Degassing performance evaluation methods, apparatus, equipment, media and procedures products

Abstract

This application relates to a method, apparatus, computer device, computer-readable storage medium, and computer program product for evaluating degassing effectiveness. The method includes: acquiring a time-domain reflectometry signal obtained by performing a terahertz time-domain spectral scan on a cable under test, the cable including an insulation layer; determining the bubble distribution in the insulation layer based on the time-domain reflectometry signal when bubbles are identified in the insulation layer; acquiring trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics of the insulation layer; and evaluating the degassing effectiveness of the cable under test based on the bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics. This method can improve the accuracy of evaluating the degassing effectiveness of cable insulation layers.

Description

Technical Field

[0001] This application relates to the field of cable manufacturing technology, and in particular to a method, apparatus, equipment, medium, and procedure for evaluating degassing effect. Background Technology

[0002] With the development of power cable manufacturing technology and operational reliability requirements, all types of cross-linked insulated cables must undergo degassing treatment during production to remove residual cross-linking byproducts in the insulation layer, ensuring the stability and reliability of the cable's insulation performance during long-term operation.

[0003] In traditional technologies, the evaluation of degassing effectiveness mainly relies on destructive sampling and testing. That is, samples are taken from the finished cable and analytical methods such as gas chromatography are used to determine the residual content of by-products in the samples.

[0004] However, the above methods not only damage the integrity of the cable product, but also only obtain local and limited information about the insulation layer, making it difficult to comprehensively and accurately assess the overall degassing state of the entire cable insulation layer, resulting in low accuracy in assessing the degassing effect of the cable insulation layer. Summary of the Invention

[0005] Therefore, it is necessary to provide a degassing effect evaluation method, device, equipment, medium, and procedure product that can improve the accuracy of the evaluation of the degassing effect of cable insulation layer in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for evaluating degassing effectiveness, including:

[0007] The time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0008] If bubbles are detected in the insulation layer based on the time-domain reflection signal, the distribution of bubbles in the insulation layer is determined based on the time-domain reflection signal.

[0009] To obtain the trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulating layer;

[0010] The degassing effect of the cable under test is evaluated based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0011] Secondly, this application also provides a degassing effect evaluation device, comprising:

[0012] The signal acquisition module is used to acquire the time-domain reflection signal, which is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0013] The bubble distribution determination module is used to determine the bubble distribution in the insulation layer based on the time-domain reflection signal when bubbles are identified in the insulation layer.

[0014] The feature acquisition module is used to acquire the trap features, space charge distribution features, and cross-linking byproduct features of the insulating layer;

[0015] The evaluation module is used to evaluate the degassing effect of the cable under test based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0016] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0017] The time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0018] If bubbles are detected in the insulation layer based on the time-domain reflection signal, the distribution of bubbles in the insulation layer is determined based on the time-domain reflection signal.

[0019] To obtain the trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulating layer;

[0020] The degassing effect of the cable under test is evaluated based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0021] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0022] The time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0023] If bubbles are detected in the insulation layer based on the time-domain reflection signal, the distribution of bubbles in the insulation layer is determined based on the time-domain reflection signal.

[0024] To obtain the trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulating layer;

[0025] The degassing effect of the cable under test is evaluated based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0026] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0027] The time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0028] If bubbles are detected in the insulation layer based on the time-domain reflection signal, the distribution of bubbles in the insulation layer is determined based on the time-domain reflection signal.

[0029] To obtain the trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulating layer;

[0030] The degassing effect of the cable under test is evaluated based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0031] The aforementioned degassing effect evaluation method, apparatus, computer equipment, computer-readable storage medium, and computer program product first utilize a terahertz time-domain spectroscopy system to comprehensively scan the cable under test, acquiring time-domain reflection signals that reflect the internal structure of the insulation layer. When bubbles are identified in the insulation layer based on the time-domain reflection signals, the bubble distribution within the insulation layer is further determined based on signal characteristics. By fully utilizing the sensitivity of terahertz waves to changes in dielectric structure, non-destructive detection and spatial localization of bubbles that may remain from the degassing process are achieved. Furthermore, trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulation layer are introduced to characterize the state of the insulation layer after degassing from multiple dimensions, considering both the electrical properties and chemical residues of the insulating material. Finally, by comprehensively analyzing and judging the bubble distribution in conjunction with the aforementioned multi-dimensional characteristics, an accurate evaluation of the overall degassing effect of the cable under test is achieved. Compared to traditional destructive sampling and testing methods that can only obtain local and single indicators, this application integrates multi-source information on physical structure, electrical properties and chemical composition to perform a comprehensive digital scan and characterization of the insulation layer. This overcomes the one-sidedness of single-dimensional and local area evaluation and improves the comprehensiveness and accuracy of the evaluation of the degassing effect of cable insulation. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating a degassing effect evaluation method in one embodiment of this application;

[0034] Figure 2 This is a schematic diagram of a terahertz pulse propagation process in one embodiment of this application;

[0035] Figure 3 This is a schematic diagram of a time-domain reflection signal in one embodiment of this application;

[0036] Figure 4 This is a graph showing the axial uniformity of the bubble density in the insulating layer in one embodiment of this application.

[0037] Figure 5 This is a flowchart illustrating the degassing effect evaluation method in another embodiment of this application;

[0038] Figure 6 This is a structural block diagram of a degassing effect evaluation device in one embodiment of this application;

[0039] Figure 7 This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0041] The degassing effect evaluation method provided in this application embodiment can be applied to computer equipment, which can be a terminal or a server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle devices, projection equipment, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0042] Cross-linked insulated cables are power cables that use cross-linked materials such as cross-linked polyethylene as the insulation layer. During the production of cross-linked insulated cables, the molecular structure of the insulation material is transformed from linear to network through a cross-linking process, which significantly improves the cable's heat resistance, mechanical strength, and electrical performance. In this production process, the insulation material, as the key carrier for realizing the cable's insulation function, has a performance quality closely related to the sufficiency and uniformity of its cross-linking reaction.

[0043] Taking XLPE (Cross-linked Polyethylene) cables as an example, the cross-linking reaction of XLPE produces low-molecular-weight cross-linking byproducts such as cumyl alcohol, acetophenone, and α-methylstyrene. The residue of these byproducts severely degrades the cable's insulation performance, leading to a large accumulation of space charge, increased dielectric loss, reduced partial discharge initiation voltage, and accelerated initiation and growth of electrical trees, ultimately threatening the cable's long-term service life and reliability. Therefore, before cross-linked insulated cables leave the factory, they must undergo rigorous degassing treatment, that is, under controlled temperature and environmental conditions, to promote the diffusion and escape of these byproducts from the insulation layer.

[0044] However, the current degassing process for cross-linked insulated cables faces many challenges.

[0045] In setting parameters such as degassing time, degassing temperature, and degassing process parameters, manufacturers often rely on their accumulated experience and lack unified scientific standards. To avoid risks, manufacturers often adopt overly conservative, excessively long degassing times, sometimes lasting for dozens of days. This not only severely hinders production efficiency but also significantly increases energy costs.

[0046] In evaluating degassing effectiveness, existing methods are mostly destructive, offline sampling and testing methods. For example, gas chromatography analysis is performed on slices of cable insulation to accurately measure the content of byproducts, or breakdown field strength tests are conducted. These methods cannot be applied to the comprehensive evaluation of long-length, large-section cables and compromise the integrity of the product.

