomography. This allows for non-contact, non-destructive, and accurate defect location, quantitative identification, and overall status assessment. Furthermore, by setting the sensor array to use multiple differential coil units, with symmetrical anti-phase series induction coils on both sides of the excitation coil of each differential coil unit, environmental common-mode interference can be effectively suppressed, retaining only the useful signal brought by the defect. This enhances the sensitivity of detecting changes in the electromagnetic response of the lead seal structure and improves the ability to identify weak defect signals. Thus, the accuracy, reliability, and early warning capability of cable lead seal structure detection are effectively improved, which helps to ensure the safe and stable operation of high-voltage cable systems.

[0007] In addition, the cable lead sealing structure detection method according to embodiments of the present invention may also have the following additional technical features:

[0008] Furthermore, applying an excitation signal to the sensor array deployed on the lead-sealed structure of the cable to be tested includes: sequentially applying an output excitation signal to the excitation coils of the plurality of differential coil units in a time-division multiplexing manner using a single coil.

[0009] Furthermore, the acquisition of the induced differential signal generated by the cable sealing structure under the excitation magnetic field includes: simultaneously acquiring the induced differential signal output by the induction coil in all the other differential coil units when the current excitation coil is working; and sequentially acquiring the induced differential signal corresponding to all excitation positions according to the excitation rotation sequence.

[0010] Furthermore, the preprocessing of the induced differential signal includes: sequentially performing differential amplification, programmable gain adjustment, and low-pass filtering conditioning on the induced differential signal.

[0011] Furthermore, the step of constructing the conductivity distribution image of the lead-sealed structure cross-section based on the preprocessed induced differential signal includes: reconstructing the conductivity distribution image of the lead-sealed structure cross-section using linear back projection, Landweber iteration method, or Tikhonov regularization method based on the amplitude data and phase data of the induced differential signal.

[0012] Furthermore, after constructing the conductivity distribution image of the lead-sealed structure cross-section based on the preprocessed differential induction signal, the process further includes: extracting multi-dimensional features from the preprocessed differential induction signal to generate a trend curve reflecting the overall state of the lead-sealed structure over time, used to determine the detection state of the lead-sealed structure; generating an instantaneous response distribution feature map based on the preprocessed differential induction signal to reflect local abnormal changes in the lead-sealed structure; performing statistical analysis of the response intensity based on the preprocessed differential induction signal to determine the signal sensitivity distribution characteristics, used for sensor array weight allocation and optimization; and calculating the spatiotemporal dimension distribution based on the preprocessed differential induction signal to generate a two-dimensional distribution image, used to distinguish the response differences between normal and defective structures.

[0013] Furthermore, the excitation signal is a 1MHz–10MHz alternating sinusoidal excitation signal.

[0014] Furthermore, the sensor array is fixed in a ring shape by a non-metallic flexible substrate, adapting to fit the surface of the cable lead sealing structure.

[0015] Secondly, based on the same inventive concept, this invention also proposes a cable lead-sealing structure detection system for implementing the cable lead-sealing structure detection method described in the above embodiments of this invention, comprising: a sensor module, disposed on the cable lead-sealing structure to be detected, for transmitting an excitation signal applied to the cable lead-sealing structure, and transmitting an induced differential signal generated by the cable lead-sealing structure under an excitation magnetic field, wherein the sensor module includes a sensor array, the sensor array includes multiple differential coil units, each differential coil unit including an excitation coil and two induction coils symmetrically distributed on both sides of the excitation coil and connected in antiphase; a control module, for generating the excitation signal and transmitting it to the sensor module, and receiving the induced differential signal transmitted by the sensor module and preprocessing it, and transmitting the preprocessed induced differential signal to a host computer module; and a host computer module, for receiving the preprocessed induced differential signal, constructing a conductivity distribution image of the lead-sealing structure cross-section based on the preprocessed induced differential signal, and performing defect location, quantitative identification, and condition assessment based on the conductivity distribution image.

[0016] The cable lead sealing structure detection system provided in the second aspect has the same technical features as the cable lead sealing structure detection method provided in the first aspect, and therefore also has similar technical effects as the first aspect, which will not be described in detail here.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The above and additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a cable lead sealing structure inspection method according to an embodiment of the present invention; Figure 2 This is a structural block diagram of a cable lead sealing structure detection system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the connection and excitation acquisition between the FPGA control module and the sensor module according to a specific embodiment of the present invention; Figure 4 This is a schematic diagram of the host computer module display when the cable sealing structure is not inserted into the sensor module according to a specific embodiment of the present invention; Figure 5 This is a schematic diagram of the normal section identification of the cable sealing structure and the display of the host computer module when the normal section enters the sensor module, according to a specific embodiment of the present invention; Figure 6 This is a schematic diagram of the identification of defective sections in the cable lead sealing structure and the display of the host computer module when the defective section enters the sensor module, according to a specific embodiment of the present invention. Figure 7 This is a schematic diagram of the upper computer module display when the cable sealing structure leaves the sensor module according to a specific embodiment of the present invention.

[0019] Figure label: 100 - Cable lead sealing structure detection system; 110 - Sensor module; 120 - Control module; 130 - Host computer module. Detailed Implementation

[0020] The embodiments of the present invention are described in detail below, and the embodiments described with reference to the accompanying drawings are exemplary.

[0021] To address the common problems of weak internal defect identification and insufficient positioning accuracy in existing lead-sealed structure inspection methods, this invention provides a method and system for inspecting cable lead-sealed structures. This method enables non-contact, high-precision, visualized, and early detection and assessment of defects in lead-sealed structures, thereby improving the safety and reliability of cable operation. (See below for reference.) Figures 1-7 A method and system for detecting the lead-sealed structure of cables according to embodiments of the present invention are described.

[0022] Figure 1 This is a flowchart of a cable lead sealing structure inspection method according to an embodiment of the present invention. Figure 1 As shown, a cable lead sealing structure inspection method according to an embodiment of the present invention includes the following steps: Step S1: Apply an excitation signal to the sensor array deployed on the lead-sealed structure of the cable to be tested. The sensor array includes multiple differential coil units, each of which includes an excitation coil and two induction coils symmetrically distributed on both sides of the excitation coil and connected in series in opposite phases.

[0023] In a specific embodiment, applying an excitation signal to a sensor array deployed on the lead-sealed structure of the cable to be tested can generate an alternating magnetic field. The alternating magnetic field acts on the lead-sealed conductor of the cable, and according to Faraday's law of electromagnetic induction, it induces an electromotive force inside the conductor. The induced electromotive force drives the formation of eddy currents inside the conductor, and the eddy currents generate a secondary induced magnetic field.

[0024] In a specific embodiment, electromagnetic interference in the environment (such as on-site power frequency interference, equipment noise, and stray magnetic fields in space) acts uniformly, simultaneously, in the same direction, and with the same amplitude on the two induction coils. This is a common-mode signal. The two induction coils are symmetrically distributed on both sides of the excitation coil, with identical mechanical positions, electrical parameters, number of coil turns, and dimensions. Therefore, any environmental interference will couple into the two induction coils simultaneously with almost the same intensity. The two induction coils are connected in series in opposite phases. The environmental common-mode interference generates voltages of equal magnitude but opposite phase in the two coils. After being connected in series, they cancel each other out, resulting in a zero output.

[0025] Meanwhile, since the cracks, air gaps, and poor contact in the cable lead sealing structure are local, asymmetrical, and located near the sensor, these defects only change the magnetic field of the induction coil on one side and do not affect the coil on the other side in an equal manner. Therefore, the electromagnetic changes caused by the defects will make the signals of the two coils unequal, and the difference will not cancel each other out, thus forming an effective differential signal. The system only amplifies the defect signal and does not amplify the interference, so the interference is canceled out and the useful signal is highlighted. The weak defect signals such as tiny cracks, micro-air gaps, and poor contact that were originally submerged by noise can be clearly detected.

[0026] In other words, by using differential coil units and symmetrically arranging the induction coils in antiphase series, the common-mode interference from the environment can generate equal and opposite induced potentials in the two coils and cancel each other out, thus achieving automatic interference suppression. At the same time, the local electromagnetic changes caused by defects in the lead-sealing structure of the cable only act asymmetrically on one side of the induction coil and will not be canceled out, thereby forming an effective differential signal that can be amplified and detected. Therefore, it can significantly enhance the detection sensitivity to changes in the electromagnetic response of the lead-sealing structure and improve the ability to identify weak defect signals.

[0027] Specifically, in this embodiment of the invention, an excitation signal is first applied to a sensor array deployed on the lead-sealing structure of the cable to be inspected, providing an alternating magnetic field excitation to the cable lead-sealing structure, thereby exciting the conductor to generate eddy current effect and providing a signal basis for defect detection. Furthermore, the sensor array is configured to employ multiple differential coil units, with symmetrical anti-phase series induction coils on both sides of the excitation coil of each differential coil unit. This can effectively suppress environmental common-mode interference and retain only the useful signal brought by the defect, thereby enhancing the sensitivity of detecting changes in the electromagnetic response of the lead-sealing structure and improving the ability to identify weak defect signals.

[0028] Step S2: Collect the induced differential signal generated by the cable lead sealing structure under the excitation magnetic field, and preprocess the induced differential signal.

[0029] In a specific embodiment, the secondary induced magnetic field, after being received, can be collected as an induced differential signal for subsequent defect detection; preprocessing includes, but is not limited to, signal amplification and filtering.

[0030] Specifically, the embodiments of the present invention acquire induced differential signals under an excitation magnetic field, that is, obtain electromagnetic response signals corresponding to the conductivity distribution inside the lead-sealed structure, avoid the limitations of single signal detection, and preprocess the induced differential signals to optimize signal quality, suppress noise interference, improve the signal-to-noise ratio, and ensure subsequent imaging accuracy.

[0031] Step S3: Based on the preprocessed inductive differential signal, construct a conductivity distribution image of the lead-sealed structure cross section, and perform defect location, quantitative identification, and condition assessment based on the conductivity distribution image.

[0032] In a specific embodiment, the amplitude and phase dual-dimensional feature data contained in the preprocessed inductive differential signal can accurately reflect the conductivity variation law at different locations inside the cable lead-sealing structure. Based on this, an intuitive and visual conductivity distribution map can be constructed, fully presenting the conductivity continuity and structural integrity inside the lead-sealing area.

[0033] In a specific embodiment, based on the abrupt changes, distortions, discontinuities, and uneven distribution characteristics of conductivity values ​​in the conductivity distribution image, the existence, location, and extent of defects can be directly determined. Based on the area, shape, and degree of numerical deviation of the conductivity abnormal region, the size and degree of damage of defects can be quantitatively calculated. Combining the difference in conductivity distribution between normal lead-sealed structures and defective structures, the accurate identification and classification of typical defects such as cracks, air gaps, and poor contact can be completed, thereby achieving a comprehensive assessment of the overall operating status of the cable lead-sealed structure and early warning of potential hazards.

[0034] Specifically, in the final embodiment of the present invention, a cross-sectional conductivity distribution image is constructed based on the preprocessed induced differential signal to visualize the internal conductivity characteristics of the lead-sealed structure, thereby intuitively reflecting the defect morphology and location. Based on the conductivity image, defect location, quantitative identification, and state assessment are performed to accurately determine the defect location and the degree of defect, thus realizing a full-state quantitative assessment of the lead-sealed structure.