[0047] The uniformity of degassing is also a key issue that is difficult to control. Studies have shown that due to differences in diffusion paths, the by-product removal efficiency varies significantly at different locations in the insulation layer, such as the inner, middle, and outer layers.

[0048] Furthermore, the risks of excessive degassing are often overlooked. Studies indicate that longer degassing time is not necessarily better; excessively long degassing times may lead to a decrease in the crystallinity and alteration of the microstructure of XLPE materials, resulting in a deterioration of their electrical properties, such as their ability to inhibit electrical treeing.

[0049] Therefore, developing a method and system that can non-destructively and comprehensively evaluate the degassing effect of cross-linked insulated cables is of great significance for improving cable manufacturing processes, ensuring product quality, and achieving cost reduction and efficiency improvement.

[0050] In one exemplary embodiment, such as Figure 1 As shown, a method for evaluating degassing effect is provided. Taking the application of this method to a server as an example, the method includes the following steps 102 to 108. Wherein:

[0051] Step 102: Obtain the time-domain reflection signal. The time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0052] In this context, time-domain reflectometry (TDRS) refers to the signal formed when a terahertz pulse propagates within the cable insulation layer and encounters an interface where the dielectric properties change (such as a bubble or layers of different materials), resulting in partial energy reflection. TDRS can reflect structural information within the insulation layer. For example, when a terahertz pulse is incident on a bubble within the cable insulation layer, the difference in dielectric constant between the gas inside the bubble and the surrounding cross-linked polyethylene creates a reflected pulse at that interface. This reflected pulse is received by a detector, forming a time-domain reflectometry signal.

[0053] Terahertz time-domain spectroscopy refers to a non-destructive testing technique that uses femtosecond lasers to generate and detect terahertz pulses. By scanning a sample point by point, the time-domain reflectance signal at various locations within the sample can be obtained, allowing for the analysis of its structure or properties.

[0054] For example, the cable under test can be placed on a rotating translation stage. The stage rotates the cable around an axis and moves it axially, causing the terahertz probe to move helically across the cable surface. At each predetermined position, the terahertz pulse excited by a femtosecond laser is focused by an optical system and incident on the surface of the cable insulation layer. The terahertz pulse propagates within the insulation layer and is reflected when it encounters a region with abrupt changes in dielectric constant. These reflected terahertz pulses are received and recorded by the detection system, forming a time-domain reflection signal. The obtained time-domain reflection signal contains structural information about the insulation layer, providing fundamental data for subsequent analysis.

[0055] Step 104: If bubbles are detected in the insulation layer based on the time-domain reflection signal, determine the distribution of bubbles in the insulation layer based on the time-domain reflection signal.

[0056] Bubble distribution can refer to the overall spatial location, size, and number of bubbles in the cable insulation layer.

[0057] For example, the presence of air bubbles in the insulating layer is first determined by processing and analyzing the acquired time-domain reflection signal.

[0058] As an example, due to the difference in dielectric constant between the air inside the bubble and the bulk insulating material, it will manifest as a specific reflection peak in the time-domain reflection signal. By identifying the appearance of these reflection peaks, it is possible to determine whether there are bubbles in the insulating layer.

[0059] As another example, due to the difference in dielectric constant between the gas inside the bubble and the insulating material, when a terahertz pulse encounters a bubble during propagation, it will pass through both interfaces of the bubble sequentially, thus generating two reflection peaks of similar intensity in the time-domain reflection signal. By identifying whether there are paired reflection peaks with similar amplitudes in the time-domain waveform, it can be determined whether there is a bubble in the insulating layer.

[0060] Furthermore, when bubbles are detected within the insulating layer, for any detected bubble, the occurrence times of the two reflection peaks in its time-domain reflection signal correspond to the times when the terahertz pulse arrives at the near and far surfaces of the bubble, respectively. Based on the time difference between the timestamps of these two reflection peaks and the terahertz pulse emission time, the radial depth positions of the two interfaces of the bubble within the insulating layer can be calculated. Based on the depth positions of the two interfaces, the radial dimension of the bubble, i.e., the thickness of the bubble in the terahertz pulse propagation direction, can also be deduced.

[0061] On the other hand, since there is a preset correspondence between the emission time of the terahertz pulse and the axial scanning position of the cable, the axial position of the bubble in the cable length direction can be determined by recording the emission time corresponding to the generation of the set of reflection peaks.

[0062] By combining the above information on radial depth, radial dimensions, and axial position, the spatial distribution of bubbles within the insulation layer can be constructed.

[0063] Step 106: Obtain the trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics of the insulating layer.

[0064] Here, a trap can refer to a microscopic defect within an insulating material that can trap electric charge. Trap characteristics can refer to the quantitative parameters of the trap, which may include at least one of the following: trap charge density and average trap energy level depth.

[0065] The space charge distribution characteristics can refer to the spatial distribution of space charge accumulated inside the insulating layer under conditions of applied voltage or no applied voltage, and can include at least one of the following: charge polarity, density and its variation with position.

[0066] Crosslinking byproducts can refer to low-molecular-weight byproducts remaining after the crosslinking reaction of insulating materials. The characteristics of crosslinking byproducts can refer to the quantitative parameters of crosslinking byproducts, which may include at least one of the following: content, type, and distribution of crosslinking byproducts.

[0067] For example, after determining the bubble distribution in the insulating layer, other key features of the insulating layer can be further obtained, including at least trap features, space charge distribution features, and crosslinking byproduct features.

[0068] Traps can be identified using electrical methods such as polarization-depolarization current testing. For example, after applying a DC voltage to the insulating layer and short-circuiting it, the depolarization current curve is recorded. By analyzing this curve, the trap charge density and average trap energy level depth can be obtained as trap characteristics.

[0069] Space charge distribution characteristics can be obtained through space charge measurement techniques such as the electroacoustic pulse method. For example, the location and density of charge accumulation inside an insulating layer can be detected under applied voltage to obtain space charge distribution characteristics.

[0070] The characteristics of crosslinking byproducts can be obtained through material analysis methods such as infrared spectroscopy. For example, by analyzing the absorption peak intensity of infrared light of a specific wavelength to the insulating material, the residual content of crosslinking byproducts such as cumyl alcohol can be determined as a characteristic of crosslinking byproducts.

[0071] Step 108: Evaluate the degassing effect of the cable under test based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

[0072] Among them, the degassing effect evaluation can refer to the process of comprehensively analyzing and judging the degree of degassing treatment and the insulation status of the cable under test based on the acquired multi-dimensional characteristics.

[0073] There is a clear correlation between the degassing effect of the insulation layer and its physical structure, electrical properties and chemical composition. Through the synergistic analysis of multi-dimensional features, a comprehensive judgment on the degassing effect can be achieved.

[0074] Specifically, terahertz waves are highly sensitive to density changes in insulating materials. When cross-linking byproducts escape from the insulating layer, they leave microscopic structural changes within the material. If the byproducts escape unevenly or remain excessively, they may form density anomalous regions such as bubbles or voids. When terahertz waves penetrate the insulating layer, they generate reflection signals upon encountering these regions. The spatial distribution of the bubbles can be obtained through three-dimensional scanning reconstruction. The fewer and sparser the bubble distribution, the more effectively the degassing process improves the uniformity of the insulating layer's microstructure.