[0035] Therefore, the cable lead-sealing structure inspection method according to an embodiment of the present invention first applies an excitation signal to a sensor array deployed on the cable lead-sealing structure to be inspected, then acquires the induced differential signal generated by the cable lead-sealing structure under the excitation magnetic field, preprocesses the induced differential signal, and finally constructs an image of the conductivity distribution of the lead-sealing structure cross-section based on the preprocessed induced differential signal. Defect location, quantitative identification, and condition assessment are then performed based on the conductivity distribution image. Thus, by applying an excitation signal, acquiring and preprocessing the induced differential signal, constructing a conductivity distribution image, and performing defect identification and condition assessment, magnetic induction tomography is used to achieve... The visualization of the internal conductivity characteristics of the cable lead-sealed structure enables non-contact, non-destructive, and precise defect location, quantitative identification, and overall condition assessment. Furthermore, by configuring the sensor array with multiple differential coil units, each with symmetrically connected anti-phase induction coils on both sides of its excitation coil, environmental common-mode interference can be effectively suppressed, retaining only useful signals from defects. This enhances the sensitivity of detecting changes in the electromagnetic response of the lead-sealed structure and improves the ability to identify weak defect signals. Consequently, the accuracy, reliability, and early warning capabilities of cable lead-sealed structure detection are significantly improved, contributing to the safe and stable operation of high-voltage cable systems.

[0036] In one embodiment of the present invention, applying an excitation signal to a sensor array deployed on the lead-sealed structure of the cable to be tested includes: sequentially applying an output excitation signal to the excitation coils of multiple differential coil units in a time-division multiplexing manner using a single coil.

[0037] In a specific embodiment, single-coil time-division rotating excitation refers to selecting only one coil unit in the sensor array as the excitation coil at any given time, while the remaining coils act as receiving coils to synchronously acquire the induced differential signal. The excitation coils are switched sequentially according to a preset order to complete the working mode of rotating excitation and acquisition of the entire array. This can achieve omnidirectional uniform excitation of the cable lead sealing structure, eliminate detection blind spots, avoid electromagnetic interference between multiple excitation sources, ensure the consistency of detection signals, simplify the hardware structure, and improve imaging accuracy and detection efficiency.

[0038] Specifically, the embodiments of the present invention, by adopting a single-coil time-division rotating excitation method, can achieve circumferential full coverage and multi-angle uniform excitation of the cable lead sealing structure, avoid excitation blind spots, and ensure that each position of the lead sealing can be effectively excited and generate a stable eddy current response. This can improve the detection coverage integrity and signal consistency, while simplifying the hardware structure and control logic, which helps to improve detection stability and efficiency.

[0039] In one embodiment of the present invention, the acquisition of the induced differential signal generated by the cable sealing structure under the excitation magnetic field includes: when the current excitation coil is working, synchronously acquiring the induced differential signal output by the induction coil in all other differential coil units; and sequentially acquiring the induced differential signal corresponding to all excitation positions according to the excitation rotation sequence.

[0040] Specifically, the embodiments of the present invention adopt a working mode in which the differential signals of the induction of all other coils are acquired simultaneously when the current excitation coil is working, and all acquisitions are completed in the order of excitation. This achieves single-coil excitation, multi-coil synchronous induction, and multi-channel orderly acquisition, thereby maximizing the acquisition of multi-angle electromagnetic response information of the lead-sealed structure. This helps to improve data richness and imaging accuracy, while ensuring strict synchronization of acquisition timing, improving signal reliability and reconstructed image quality.

[0041] In one embodiment of the present invention, the preprocessing of the induced differential signal includes: performing differential amplification, programmable gain adjustment and low-pass filtering conditioning on the induced differential signal in sequence.

[0042] In a specific embodiment, differential amplification involves subtracting and amplifying a pair of differential signals (positive phase + negative phase) output by the sensor, amplifying only the difference between the two signals and not the common signal; this can suppress common-mode interference (on-site electromagnetic noise, power supply interference, spatial radiation interference), highlight the weak and effective signal that truly comes from the lead sealing defect, improve the signal-to-noise ratio, and make the defect signal easier to detect.

[0043] In a specific embodiment, the programmable gain adjustment automatically / programmably adjusts the amplification factor according to the strength of the current induced differential signal: it automatically amplifies when the signal is weak and automatically reduces the amplification when the signal is strong, ensuring that the signal is always within the optimal acquisition amplitude range; in this way, signals of different strengths can be stably acquired without being lost due to excessively small signals or overflowing due to excessively large signals, thereby adapting to signal changes at different locations and with different defect sizes, ensuring a wide acquisition range, good consistency, and more stable imaging.

[0044] In a specific embodiment, low-pass filtering conditioning removes high-frequency noise by filtering out high-frequency noise, retaining only the low-frequency effective inductive differential signal that reflects the lead-sealed structure. This removes high-frequency electromagnetic interference and circuit noise, making the signal smoother and cleaner, facilitating subsequent acquisition and imaging, improving image reconstruction quality, and reducing artifacts and errors.

[0045] Specifically, the embodiments of the present invention can effectively amplify weak effective signals, adaptively match responses of different intensities, and filter out high-frequency noise and environmental interference by sequentially performing differential amplification, programmable gain adjustment, and low-pass filtering on the induced differential signal, thereby significantly improving signal quality and signal-to-noise ratio, and providing a clean, stable, and high-quality data foundation for subsequent conductivity image reconstruction and defect identification.

[0046] In one embodiment of the present invention, a conductivity distribution image of a sealed lead structure cross section is constructed based on the preprocessed induced differential signal, including: reconstructing the conductivity distribution image of the sealed lead structure cross section by using linear back projection, Landweber iterative method or Tikhonov regularization method based on the amplitude data and phase data of the induced differential signal.

[0047] In a specific embodiment, the preprocessed induced differential signal contains amplitude and phase dual-dimensional feature data, which can accurately reflect the conductivity variation law at different locations inside the cable lead-sealed structure. The multi-channel induced data can be integrated into a complete measurement matrix, and the electromagnetic inverse problem can be solved using linear back projection, Landweber iteration, or Tikhonov regularization algorithm as the core. Through spatial mapping and numerical iteration, the conductivity distribution image of the cable lead-sealed structure cross section is reconstructed, transforming the abstract electromagnetic signal into an intuitive and visual conductivity distribution map, fully presenting the conductivity continuity and structural integrity inside the lead-sealed area.

[0048] Specifically, the linear back-projection algorithm is the simplest and fastest magnetic induction tomography image reconstruction algorithm. It projects the electromagnetic response obtained by each excitation receiving channel back along the ray direction to the cross-sectional area. The projections of all channels are superimposed to form a complete cross-sectional distribution. The linear back-projection algorithm has an extremely fast imaging speed, is suitable for real-time detection, and has a simple structure and low computational load. It can quickly obtain the overall conductivity distribution of the lead-sealed cross section and make a preliminary judgment on the presence or absence of defects.

[0049] Specifically, the Landweber iterative algorithm is an optimization-type imaging algorithm that gradually approximates the true conductivity distribution through multiple iterations. It repeatedly corrects the image, making the reconstructed result closer and closer to the real structure. The specific process includes "establishing a forward field model, calculating the response error of the current image, updating the image along the gradient descent direction, and repeating the iteration until the error is small enough." The Landweber iterative algorithm is clearer and has fewer artifacts than linear backprojection images. At the same time, the defect edges are more accurate and the positioning accuracy is higher, making it suitable for lead sealing defect detection that requires high imaging quality.

[0050] Specifically, the Tikhonov regularization algorithm is a constrained stable reconstruction algorithm specifically designed to solve ill-posed problems in magnetic induction tomography (noise amplification, image oscillation). By adding regularization constraints during the solution process, it balances the fitting data and image smoothness, suppressing noise amplification and making the image smoother and more stable. The Tikhonov regularization algorithm has the strongest anti-interference ability, producing stable images even with high noise levels. It can effectively suppress artifacts, jitter, and false defects, ensuring more accurate quantitative analysis of defects and preventing misjudgments of size due to noise.

[0051] Specifically, the embodiments of the present invention reconstruct conductivity distribution images based on the amplitude and phase data of the inductive differential signal using linear back projection, Landweber iteration, or Tikhonov regularization algorithms. This allows for the selection of appropriate algorithms according to different detection scenarios and accuracy requirements, improving image resolution and anti-distortion capabilities while ensuring imaging speed. It enables accurate presentation and quantitative analysis of defect location, shape, and size, thereby helping to improve the accuracy of defect identification, location, and quantification.

[0052] In one embodiment of the present invention, after constructing the conductivity distribution image of the lead-sealed structure cross-section based on the preprocessed differential induction signal, the method further includes: extracting multi-dimensional features based on the preprocessed differential induction signal to generate a trend curve reflecting the overall state of the lead-sealed structure changing over time, used to determine the detection state of the lead-sealed structure; generating an instantaneous response distribution feature map based on the preprocessed differential induction signal to reflect local abnormal changes in the lead-sealed structure; performing response intensity statistical analysis based on the preprocessed differential induction signal to determine the signal sensitivity distribution characteristics, used for sensor array weight allocation and optimization; and performing spatiotemporal dimension distribution calculation based on the preprocessed differential induction signal to generate a two-dimensional distribution image, used to distinguish the response differences between normal and defective structures.

[0053] In a specific embodiment, based on the differential signals from all channels, a comprehensive index representing the overall health of the lead seal is calculated. This index is plotted as a continuously changing curve over time, automatically determining whether the lead seal has entered or left the detection area. It visually reflects whether the overall structure is abnormal or defective, achieving automatic monitoring and automatic alarm for any anomalies. Specifically, this involves normalizing, smoothing, filtering, and statistically fusing the differential signals from each channel to form a global change index. The curve is stable when there are no defects, and abruptly changes when defects enter or leave the area.

[0054] In a specific embodiment, the induction intensity of all effective channels at a given moment is displayed in real time using a bar chart or channel distribution map. This allows for quick identification of any anomalies in the current lead sealing area, visually pinpointing which angle / channel is experiencing the anomaly, and facilitating rapid on-site determination of the approximate location of the defect. Specifically, the electromagnetic responses of all channels to the lead sealing structure at a given moment are simultaneously collected and displayed. A normal structure is characterized by uniform responses across all channels, while a defect location is characterized by a significant abrupt change in the response of the corresponding channel.

[0055] In a specific embodiment, the response strength of all channels in the defect / normal segment is statistically analyzed to determine which channel is the most sensitive and which contributes the most, forming a channel sensitivity ranking. This identifies the key channel that best identifies defects, which is then used for subsequent sensor array optimization and weight allocation, and can also improve the overall detection accuracy and efficiency of the system. Specifically, this involves quantifying the sensitivity of each channel to defects by statistically analyzing the response amplitude, fluctuation range, and anomaly contribution rate of each channel.

[0056] In a specific embodiment, spatial angle is used as one dimension, and time / scan frame as another dimension to create a two-dimensional grayscale / color image. This visually distinguishes the response differences between normal and defective segments. Defects appear as distinct color blocks / bright bands / dark bands, providing strong visualization and facilitating manual interpretation, archiving, and comparative analysis. Specifically, the response intensity of each channel and each frame is converted into grayscale or color, with the horizontal axis representing time / scan position and the vertical axis representing the annular spatial angle, forming a spatial-temporal two-dimensional response distribution map.

[0057] Specifically, the embodiments of the present invention further generate an overall state trend curve, an instantaneous response distribution feature map, a signal sensitivity distribution feature, and a spatiotemporal two-dimensional distribution image based on the preprocessed signal. The lead-sealing structure is comprehensively analyzed from four dimensions: overall trend, local changes, channel contribution, and spatiotemporal distribution. This enables automatic determination of the detection status, rapid identification of local anomalies, guidance for sensor array optimization, and intuitive distinction between normal and defective structures. It improves the detection and evaluation system and enhances the system's intelligence level and engineering practicality.

[0058] In one embodiment of the present invention, the excitation signal is a 1MHz–10MHz alternating sinusoidal excitation signal.