[0075] At the electrical properties level, degassing directly affects the trap state and space charge behavior of insulating materials. Crosslinking byproduct molecules themselves can act as charge trapping centers, introducing numerous shallow traps into the insulating layer. The more thorough the degassing, the fewer byproducts remain, and the significantly lower the charge density of the shallow traps, thus restoring the charge transport properties of the insulating material. Simultaneously, the accumulation of space charge is closely related to byproduct residue. Residual byproducts are easily ionized or migrated under an electric field, leading to a large accumulation of space charge. Better degassing results in weaker space charge accumulation.

[0076] In areas with a high concentration of byproducts, terahertz scanning typically identifies higher bubble densities, often accompanied by higher trap charge densities and significant space charge accumulation. Infrared spectroscopy, on the other hand, detects strong characteristic absorption peaks of these byproducts. When the detection results from multiple dimensions show a consistent trend, a complete chain of evidence can be formed, enabling an accurate and reliable assessment of the overall degassing effect of the cable insulation layer.

[0077] For example, after determining the bubble distribution in the insulation layer and obtaining the trap characteristics, space charge distribution characteristics, and cross-linking by-product characteristics of the insulation layer, the degassing effect of the insulation layer can be comprehensively evaluated based on the bubble distribution, trap characteristics, space charge distribution characteristics, and cross-linking by-product characteristics according to the preset degassing effect evaluation rules or degassing effect evaluation model. Based on the comprehensive evaluation results of the degassing effect, it can be determined whether the cable under test meets the factory requirements.

[0078] In the aforementioned degassing effect evaluation method, a terahertz time-domain spectroscopy system is first used to comprehensively scan the cable under test, acquiring time-domain reflection signals that reflect the internal structure of the insulation layer. When bubbles are identified in the insulation layer based on the time-domain reflection signals, the bubble distribution within the insulation layer is further determined based on signal characteristics. By fully utilizing the sensitivity of terahertz waves to changes in dielectric structure, non-destructive detection and spatial localization of bubbles that may remain from the degassing process are achieved. Furthermore, trap features, space charge distribution features, and cross-linking byproduct features of the insulation layer are introduced to characterize the state of the insulation layer after degassing from multiple dimensions, considering both the electrical properties and chemical residues of the insulating material. Finally, by comprehensively analyzing and judging the bubble distribution and the aforementioned multi-dimensional features, an accurate evaluation of the overall degassing effect of the cable under test is achieved. Compared to traditional destructive sampling and testing methods that can only obtain local, single-indicator results, this application integrates multi-source information on physical structure, electrical properties, and chemical composition to perform a comprehensive digital scan and characterization of the insulation layer, overcoming the limitations of single-dimensional and localized evaluations and improving the comprehensiveness and accuracy of the degassing effect evaluation of cable insulation.

[0079] In one exemplary embodiment, determining the bubble distribution in the insulating layer based on the time-domain reflection signal includes:

[0080] From the time-domain reflection signal, the propagation time interval and reflection intensity of the terahertz pulse in the insulating medium are identified; based on the propagation time interval and reflection intensity, the bubble surface position of at least one bubble in the insulating layer is located; based on the bubble surface position of each bubble, the bubble distribution in the insulating layer is generated.

[0081] The propagation time interval refers to the time difference between different reflection peaks in the time-domain reflected signal, reflecting the time required for the terahertz pulse to propagate between different interfaces. The reflection intensity refers to the energy amplitude of the terahertz pulse reflected back at an interface with a sudden change in dielectric constant, which can be expressed as voltage amplitude or relative intensity, reflecting the magnitude of the difference in dielectric properties at the interface.

[0082] For example, such as Figure 2 As shown, the dashed line 201 represents the propagation path of the terahertz pulse. When the terahertz pulse is incident on the surface 202 of the insulating layer 206, a strong reflection peak P0 is generated. If there is a bubble 205 in the insulating layer 206, two secondary reflection peaks P1 and P2 will appear after P0, corresponding to the upper interface 203 and lower interface 204 of the bubble 205, respectively. Assuming the time difference between P0 and P1 is Δt1, it corresponds to the time it takes for the terahertz pulse to propagate from the surface 202 of the insulating layer 206 to the upper interface 203 of the bubble 205 and back to the surface 202 of the insulating layer 206. Assuming the time difference between P1 and P2 is Δt... 12 This corresponds to the duration of the terahertz pulse propagating from its upper interface 203 to its lower interface 204 and back to its upper interface 203 inside the bubble 205.

[0083] For any terahertz pulse, if multiple reflection interfaces exist along its propagation path, the propagation time interval corresponding to each reflection interface can be identified from the time-domain reflection signal. When the same terahertz pulse is reflected back to the detection system at different reflection interfaces, the detection system can detect different reflection intensities. That is, if N reflection interfaces exist along the propagation path of a terahertz pulse, N propagation time intervals and N reflection intensities can be identified from its corresponding time-domain reflection signal.

[0084] The location of the bubble surface can refer to the spatial position of the interface between the bubble and the surrounding insulating material.

[0085] For example, the acquired time-domain reflection signal is first analyzed using waveform analysis. Figure 3 As shown, in the time-domain reflection signal, the surface of the insulating layer (air-insulating layer interface) can generate a primary reflection pulse P0, which arrives earliest. If there is a bubble in the insulating layer, two secondary reflection pulses P1 and P2 will appear after P0. P1 is the reflection pulse generated by the upper interface of the bubble (insulating layer-air interface), and P2 is the reflection pulse generated by the lower interface of the bubble (air-insulating layer interface).

[0086] Waveform analysis algorithms can automatically detect all reflection peaks in a time-domain reflected signal. For each detected reflection peak, the absolute time of its occurrence relative to the pulse emission time, and the relative time difference between adjacent reflection peaks are recorded. The time difference between P0 and P1 can be denoted as Δt1, and the time difference between P1 and P2 as Δt. 12 Simultaneously, record the amplitude of each reflection peak. Since P1 and P2 both originate from the interface between air and insulating material, their reflection intensities should be approximately equal. The aforementioned propagation time intervals (Δt1, Δt...) 12 The reflection intensity (the amplitudes of P1 and P2) is the basic data for subsequent positioning of the bubble surface.

[0087] Furthermore, the spatial location of the bubble is achieved using the identified propagation time interval and reflection intensity. First, based on the time difference Δt1 between P0 and P1, and combined with the known propagation speed of the terahertz pulse in the insulating material, the radial depth of the bubble's upper surface within the insulating layer can be calculated using the distance calculation formula. Next, based on the time difference Δt1 between P1 and P2... 12 By combining the propagation speed of the terahertz pulse in air, the radial thickness of the bubble can be calculated. Furthermore, based on the amplitudes of P1 and P2 and the temporal proximity of the two reflection peaks, it can be further confirmed that P1 and P2 belong to two interfaces of the same bubble.

[0088] As an example, the burial depth d of the bubble can be calculated by measuring the time difference Δt1 between the primary reflected pulse corresponding to the surface of the insulating layer and the secondary reflected pulse corresponding to the interface on the bubble:

[0089] d=(c×Δt1) / (2×n)

[0090] Where d is the distance between the upper interface of the bubble and the upper surface of the insulation layer; c is the speed of light in a vacuum; Δt1 is the time difference between P0 and P1; and n is the refractive index of the insulation layer in the terahertz frequency band, which can be obtained by detecting the degassed cable sample without bubbles.