[0059] In a specific embodiment, significant eddy currents can only be generated inside the conductor under a high-frequency alternating magnetic field. The 1MHz–10MHz range is within the quasi-static eddy current field range, where the electromagnetic field propagation effect is negligible, satisfying the imaging theoretical model and ensuring the accuracy of the reconstructed conductivity image. Simultaneously, due to the skin effect, the conductor eddy currents are concentrated only on the surface layer. Cable lead sealing defects appear precisely at the interface between the surface layer and the bonding layer, and the skin depth of 1MHz–10MHz just covers the defect area. Furthermore, the phase change of the induced differential signal is significant in the 1MHz–10MHz high-frequency band, making it extremely sensitive to minute changes in conductivity and capable of capturing early, weak defects such as cracks, air gaps, and poor contact. Specifically, the excitation signal is preferably a 10MHz alternating sinusoidal excitation signal. At this frequency, the skin depth is the smallest, which allows the eddy currents to be most concentrated on the lead seal surface. This is also the location where cracks, poor solder joints, and gaps are most likely to occur, meaning that the defect detection sensitivity is the highest. Furthermore, the higher the frequency, the more sensitive the phase is to conductivity. At 10MHz, the phase of the intact area is stable, while the phase of the defective area changes drastically, resulting in the highest image contrast and the clearest defect outline. At the same time, 10MHz can excite the strongest eddy currents and secondary magnetic fields, ensuring that the acquired signal amplitude and signal-to-noise ratio reach the optimal range.

[0060] Specifically, the embodiments of the present invention employ an alternating sinusoidal excitation signal of 1MHz-10MHz, which conforms to the quasi-static eddy current field theory and skin effect law of magnetic induction tomography, so that the eddy currents are concentrated on the surface of the cable lead seal structure and the key bonding interface, thereby improving the detection sensitivity and imaging resolution of internal defects of the lead seal, thus ensuring the balance between detection depth and accuracy, and adapting to the actual detection needs of the lead seal structure.

[0061] In one embodiment of the present invention, the sensor array is fixed as a ring array by a non-metallic flexible substrate, which is adapted to fit the surface of the cable lead sealing structure.

[0062] In a specific embodiment, the sensor array is fabricated using a non-metallic flexible substrate and fixed in a ring array, which can tightly surround and conform to the irregular circular curved surface of the cable's lead-sealing structure. Specifically, the non-metallic substrate avoids the generation of eddy currents and magnetic field interference, ensuring that the excitation magnetic field and the induced differential signal are not shielded and do not become distorted; the flexible structure can adapt to the cable's shape, maintaining a stable and uniform lift-off distance between each coil unit and the lead-sealing surface, eliminating signal errors caused by gap fluctuations, and improving detection consistency and repeatability; the ring array layout can achieve 360° full coverage excitation and signal acquisition of the cable's lead-sealing structure, eliminating detection blind spots, providing sufficient multi-angle response data for cross-sectional conductivity imaging, thereby improving the integrity, stability, and imaging accuracy of defect detection.

[0063] Specifically, the embodiments of the present invention use a non-metallic flexible substrate to form a ring sensor array, which can closely fit the irregular curved surface of the cable lead sealing structure, maintain a stable and uniform lift-off distance between the sensor and the measured surface, reduce signal fluctuations and errors caused by poor fitting, improve the consistency, stability and repeatability of the detection signal, thereby enhancing the adaptability to cable lead sealing structures of different specifications and shapes.

[0064] A further embodiment of the present invention discloses a cable lead seal structure detection system for implementing the cable lead seal structure detection method as described in the above embodiments of the present invention. Figure 2 This is a structural block diagram of a cable lead sealing structure detection system according to an embodiment of the present invention. Figure 2 As shown, in one embodiment of the present invention, the cable lead sealing structure detection system 100 includes: a sensor module 110, a control module 120, and a host computer module 130.

[0065] Specifically, the sensor module 110 is disposed on the lead-sealing structure of the cable to be tested, and is used to transmit the excitation signal applied to the lead-sealing structure of the cable, as well as the induced differential signal generated by the lead-sealing structure of the cable under the excitation magnetic field. The sensor module 110 includes a sensor array, which includes multiple differential coil units. Each differential coil unit includes an excitation coil and two induction coils symmetrically distributed on both sides of the excitation coil and connected in series in opposite phases. The control module 120 is used to generate an excitation signal and transmit it to the sensor module, as well as to receive the differential sensing signal transmitted by the sensor module, perform preprocessing, and transmit the preprocessed differential sensing signal to the host computer module. The host computer module 130 is used to receive the pre-processed inductive differential signal, construct a conductivity distribution image of the lead-sealed structure cross section based on the pre-processed inductive differential signal, and perform defect location, quantitative identification and condition assessment based on the conductivity distribution image.

[0066] It should be noted that the specific implementation of the cable lead seal structure detection system 100 in this embodiment of the invention is similar to the specific implementation of the cable lead seal structure detection method described in the above embodiment of the invention, and therefore has similar technical effects. For details, please refer to the description of the cable lead seal structure detection method section. To reduce redundancy, it will not be repeated here.

[0067] The cable lead seal structure detection method and system of the present invention described above will be explained below with reference to a specific embodiment. In this specific embodiment, a cable lead seal structure detection system is provided.

[0068] It should be noted that the Altera Cyclone III series chips provided in this specific embodiment for building a high-speed parallel processing platform for FPGA (Field-Programmable Gate Array) are only one design under this specific embodiment. For example, the AD9754 and ULN2803 chips are used for time-division multiplexing of excitation signals, and the AD9648 and THS7001 chips are used for signal conditioning, etc. The protection scope of the cable lead sealing structure detection system provided in this specific embodiment is not limited to the specific hardware model design under this specific embodiment.

[0069] In this specific embodiment, the cable lead sealing structure detection system includes a sensor module (i.e., sensor module 110 in the above embodiment of the present invention), an FPGA control module (i.e., control module 120 in the above embodiment of the present invention), and a host computer module (i.e., host computer module 130 in the above embodiment of the present invention).

[0070] Specifically, the FPGA control module is used to realize the cyclic emission of sinusoidal excitation signals and the acquisition of induced voltage signals in the megahertz frequency band. It constructs a high-speed parallel processing platform based on the Altera Cyclone III series chips, integrating DDS (Direct Digital Frequency Synthesis), time-division gating, and differential acquisition mechanisms to establish a multi-channel high-precision architecture for cyclic emission of excitation signals and acquisition of induced differential signals. The FPGA control module is electrically connected to the sensor module through a coaxial cable to transmit the sinusoidal excitation signal and induced voltage signal of the required frequency. The FPGA control module communicates with the host computer module through a data transmission interface to transmit a 14-bit ADC (Analog-to-Digital Converter) differential digital signal, which is obtained by processing the induced voltage signal by the FPGA control module.