[0091] Similarly, the time difference Δt between P1 and P2 can be used as a basis. 12 The radial thickness of the bubble is calculated.

[0092] Meanwhile, since there is a preset correspondence between the emission time of the terahertz pulse and the axial scanning position of the cable, the axial position of the bubble along the cable length can be determined by recording the scanning positions corresponding to the generation of P1 and P2. Combining the radial depth position, radial thickness, and axial position, the bubble surface positions of the upper and lower surfaces can be located.

[0093] Furthermore, by collecting information such as the upper surface position, lower surface position, radial thickness, and axial position of all detected bubbles, and arranging and reconstructing them according to their spatial coordinates, the bubble distribution in the insulation layer can be generated.

[0094] The bubble distribution can be presented in several ways: by recording parameters such as the number of each bubble, the depth of the upper surface, the depth of the lower surface, the radial thickness, and the axial position in the form of a data table; or by displaying the bubble distribution cross section along the length of the cable in the form of a two-dimensional cross-sectional view; or by reconstructing the spatial distribution image of the bubbles in the insulation layer in the form of a three-dimensional visualization.

[0095] In some embodiments, the bubble density ρ can be further calculated based on the bubble distribution:

[0096] ρ=N / (π×D×L)

[0097] Where ρ is the bubble density (unit: bubbles / m²); N is the total number of bubbles identified in the insulation layer or the insulation material segment scanned by terahertz time-domain spectroscopy; D is the diameter of the insulation layer or the insulation material segment scanned by terahertz time-domain spectroscopy (unit: m); and L is the axial length of the insulation layer or the insulation material segment scanned by terahertz time-domain spectroscopy (unit: m). The bubble density is calculated by unfolding the cylindrical cable surface into a plane, reflecting the number of bubbles per unit insulation surface area in the insulation material segment.

[0098] In some embodiments, the bubble density can be in the form of a bubble density axial uniformity curve. Figure 4 This is a graph illustrating the axial uniformity of bubble density in the insulation layer in one embodiment. The graph uses the axial length of the cable insulation layer as the abscissa (0-100 meters) and the local bubble density as the ordinate (bubbles / meter). The undulating curve quantitatively displays the distribution trend of bubbles within the insulation layer along the cable length. The curve shows a distinct peak in the 40-70 meter range, indicating that the defect density in this area is significantly higher than in other sections, suggesting axial non-uniformity in the degassing effect of the insulation layer.

[0099] In this embodiment, by identifying characteristic reflection peaks such as P0, P1, and P2 and their time intervals from the time-domain reflection signal, the precise spatial positioning of bubbles in the insulation layer is achieved. The discrete positioning results are integrated into a dataset that reflects the overall distribution characteristics, providing accurate spatial distribution data support for subsequent evaluation of the impact of bubbles on insulation performance and optimization of the degassing process. This significantly improves the detection capability of degassing uniformity and microstructural defects.

[0100] In one exemplary embodiment, obtaining the trap features of the insulating layer includes:

[0101] For any preset first test site on the insulating layer, depolarization current data is acquired. The depolarization current data is obtained by performing polarization-depolarization current tests on the insulating layer from the preset first test site. The depolarization current component is extracted from the depolarization current data. By performing staged fitting on the depolarization current component, the trap charge density and trap energy level depth of the insulating layer at the preset first test site are determined. The trap charge density and trap energy level depth are determined as the trap characteristics of the insulating layer.

[0102] The preset first test site can refer to a specific location on the insulation layer that is pre-selected for polarization-depolarization current testing. In some embodiments, multiple preset first test sites can be selected along the axial direction at preset intervals. For example, assuming the cable length is L, the two ends of the cable insulation layer and the point at L / 2 can be determined as preset first test sites.

[0103] Depolarization current data refers to the sequence of current values ​​that decay over time after the polarization process is completed and the electrodes are short-circuited. It reflects the relaxation process of the polarized medium after the external electric field is removed.

[0104] Polarization-depolarization current testing is a method for testing the properties of dielectric materials. It involves applying a DC voltage (polarization) to a sample and then short-circuiting it (depolarization), recording the current response throughout the process, and using this data to analyze the dielectric response characteristics and trap states of the material.

[0105] The de-trapping current component can refer to the current component separated from the total depolarization current, contributed by the thermal release of the trapped charges after the electric field is removed.

[0106] Phased fitting can refer to the process of dividing the current decay curve into several intervals according to the time axis, and fitting each interval with an exponential function to obtain the fitting parameters corresponding to different relaxation times.

[0107] Trapping charge density can refer to the total amount of charge trapped by traps in a unit volume of insulating material.

[0108] The depth of the trap energy level can refer to the binding strength of the trap on the charge, that is, the energy barrier that the charge needs to overcome to escape from the trap.

[0109] For example, at each preset first test point on the insulating layer, a polarization-depolarization current test electrode is connected to ensure good contact between the electrode and the surface of the insulating layer. Then, a polarization-depolarization current test is performed according to preset test parameters. Specifically, a DC voltage is first applied to the insulating layer and maintained for a preset polarization time, causing polarization and charge injection within the insulating material, filling the traps with charge. After polarization, the test electrode is immediately short-circuited, and the depolarization current is recorded simultaneously. The recording process lasts for a preset time, resulting in a current decay curve with time on the horizontal axis and current value on the vertical axis. This current decay curve can be used as the depolarization current data for that test point.

[0110] Next, the depolarization current data undergoes signal separation processing. The measured depolarization current is a superposition of contributions from multiple physical processes, including the trap release current contributed by the trap release charge. To obtain information relevant only to the trap state, the trap current component needs to be extracted from the total current.

[0111] In some embodiments, a staged fitting method can be used to extract the depolarization current component from the total current. Specifically, the total depolarization current is considered as the superposition of multiple exponential decay processes, expressed as I(t) = ΣAm·exp(-t / τm), where m represents different relaxation processes. A set of fitting coefficients Am and relaxation time constant τm are solved using a numerical fitting algorithm, ensuring that the curve of the above expression best matches the measured depolarization current curve. After fitting, based on the physical meaning of the relaxation time constant τm, the component corresponding to the slow decay process of trap charge release is separated, and the sum of these components is the depolarization current component.

[0112] After extracting the detrapping current component, further parameter inversion is performed on it. The detrapping current component is divided into several stages on a logarithmic time coordinate, each stage corresponding to the process of releasing charge from the trap within a specific energy level range. An exponential function is used to fit each stage separately, obtaining each set of fitting parameters, including the fitting coefficients Am and the relaxation time constant τm.

[0113] According to dielectric physics theory, the relaxation time constant τm and the trap energy level depth Et satisfy the Arrhenius relation: τm = τ0 · exp(Et / (k · T)). Here, k is the Boltzmann constant, T is the absolute temperature, and τ0 is the reciprocal of the escape attempt frequency. Substituting the fitted τm into the above relation, the average trap energy level depth for the corresponding stage can be deduced. By taking a weighted average or a representative value of the trap energy level depths for each stage, the trap energy level depth at the test site can be obtained.

[0114] Furthermore, the trap charge density can be obtained by integrating the trap current component over time.

[0115] The trap charge density and trap energy level depth calculated above are used as the trap characteristics of the preset first test site for subsequent comprehensive evaluation of degassing effect.

[0116] In this embodiment, depolarization current data is obtained through polarization-depolarization current testing, and the detrapping current component is extracted from it. Then, trap charge density and trap energy level depth are determined through staged fitting, achieving non-destructive quantitative characterization of the trap state inside the insulating material. Compared to traditional methods that can only indirectly reflect insulation performance through macroscopic indicators such as breakdown field strength, this scheme directly quantifies the trap state introduced by residual crosslinking byproducts at the microscopic level, providing characteristic parameters reflecting the microscopic root causes of electrical properties for degassing effect evaluation. By using trap characteristics as one of the dimensions for degassing effect evaluation, it can be subsequently combined with other dimensions such as bubble distribution, space charge distribution, and crosslinking byproduct characteristics to form multi-source data fusion, significantly improving the comprehensiveness and accuracy of degassing effect evaluation.

[0117] In one exemplary embodiment, obtaining space charge distribution characteristics includes:

[0118] The space charge distribution of the insulating layer at a preset second test site is obtained by performing an electroacoustic pulse test on the insulating layer from the preset second test site. Based on the space charge distribution, the level of opposite polarity charge accumulation and the internal electric field distortion rate of the insulating layer at the preset second test site are determined. The level of opposite polarity charge accumulation and the internal electric field distortion rate are determined as the space charge distribution characteristics of the insulating layer.

[0119] The preset second test site can refer to a specific location on the cable insulation layer pre-selected for performing electroacoustic pulse testing. In some embodiments, multiple preset second test sites can be selected along the axial direction at preset intervals. For example, assuming the cable length is L, the two ends of the cable insulation layer and the point at L / 2 can be determined as preset second test sites. The preset second test sites can be the same as or different from the preset first test sites, and this embodiment does not impose any restrictions.

[0120] Space charge distribution refers to the radial distribution of the space charge density accumulated inside the insulating layer under conditions of applied voltage or no applied voltage. It can be represented as a curve with charge density as the vertical axis and depth as the horizontal axis.

[0121] Electroacoustic pulse testing is a space charge measurement technique that uses a high-voltage pulse applied to an insulation layer to generate sound waves from the space charge. By detecting these sound waves, the distribution of space charge within the insulation layer can be deduced. For example, when a high-voltage pulse is applied to the surface of a cable insulation layer, the pulse voltage acts on the accumulated space charge, causing it to produce weak mechanical vibrations and propagate sound waves outward. A sensor located on the other side of the insulation layer receives this sound wave signal, and a signal processing algorithm converts the sound wave signal into a curve showing the distribution of charge density with depth. This testing process is called electroacoustic pulse testing.

[0122] The level of opposite polarity charge accumulation can refer to the degree to which space charge opposite to the polarity of the electrode accumulates in the region near the electrode, and can be quantified by the peak value or the total integral of the space charge density in that region.

[0123] The internal electric field distortion rate can refer to the degree to which the actual electric field distribution inside the insulation layer deviates from the design electric field when there is no space charge due to the presence of space charge. It can be expressed as the ratio or percentage of the maximum distorted electric field to the design electric field.

[0124] For example, at each predetermined second test point on the insulating layer, a test system is arranged according to the requirements of the electroacoustic pulse test method. Typically, a high-voltage electrode is applied to one side of the insulating layer, and a piezoelectric sensor is placed on the other side. During testing, a high-voltage pulse is applied to the insulating layer. This pulse acts on the space charge accumulated inside the insulating layer, causing an instantaneous change in electrostatic force, thereby exciting an acoustic signal. The acoustic signal propagates within the insulating layer and is received by the piezoelectric sensor located on the other side, converting it into an electrical signal. After amplification and digitization, the acquired electrical signal is converted into a distribution curve of space charge density as a function of depth using an inversion algorithm (e.g., deconvolution). This curve, with the radial depth of the insulating layer as the abscissa and charge density as the ordinate, can be used as the space charge distribution at the predetermined second test point.

[0125] Next, the charge accumulation near the electrodes in the space charge distribution is identified. Specifically, the polarity of the charge accumulated in the region near the electrodes is determined based on the polarity of the voltage applied during the test. If negative charge accumulation occurs near a positive electrode, or positive charge accumulation occurs near a negative electrode, it is determined that opposite polarity charge accumulation has occurred. The peak value of the space charge density in this region, or the total charge obtained by integrating the charge density curve in the region near the electrodes, can be used as the quantitative value of the level of opposite polarity charge accumulation.

[0126] Simultaneously, the internal electric field distortion rate can be determined. Specifically, based on the obtained space charge distribution, the electric field distribution inside the insulation layer is recalculated using the Poisson equation. Starting from one electrode, the space charge distribution is integrated to progressively calculate the electric field intensity at each depth, obtaining the actual electric field distribution curve. This actual electric field distribution is compared with the designed electric field distribution without space charge to find the maximum value of the actual electric field and the corresponding designed electric field value. The ratio of the maximum value of the actual electric field to the corresponding designed electric field value is determined as the electric field distortion rate.

[0127] The calculated level of heteropolar charge accumulation and internal electric field distortion rate are used as the spatial charge distribution characteristics of the preset second test site for subsequent comprehensive evaluation of degassing effect.

[0128] In this embodiment, the space charge distribution is obtained through electroacoustic pulse testing, and based on this, the level of heteropolar charge accumulation and the internal electric field distortion rate are determined, achieving a quantitative characterization of the ability of degassing treatment to suppress space charge. Residual cross-linking byproducts are a significant cause of space charge accumulation; the more thorough the degassing, the fewer byproducts remain, the weaker the space charge accumulation, and the smaller the resulting electric field distortion. This method directly quantifies this effect from the perspective of space charge behavior, providing a key parameter reflecting the stability of insulating materials under an electric field for evaluating the degassing effect. By using the space charge distribution characteristics as one dimension of the degassing effect evaluation, it can be subsequently combined with other dimensions such as bubble distribution, trap characteristics, and cross-linking byproduct characteristics to form multi-source data fusion, jointly verifying the sufficiency and uniformity of the degassing treatment from different perspectives, significantly improving the comprehensiveness and reliability of the evaluation results.

[0129] In one exemplary embodiment, obtaining crosslinking byproduct characteristics includes:

[0130] Based on the bubble distribution, the test area is located on the insulation layer. The bubble distribution density in the test area is less than the preset density. By performing infrared spectral analysis on the test area, the intensity of the characteristic absorption peak of the preset cross-linking byproduct is detected, and the intensity of the characteristic absorption peak is determined as the characteristic of the cross-linking byproduct of the insulation layer.

[0131] The test area can refer to a specific location on the cable insulation layer selected for infrared spectral analysis based on bubble distribution information.

[0132] Bubble distribution density can refer to the number of bubbles or the volume ratio of bubbles in an insulation layer per unit volume or unit area, and is used to quantify the density of bubble distribution.

[0133] The preset density threshold can refer to a pre-set boundary value used to judge the density of bubble distribution. Areas exceeding the threshold can be considered as dense bubble areas, while areas below the threshold can be considered as sparse bubble areas.

[0134] Infrared spectroscopy is a detection technique that utilizes the selective absorption of infrared light by substances to analyze their chemical composition. Different chemical bonds or functional groups exhibit characteristic absorption of infrared light at specific wavelengths. The probe of an infrared spectrometer is attached to the surface of an insulating layer, emitting infrared light into the material and receiving the reflected spectrum. By analyzing the position and intensity of absorption peaks in the spectrum, the presence and relative abundance of specific chemical components in the material can be determined.