[0071] Specifically, the FPGA control module has sub-modules deployed below it. The sub-modules include an excitation signal generation module, a signal conditioning module, and a data acquisition and transmission module. The excitation signal generation module and the signal conditioning module are electrically connected.

[0072] Specifically, the excitation signal generation module is used to generate a sinusoidal excitation signal with a frequency of 10MHz required by the cable lead sealing structure defect detection system. The selection of the 10MHz excitation frequency is determined based on the quasi-static eddy current field theory of magnetic induction tomography and the skin effect principle. At high frequencies, the secondary induced magnetic field is mainly characterized by phase change, and the skin depth decreases as the frequency increases. The 10MHz excitation frequency can concentrate the eddy current on the lead sealing area. The DDS signal generation scheme, which features anti-interference, low power consumption, and phase controllability, is selected. The core chip is a high-performance, low-power CMOS (Complementary Metal-Oxide-Semiconductor) digital-to-analog converter AD9754. Controlled by the FPGA chip, the AD9754 can output a 10MHz sine wave excitation signal based on the megahertz-level main frequency emitted by the FPGA phase-locked loop module. The signal is then processed by differential-to-single-ended conversion and second-order low-pass filtering. The ADG1207 and ULN2803 chips are used to form a switch selection array. The FPGA controls eight dual-input channels to achieve high-speed rotation. Different states are selected by a counter that increments by one with each clock cycle. The default state is the latch of the current state. Each state corresponds to one of the eight channels, and the selected channel is selected, realizing the time-division multiplexing excitation and reception rotation strategy of the sensor coil.

[0073] Specifically, the signal conditioning module is used to filter and improve the electrically induced differential signal. The lead-sealed structure under test generates a secondary eddy current induced differential signal under the action of the excitation magnetic field. This induced differential signal is synchronously acquired by seven receiving coils. The received signal is then sequentially converted from differential to single-ended and amplified by an AD8429 instrumentation amplifier, and then amplified by a THS7001 programmable gain amplifier and Butterworth low-pass filter. This achieves single-ended processing and filtering of the complementary differential signal output from the front-end differential coils, improving signal transmission quality.

[0074] Specifically, the data acquisition and transmission module samples the received analog signal into a digital signal recognizable by the host computer and returns it to the host computer via a data transmission protocol. The conditioned received signal is sent to the AD9648 analog-to-digital converter for 14-bit differential digital signal conversion. A complete measurement cycle is formed by sequentially acquiring data from 8 excitation channels and 7 receiving channels, resulting in 56 original measurement combinations (8×7). After deduplication and geometric symmetry optimization, these are mapped to 28 independent and effective measurement channels. The acquired digital signals are uploaded to the host computer via a USB 2.0 interface. After amplitude and phase information extraction, they are used for imaging inversion calculations. For a single acquisition cycle set to T = 25×n milliseconds, the acquired data volume is 28×2×n, where 28 is the number of independent and effective measurement channels after system optimization, 2 represents the simultaneous output of amplitude and phase data for each channel, and n is the total number of frames (rounds) acquired.

[0075] Figure 3 This is a schematic diagram illustrating the connection and excitation acquisition of an FPGA control module and a sensor module according to a specific embodiment of the present invention. In the diagram, the Tx terminal provides an excitation signal to the excitation coil, and the Rx terminal acquires the induced voltage from the receiving coil. Figure 3 As shown, the sensor module is used to transmit the sinusoidal signal emitted by the FPGA control module to excite the cable sealing structure, and simultaneously receive the electrical signal induced by the cable sealing structure. The sensor module consists of eight independent coil units. Each coil unit adopts a differential structure design, that is, three identical coils (parameters: inner diameter 27mm, wire width 0.25mm, number of turns 9, wire spacing 0.25mm) are placed coaxially and at equal intervals perpendicularly. The middle coil serves as the excitation coil, with a layer spacing of 0.25mm between it and the receiving coils on both sides. Its positive and negative terminals are connected to the excitation signal generation module to receive the sinusoidal excitation signal. The negative terminal of one coil is connected to the positive terminal of the other coil. The positive terminals of one coil and the negative terminals of the other coil are connected to the FPGA control module to suppress spatial distortion. The sensor module is made of flexible printed circuit board technology. It consists of eight flexible differential coil units fixed at equal intervals by a non-metallic thin film. The eight coil units together form a circular detection range with a diameter of 114mm, forming a ring array layout that can surround the cable lead seal structure. It can closely fit the irregular curved surface of the cable lead seal structure. Each coil unit maintains a lift distance of 1mm from the surface of the cable lead seal structure, with an allowable fluctuation range of ±0.25mm, to ensure stable and consistent detection signal.

[0076] Specifically, the host computer module receives the 14-bit ADC differential digital signal and generates a complete measurement voltage matrix. This matrix serves as input data for the magnetic induction tomography image reconstruction algorithm. After inversion calculation by the host computer module, the conductivity distribution image of the cable lead-sealed structure can be reconstructed. The host computer module employs algorithms such as linear back projection, Landweber iteration, and Tikhonov regularization to reconstruct and analyze the induction data. The reconstructed conductivity distribution image of the cable lead-sealed structure cross-section reflects the conductivity characteristics, enabling the location and quantitative analysis of defects in the cable lead-sealed structure. Simultaneously, it evaluates the reconstruction capability and error sensitivity of the image quality under each algorithm, providing a comparative basis and technical reserves for subsequent FPGA-based image processing optimization.

[0077] More specifically, the host computer module can also generate a trend curve of the overall state index of the measured object changing over time, which is used to determine whether the measured object has entered the sensor and the duration of the change. By selecting the detection start point and end point to determine the reference interval, calculating the median and standard deviation of the response signals of each channel, constructing a reference statistical model, and performing standardization and smoothing filtering, a "measured object overall state index - time" curve is formed.