[0135] Pre-defined cross-linking byproducts can refer to specific low-molecular-weight compounds generated during the cross-linking reaction and that need to be removed during degassing. These can be predetermined based on the type of insulating material.

[0136] The intensity of a characteristic absorption peak refers to the height or area of ​​the absorption peak corresponding to a specific functional group or chemical bond in an infrared spectrum, reflecting the relative abundance of that functional group or chemical bond.

[0137] For example, firstly, based on bubble distribution information, the bubble distribution density of each local region in the insulation layer is calculated. Bubble distribution density can be quantified in various ways, such as the number of bubbles per unit length, the number of bubbles per unit volume, or the volume fraction occupied by bubbles. According to a pre-set density threshold, at least one continuous region in the insulation layer with a bubble distribution density lower than the preset threshold is identified, and the specific location range of these regions on the cable, including the starting and ending axial coordinates, is recorded to determine the test area for infrared spectral analysis. Since the presence of bubbles alters the propagation path of infrared light and affects the accurate measurement of absorption peaks, locating the test area in a bubble-sparse region can effectively reduce the interference of the bubbles themselves on infrared spectral measurements.

[0138] Next, an infrared spectrometer equipped with an ATR (Attenuated Total Reflectance) accessory was used, with the ATR probe tightly attached to the surface of the insulating layer in the test area. ATR technology utilizes the principle of total internal reflection, causing infrared light to form an evanescent wave at the contact surface between the probe and the sample, penetrating to the shallow surface of the sample, carrying information about the sample's chemical composition, and then returning to the detector.

[0139] The infrared spectrometer is activated, and infrared light is emitted onto the surface of the insulating layer through the ATR probe. The reflected spectral signal is then received to obtain the infrared absorption spectrum of the test area. Characteristic absorption peaks of the pre-defined cross-linking byproducts are identified in the acquired spectrum. The positions of these characteristic absorption peaks are automatically or manually marked using spectral analysis software, and their peak heights are measured or their peak areas are calculated to obtain the intensity of the characteristic absorption peaks.

[0140] If multiple measurements are performed on the same test area or the area contains multiple test points, the multiple intensity values ​​detected within the same test area can be averaged, and the resulting average value can be used as the characteristic absorption peak intensity of that test area. This characteristic absorption peak intensity of the test area is then used as a characteristic of the crosslinking byproducts in that test area for subsequent comprehensive evaluation of the degassing effect.

[0141] In this embodiment, infrared spectroscopy analysis of sparsely distributed bubble test areas enables non-destructive detection and targeted verification of crosslinking byproduct residues. Compared to traditional destructive gas chromatography methods, this approach utilizes a portable infrared spectrometer with an ATR probe, allowing direct chemical composition analysis of the insulation surface without cutting cable samples, thus preserving product integrity. More importantly, by acquiring bubble distribution information before locating the test area, infrared spectroscopy analysis can focus on sparse bubble regions, effectively avoiding bubble interference with spectral measurements. Simultaneously, it establishes a correspondence between microscopic physical structure (bubbles) and chemical composition (byproduct residues). The measured characteristic absorption peak intensity serves as a feature of crosslinking byproducts, providing direct chemical evidence for degassing effect evaluation. This data can then be fused with other dimensions such as bubble distribution, trap characteristics, and space charge distribution characteristics to form multi-source data, jointly verifying the adequacy of the degassing treatment from different perspectives and significantly improving the comprehensiveness and reliability of the evaluation results.

[0142] In one exemplary embodiment, the degassing effect of the cable under test is evaluated based on bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics, including:

[0143] The bubble distribution, trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics are input into a trained comprehensive evaluation model for degassing effect. The degassing effect of the cable under test is evaluated to obtain the degassing uniformity index and comprehensive degassing grade.

[0144] Among them, the comprehensive evaluation model of degassing effect can refer to a mathematical model that is trained by machine learning algorithm and can automatically output the evaluation result of degassing effect based on the multi-dimensional features of the input.

[0145] The degassing uniformity index is a numerical indicator that quantifies the degree of uniformity in degassing treatment along the length or radial direction of cable insulation. The degassing uniformity index is obtained by a comprehensive evaluation model of degassing effect, which integrates bubble distribution, trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics.

[0146] The comprehensive degassing level can refer to the result of classifying and evaluating the overall degassing effect of the cable based on the degassing uniformity index and the preset degassing level threshold.

[0147] For example, the four types of characteristic data of the cable under test—bubble distribution, trap features, space charge distribution features, and cross-linking byproduct features—are processed and normalized according to the format required by the model. The processed characteristic data is then input into the pre-trained comprehensive evaluation model for degassing effect. Based on the mapping relationships learned during the training phase, the comprehensive evaluation model calculates the input multidimensional features and outputs the degassing uniformity index and the overall degassing level.

[0148] In some embodiments, the comprehensive evaluation model for degassing performance can be a generative large model. After determining the degassing uniformity index and the overall degassing level, the comprehensive evaluation model for degassing performance can further generate a visual evaluation report and degassing optimization suggestions.

[0149] The above-mentioned comprehensive evaluation model for degassing effect can be iteratively optimized before practical application. For example... Figure 5 As shown, its iterative optimization process includes:

[0150] Step 502: Obtain the training time-domain reflection signal. The training time-domain reflection signal is obtained by performing a terahertz time-domain spectral scan on the training cable. The training cable includes a training insulation layer, and there are air bubbles in the training insulation layer.

[0151] Step 504: Determine the distribution of training bubbles in the training insulating layer based on the training time-domain reflection signal;

[0152] Step 506: Obtain the training trap features, training space charge distribution features, and training crosslinking byproduct features of the training insulating layer;

[0153] Step 508: By inputting the training bubble distribution, training trap features, training space charge distribution features, and training crosslinking byproduct features into the machine learning model to be trained, the degassing effect of the training cable is evaluated, and the training degassing uniformity index and training comprehensive degassing level are obtained.

[0154] Step 510: Obtain the degassing uniformity index label and the comprehensive degassing level label of the training insulation layer. The degassing uniformity index label and the comprehensive degassing level label are obtained by performing destructive testing on the training cable.

[0155] Step 512: Based on the difference between the training degassing uniformity index and the degassing uniformity index label, and the difference between the training comprehensive degassing level and the comprehensive degassing level label, iteratively optimize the machine learning model to be trained to obtain the trained comprehensive evaluation model of degassing effect.

[0156] Among them, training cables can refer to cross-linked insulated cable samples used to collect training data for training machine learning models.

[0157] The training insulation layer can refer to the insulation layer in the training cable, which is the object of various tests and feature extractions.

[0158] The training time-domain reflectance signal can refer to the time-domain reflectance signal obtained by performing a terahertz time-domain spectral scan on the training cable.

[0159] Training bubble distribution can refer to the spatial distribution information of bubbles in the training insulating layer, determined based on the training time-domain reflection signal.

[0160] Training trap features can refer to the trap features obtained by performing polarization-depolarization current tests on the training insulation layer.

[0161] The training space charge distribution characteristics can refer to the space charge distribution characteristics determined by electroacoustic pulse testing of the training insulation layer.

[0162] The characteristics of cross-linking byproducts in training can refer to the characteristics of cross-linking byproducts obtained by infrared spectroscopy analysis of the training insulation layer.

[0163] The training degassing uniformity index refers to the predicted value of the degassing uniformity index output by the machine learning model to be trained after the training features are input into the model during the model training process.