[0078] More specifically, the host computer module can also display and analyze the response signals of each channel at any moment during the detection process in real time. By analyzing the amplitude and phase fluctuation characteristics of the induced differential signal, it presents the instantaneous response of each channel to the test object in the form of a bar chart, completing the visualization of the channel spatial sensitivity distribution, which is used to quickly observe signal abrupt changes and changes in the local characteristics of the test object.

[0079] More specifically, the host computer module can also quantify the sensitivity of different channels to changes in the structure of the object under test, which is used for subsequent sensor array weight allocation and optimization. Based on the exponential curve and adaptive threshold, an event flag is generated to obtain a set of effective abnormal frames. The mean absolute value of the signal of each channel on the abnormal frames after standardization is extracted as a sensitivity index to reflect the response intensity of each channel to defects. The top eight sensitive channels are selected to show their contribution.

[0080] More specifically, the host computer module can also present the distribution characteristics of the response signal in the spatial and temporal dimensions as a two-dimensional image, which can be used to intuitively reflect the difference in spatial response between intact and defective structures. Energy calculation is performed on the response signal along the time dimension to form a two-dimensional energy image with the spatial angle as the vertical axis and the scan frame as the horizontal axis, while grayscale variations are set to enhance contrast.

[0081] The following combination Figures 4-7 Describe the different states of the host computer module, in Figures 4-7In the demo, MIT represents the interface for Magnetic Induction Tomography (MMT). Frame represents frame data, HI represents the Health Index, characterizing the overall state of the object under test, Global change indicator represents the global change index curve, reflecting the trend of the overall state of the object under test over time, Current 28-channel fingerprint represents the instantaneous response fingerprint of the current 28 channels, displaying the real-time status of the sensing signals of each channel, Most sensitive channels represent the most sensitive channel response histogram, quantifying the sensitivity and contribution of each channel to defect signals, and MIT circumferential B-scan like map represents the circular B-scan distribution map of MMT, visually presenting the electromagnetic response distribution of the object under test along the circumferential direction. The vertical axis, Health Index, represents the health index, corresponding to the quantified value of the global change index; the horizontal axis, Frame, represents the scan frame, corresponding to the scan sequence number in the time dimension; the horizontal axis, Channel, represents the channel, corresponding to the detection channel number of the sensor array; and the vertical axis, Circumferential angle (deg), represents the circumferential angle (unit: degrees), corresponding to the circumferential position of the object under test.

[0082] It should be noted that, Figures 4-7 This is the actual display interface state of the host computer module; therefore, the following should be... Figures 4-7 Each image is treated as an independent display interface, rather than each attached image being a simple combination or patchwork of multiple images.

[0083] Figure 4 This is a schematic diagram of the host computer module display when the cable sealing structure is not inserted into the sensor module, according to a specific embodiment of the present invention. Figure 4 As shown, when the cable lead sealing structure does not enter the sensor detection area, the global change index value displayed by the host computer is close to the benchmark value, the response signal of each channel does not fluctuate significantly, the instantaneous response fingerprint of 28 channels remains stable, the sensitive channel has no significant response, and the ring-shaped B-scan distribution image has no abnormal features. This state indicates that no target is currently entering the detection area, the system is in a stable standby state, and can be used as a benchmark reference for subsequent detection.

[0084] Figure 5 This is a schematic diagram of the normal section identification of the cable sealing structure and the display of the host computer module when the normal section enters the sensor module, according to a specific embodiment of the present invention. Figure 5As shown, when a normal section of the cable with an intact lead-sealing structure enters the sensor detection area, the global change index of the host computer increases significantly. The 28-channel instantaneous response fingerprint shows a uniform and symmetrical distribution. The response intensity of each channel is consistent and there are no abrupt changes. The ring-shaped B-scan distribution image shows a continuous, smooth, and regular response distribution with no local distortion or abnormal areas. This state indicates that the lead-sealing structure has continuous conductivity, good interface bonding, and no defects.

[0085] Figure 6 This is a schematic diagram of the identification of defective sections in the cable lead sealing structure and the display of the upper computer module when the defective section enters the sensor module, according to a specific embodiment of the present invention. Figure 6 As shown, when a defective section of the cable lead sealing structure, which has defects such as cracks, air gaps, or poor contact, enters the sensor detection area, the global change index of the host computer shows characteristic fluctuations, the 28-channel instantaneous response fingerprint shows obvious distortion, sudden changes in local peaks, or depressions, the sensitive channel shows a significant response, and the annular B-scan distribution image shows obvious abnormal bright spots, dark bands, or discontinuous distortion areas, which are significantly different from the distribution characteristics of the normal section. This state can intuitively and accurately identify the location and degree of abnormality of the defect, and realize the location and qualitative judgment of the defect.

[0086] Figure 7 This is a schematic diagram of the upper computer module display when the cable sealing structure leaves the sensor module according to a specific embodiment of the present invention. Figure 7 As shown, when the cable lead sealing structure completely leaves the sensor detection area, the global change index of the host computer quickly returns to the baseline level, the 28-channel instantaneous response fingerprint returns to a stable and unfluctuating state, the response of the sensitive channel disappears, and the ring-shaped B-scan distribution image returns to a featureless baseline background; this state indicates that the target under test has left the detection area, the system returns to standby state, and a complete detection process is completed.

[0087] In summary, the cable lead sealing structure detection method and system of the present invention have the following significant advantages compared with related technologies: 1. Improved accuracy and reliability of detection: Based on the principle of magnetic induction tomography, it directly reflects the internal conductivity distribution of the cable lead sealing structure, outputs a true physical mapping rather than relying on defect model judgment, avoids misjudgment of unknown defects from the root, and can accurately locate, quantitatively identify and assess the condition of defects such as cracks, air gaps and poor contact, significantly improving detection accuracy and early warning capability.

[0088] 2. Significantly enhanced anti-interference and sensitivity: The differential coil unit and anti-phase series structure, combined with differential amplification, filtering and conditioning and other preprocessing processes, effectively suppress environmental common-mode interference, amplify the electromagnetic response signal of weak defects, and greatly improve the system signal-to-noise ratio and the sensitivity of small defect detection. It is suitable for stable detection in complex electromagnetic environments.