[0164] The training comprehensive degassing level can refer to the comprehensive degassing level prediction value output by the machine learning model to be trained after the training features are input into the machine learning model to be trained during the model training process.

[0165] The degassing uniformity index label can refer to the degassing uniformity index obtained through destructive testing methods and used as the true value of the training data.

[0166] In some embodiments, the degassing uniformity index label P can be represented as:

[0167] P = Q_{av} / Q_{max}

[0168] Where Q_{av} is the average value of the normalized light intensity coefficient, and Q_{max} is the maximum value of the normalized light intensity coefficient.

[0169] The two key parameters in the above formula can be obtained and processed through experiments. The specific steps are as follows:

[0170] Step 1: Sample preparation and processing: Take the insulated core of the cable, retain the insulation layer and inner shielding layer, grind the outer surface, and then treat it in heated silicone oil until the insulation layer becomes transparent.

[0171] Step 2: Acquire light scattering images: In a dark room, illuminate the transparent sample horizontally with a monochromatic laser and use a CCD (charge coupled device) camera to capture images of the scattered light spots within the insulating layer from a vertical direction.

[0172] Step 3: Extraction and normalization of light intensity: Extract light intensity data from the image along the radial direction of the insulating layer, and normalize it based on the light intensity near the outer surface of the insulating layer to obtain the normalized light intensity coefficient Q.

[0173] Step 4: Determine the calculation parameters: Within the preset insulation layer thickness range, find the maximum value of Q, Q_{max}, and calculate its average value, Q_{av}.

[0174] The comprehensive degassing level label can refer to the comprehensive degassing level obtained through destructive testing methods, which serves as the true value of the training data.

[0175] Destructive testing refers to testing methods that require causing permanent damage to the cable by cutting samples or insulation layers. Compared to destructive testing methods used in actual degassing effect evaluation, destructive testing methods used in the model training phase can perform more intensive sampling to improve the accuracy of the test results. This, in turn, improves the accuracy of the degassing uniformity index label and the comprehensive degassing level label used for training, thereby improving the model training effect.

[0176] For example, several degassed cables are selected as training cables. For each training cable, a terahertz time-domain spectroscopy scanning method, similar to that used for the cable under test, is employed to scan along its length, acquiring the training time-domain reflectance signal at each scanning position. Then, using the same bubble distribution determination method as for the cable under test, reflection peaks are identified from the training time-domain reflectance signal, and the bubble surface positions are located, thereby generating the training bubble distribution in the training insulation layer. Following the same method as for the cable under test, its training trap characteristics, training space charge distribution characteristics, and training crosslinking byproduct characteristics are acquired.

[0177] Subsequently, for each training cable, the four types of training feature data acquired—training bubble distribution, training trap features, training space charge distribution features, and training crosslinking byproduct features—were processed and normalized according to the format required by the model. The processed training feature data was then fed into the machine learning model to be trained. The machine learning model, under its current parameter state, performed forward calculations on the input features and output the training degassing uniformity index and the training comprehensive degassing level.

[0178] After non-destructive testing of the training cable, destructive testing can be performed to obtain its true degassing uniformity index label and comprehensive degassing grade label.

[0179] Subsequently, for the continuous output of the degassing uniformity index, the mean squared error can be used as the loss function; for the categorical output of the comprehensive degassing level, the cross-entropy loss function can be used. The predicted values ​​of the machine learning model to be trained and the obtained label values ​​are substituted into the corresponding loss functions to calculate the current loss value of the machine learning model. Based on the loss value, the gradients of each model parameter of the machine learning model to be trained are calculated using the backpropagation algorithm, and the model parameters are updated according to optimization algorithms such as gradient descent. The above process is repeated for each training cable, or batch training is used for multiple iterations until the loss converges or the preset number of training rounds is reached. The machine learning model to be trained after iterative optimization can be used as a trained comprehensive evaluation model for degassing effect, for rapid and accurate evaluation of the degassing effect of the tested cable.

[0180] In this embodiment, a comprehensive evaluation model for degassing effect is constructed and trained, achieving deep fusion and automatic judgment of multi-source heterogeneous data such as bubble distribution, trap characteristics, space charge distribution characteristics, and cross-linking by-product characteristics. Compared with traditional methods that rely on manual experience or simple weighting for comprehensive evaluation, this scheme uses machine learning algorithms to automatically learn the complex nonlinear mapping relationship between each dimension of features and degassing effect, avoiding evaluation bias caused by subjective factors and linear assumptions. By introducing destructive detection results as training labels, the model can continuously improve the accuracy of evaluation by using real and accurate degassing effect as the learning target. The degassing uniformity index and comprehensive degassing level output by the model not only provide a judgment on whether the overall performance is qualified, but also provide a quantitative description of degassing uniformity and a graded evaluation of the overall effect, providing richer decision-making basis for the refined control of the degassing process and the graded management of product quality.

[0181] In one exemplary embodiment, a 220kV XLPE insulated submarine cable that has just undergone degassing treatment is used as the evaluation object to specifically illustrate the degassing effect evaluation method.

[0182] The cable under test is a 100-meter-long 220kV cross-linked polyethylene insulated submarine cable with a conductor cross-section of 800mm². 2 The insulation layer thickness is 22mm. This cable undergoes a staged heating and degassing process, with a degassing temperature of 70℃ and a degassing time of 14 days, in a nitrogen-circulating environment. This evaluation aims to non-destructively assess the degassing effect and uniformity of the cable before it leaves the factory. The specific process is as follows:

[0183] Step 1: Terahertz time-domain spectral scanning and bubble distribution determination.

[0184] A cable approximately 100 meters long was placed on a specially designed rotating and translating support. The terahertz time-domain spectroscopy scanning module was activated, controlling the cable to rotate uniformly around its axis while simultaneously moving axially in a step-like motion to perform a 360-degree transmission-type scan of the cable insulation layer. Time-domain reflection signals were collected at each scanning position, and the signals were analyzed using data processing software to identify paired reflection peaks generated when the terahertz pulse encounters bubble interfaces within the insulation layer. The surface positions of each bubble were located based on the timestamps of the reflection peaks, thus generating a three-dimensional distribution cloud map of the microscopic bubbles within the insulation layer. Based on the bubble distribution information, a total of 127 bubbles with an equivalent diameter greater than 50 μm were identified in the entire cable reel, with an overall bubble density of 3.05 bubbles / dm³. The bubbles exhibited a non-uniform distribution within the insulation layer, with a significantly higher bubble density between 40 and 70 meters axially than at both ends, and a greater concentration in the central region of the insulation layer in terms of radial depth. The coefficient of variation for the bubble density along the axial direction was calculated to be 28.5%, indicating that axial uniformity needs improvement.

[0185] Step 2: Electrical characteristics test.

[0186] Polarization-depolarization current test and space charge distribution test were performed at three locations on the cable: the beginning (0 meters), the middle (50 meters), and the end (100 meters).

[0187] For polarization-depolarization current testing, a small section of the outer sheath and metal shield was stripped from each test site to expose the insulating layer surface, and test electrodes were connected. A 500V DC voltage was applied, and after polarization for 1000 seconds, a short-circuit discharge was performed, recording the depolarization current data. The trapping current component was extracted from the depolarization current data, and the trapping current component was fitted in stages to determine the trap charge density and average trap level depth at each test site. The test results showed that the trapping charge density at the initial stage was 1.32 × 10⁻⁶. 3 C / m³, trap energy level depth is 0.86 eV; middle trap charge density is 1.65 × 10⁻⁶ eV. 3 C / m³, trap energy level depth is 0.89 eV; terminal trap charge density is 1.28 × 10⁻⁶. 3 C / m 3 The trap energy level depth is 0.84 eV. The trap charge density in the central region is slightly higher than that at both ends, corresponding to the location of the dense bubble region found by terahertz scanning.