[0089] 3. Improved adaptability and detection stability: The flexible ring sensor array can closely fit the irregular curved surface of the cable seal, maintain a stable lift distance, eliminate signal fluctuations caused by poor fit, adapt to different specifications of cable seal structures, and greatly improve the consistency, stability and repeatability of detection signals.

[0090] 4. Improved detection coverage and efficiency: The single-coil time-division rotating excitation and multi-coil synchronous induction acquisition mode are adopted to achieve 360° full coverage detection of the lead-sealed structure, with no excitation or detection blind spots. Combined with FPGA high-speed parallel processing, it can realize rapid acquisition and imaging of multi-channel signals, significantly improving detection efficiency and data integrity.

[0091] 5. Enhanced Analytical Dimensions and Intelligence: Generates multi-dimensional results such as cross-sectional conductivity imaging, overall state trend curves, instantaneous response distribution, channel sensitive characteristics, and spatiotemporal two-dimensional distribution. It also has functions such as defect location and quantification, state monitoring, anomaly identification, and sensor optimization, forming a complete detection and evaluation system with enhanced intelligence and engineering practicality.

[0092] 6. Safety and non-destructive testing advantages: The non-contact electromagnetic testing method eliminates the need for disassembly, does not damage the lead sealing structure, and does not affect the normal operation of the cable, enabling all-weather online non-destructive testing and significantly improving the safety and convenience of operation and maintenance.

[0093] 7. Improved system integration and practicality: The system adopts a modular integrated design of sensor module, FPGA control module and host computer module to realize full automation of excitation, sensing, conditioning, acquisition, imaging and evaluation. It has a simple structure, stable operation and convenient deployment, and can be directly used for actual on-site operation and maintenance testing.

[0094] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.

[0095] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for detecting the lead-sealing structure of cables, characterized in that, Includes the following steps: An excitation signal is applied to a sensor array deployed on the lead-sealed structure of the cable to be tested. The sensor array includes multiple differential coil units, each of which includes an excitation coil and two induction coils symmetrically distributed on both sides of the excitation coil and connected in series in opposite phases. The induced differential signal generated by the cable lead sealing structure under the excitation magnetic field is collected, and the induced differential signal is preprocessed. Based on the preprocessed inductive differential signal, a conductivity distribution image of the lead-sealed structure cross section is constructed, and defect location, quantitative identification, and condition assessment are performed based on the conductivity distribution image.

2. The cable lead sealing structure inspection method according to claim 1, characterized in that, Applying an excitation signal to the sensor array deployed on the lead-sealing structure of the cable to be tested includes: The output excitation signal is applied sequentially to the excitation coils of multiple differential coil units using a single-coil time-division rotating method.

3. The cable lead sealing structure inspection method according to claim 1 or 2, characterized in that, The acquisition of the induced differential signal generated by the cable lead sealing structure under the excitation magnetic field includes: While the current excitation coil is working, the induced differential signals output by the induction coils in all the other differential coil units are simultaneously acquired; Following the excitation rotation sequence, the inductive differential signals corresponding to all excitation positions are acquired sequentially.

4. The cable lead sealing structure inspection method according to claim 1, characterized in that, The preprocessing of the inductive differential signal includes: The induced differential signal is sequentially subjected to differential amplification, programmable gain adjustment, and low-pass filtering conditioning.

5. The cable lead sealing structure inspection method according to claim 1, characterized in that, The step of constructing a conductivity distribution image of the lead-sealed structure cross-section based on the preprocessed induced differential signal includes: Based on the amplitude and phase data of the induced differential signal, the conductivity distribution image of the lead-sealed structure cross section is reconstructed by using linear back projection, Landweber iteration method, or Tikhonov regularization method.

6. The cable lead sealing structure inspection method according to claim 1, characterized in that, After constructing the conductivity distribution image of the lead-sealed structure cross-section based on the preprocessed differential induction signal, the following steps are also included: Multi-dimensional features are extracted from the preprocessed differential induction signal to generate a trend curve reflecting the overall state of the lead-sealed structure over time, which is used to determine the detection status of the lead-sealed structure. An instantaneous response distribution feature map is generated based on the preprocessed inductive differential signal to reflect local abnormal changes in the lead-sealing structure; Statistical analysis of response intensity is performed on the preprocessed inductive differential signal to determine the signal sensitivity distribution characteristics, which are then used for sensor array weight allocation and optimization. The spatiotemporal dimension distribution is calculated based on the preprocessed inductive differential signal to generate a two-dimensional distribution image, which is used to distinguish the response differences between normal and defective structures.

7. The cable lead sealing structure inspection method according to claim 1, characterized in that, The excitation signal is an alternating sinusoidal excitation signal of 1MHz–10MHz.

8. The cable lead sealing structure inspection method according to claim 1, characterized in that, The sensor array is fixed in a ring shape by a non-metallic flexible substrate, adapting to the surface of the cable lead sealing structure.

9. A cable lead seal structure testing system for implementing the cable lead seal structure testing method as described in any one of claims 1-8, characterized in that, include: A sensor module is installed on the lead-sealing structure of the cable to be tested. It is used to transmit the excitation signal applied to the lead-sealing structure of the cable and the induced differential signal generated by the lead-sealing structure of the cable under the excitation magnetic field. The sensor module includes a sensor array, which includes multiple differential coil units. Each differential coil unit includes an excitation coil and two induction coils that are symmetrically distributed on both sides of the excitation coil and connected in series in opposite phases. The control module is used to generate the excitation signal and transmit it to the sensor module, and to receive the inductive differential signal transmitted by the sensor module, perform preprocessing, and transmit the preprocessed inductive differential signal to the host computer module. The host computer module is used to receive the pre-processed inductive differential signal, construct a conductivity distribution image of the lead-sealed structure cross section based on the pre-processed inductive differential signal, and perform defect location, quantitative identification, and status assessment based on the conductivity distribution image.

Images