[0188] For space charge distribution testing, the electroacoustic pulse method was used for measurement at each test site. A DC electric field of 50 kV / mm was applied and maintained for 30 minutes, and the charge density distribution curve along the thickness direction of the insulating layer was measured. Based on the space charge distribution, the maximum electric field distortion rate and the average opposite-polarity charge density at each test site were determined. The test results showed that the maximum electric field distortion rate at the starting end was 8%, and the average opposite-polarity charge density was 1.05 × 10⁻⁶.2 C / m³; the maximum electric field distortion rate in the central part is 12%, and the average opposite-polarity charge density is 1.48 × 10⁻⁶. 2 C / m 3 The maximum electric field distortion rate at the end is 7%, and the average opposite-polarity charge density is 0.98 × 10⁻⁶. 2 C / m³. The electric field distortion rate at all test sites was below the safety threshold of 15%, but the distortion rate and charge density in the central region were still slightly higher than those at the ends, consistent with the PDC test and terahertz scan results.

[0189] Step 3: Infrared spectroscopy verification.

[0190] Based on the bubble distribution generated by terahertz scanning, test areas were located on the insulation layer. Regions with high bubble density (75 meters axially) and regions with low bubble density (20 meters axially) were selected as test areas. Infrared spectral analysis was performed on the outer, middle, and inner layers of the insulation in each test area. An infrared spectrometer equipped with a micro-ATR probe was used to acquire the infrared spectrum of each micro-region, and the intensity of the characteristic absorption peaks of the preset crosslinking byproducts was detected. The test results showed that cumyl alcohol was detected at 3400 cm⁻¹ at all test points. -1 The characteristic absorption peak of hydroxyl groups and acetophenone at 1690 cm⁻¹ are nearby. -1 The carbonyl characteristic absorption peaks were observed nearby, but the signal intensities differed significantly. The intensity of the cumulol characteristic peak in the middle layer of the high bubble density region was approximately 2.3 times that of the middle layer of the low bubble density region, indicating a significant radial gradient in the byproduct residue. Semi-quantitative analysis showed that the cumulol residue at all test sites was below the sufficient threshold of 2 mg / g, but the uneven distribution was quite evident.

[0191] Step 4: Multi-source information fusion and comprehensive evaluation of degassing effect.

[0192] The multi-dimensional feature data obtained in the above steps are input into the pre-trained comprehensive evaluation model for degassing effect. After normalizing the input features, the comprehensive evaluation model performs a comprehensive calculation based on the mapping relationship learned during the training phase, and outputs the degassing uniformity index and the comprehensive degassing level.

[0193] Model evaluation showed that the degassing uniformity index of this cable reel was 0.78 (out of 1.0). This index comprehensively reflects the spatial consistency of bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics in both the axial and radial directions, indicating good overall uniformity but still room for improvement. The overall degassing level is Level 2 (sufficient). Based on the degassing uniformity index and the residual threshold of key byproducts, the space charge distortion rate at all test points has been reduced to a low level. Infrared spectroscopy shows that the characteristic peak of cumyl alcohol in the inner layer is relatively weak, and the electrical performance meets the factory requirements, making it safe for shipment. The model analysis also indicates that the degassing efficiency in the middle region of the cable is slightly lower than that at both ends, consistent with the trend of the test results across all dimensions.

[0194] A comprehensive evaluation report can be generated based on the evaluation results. The comprehensive evaluation report may include the following:

[0195] In general, the degassing effect of the 220kV XLPE cable in this panel is sufficient, with an overall degassing level of "sufficient," and the key electrical performance meets the factory requirements. However, the evaluation also revealed a clear uneven axial degassing phenomenon in the cable, with the middle region (approximately 40 to 70 meters) showing inferior performance compared to the two ends in terms of microbubble density, trapped charge level, and residual by-product gradient.

[0196] Based on the process data, the root cause of the non-uniformity may stem from two factors: First, the cables are stacked in a disc shape in the degassing chamber, and the heat dissipation and gas diffusion path in the middle area is longer, resulting in a lower actual degassing efficiency than the surface layer; second, the diffusion path of by-products in the middle of the XLPE insulation layer is the longest, which is theoretically the most difficult part to degas completely.

[0197] For this cable reel, the long-term operating status of the central section can be marked and monitored, but rework is not required. Regarding the production process, it is recommended to optimize the degassing chamber design, enhancing the forced convection efficiency of the internal airflow circulation, especially directional airflow to the central area of ​​the cable reel, to reduce temperature and concentration gradients. For cables of the same specification, consider increasing the temperature by 3°C to 5°C at the current degassing duration, or adjusting the temperature profile in the later stages to accelerate the diffusion of byproducts in the central section without affecting the overall crystallinity. Simultaneously, the degassing uniformity index evaluated in this study can be correlated with process parameters to establish a benchmark database of "process parameters - evaluation index" for subsequent cables of the same type.

[0198] This embodiment, through the aforementioned multi-dimensional and non-destructive evaluation methods, achieves a leap from experience-based judgment to data-driven assessment, and from result sampling to full-process transparency in evaluating the degassing effect of XLPE cables, providing a reliable basis for optimizing cable manufacturing processes and controlling product quality.

[0199] In one exemplary embodiment, a comprehensive degassing effect evaluation report can be further generated using a comprehensive degassing effect evaluation model. An example of a comprehensive degassing effect evaluation report is as follows:

[0200]

[0201]

[0202]

[0203]

[0204] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0205] Based on the same inventive concept, this application also provides a degassing effect evaluation device for implementing the degassing effect evaluation method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the degassing effect evaluation device provided below can be found in the limitations of the degassing effect evaluation method described above, and will not be repeated here.

[0206] In one exemplary embodiment, such as Figure 6 As shown, a degassing effect evaluation device is provided, including: a signal acquisition module 602, a bubble distribution determination module 604, a feature acquisition module 606, and an evaluation module 608, wherein:

[0207] The signal acquisition module 602 is used to acquire the time-domain reflection signal, which is obtained by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer.

[0208] The bubble distribution determination module 604 is used to determine the bubble distribution in the insulation layer based on the time-domain reflection signal when the presence of bubbles in the insulation layer is identified based on the time-domain reflection signal.

[0209] The feature acquisition module 606 is used to acquire the trap features, space charge distribution features, and cross-linking byproduct features of the insulating layer;

[0210] Evaluation module 608 is used to evaluate the degassing effect of the cable under test based on bubble distribution, trap characteristics, space charge distribution characteristics and crosslinking byproduct characteristics.

[0211] In one exemplary embodiment, the evaluation module 608 is further configured to:

[0212] The propagation time interval and reflection intensity of the terahertz pulse in the insulating medium can be identified from the time-domain reflection signal.

[0213] Based on the propagation time interval and reflection intensity, locate the surface position of at least one bubble in the insulation layer;

[0214] The bubble distribution in the insulating layer is generated based on the position of each bubble on its surface.

[0215] In an exemplary embodiment, the feature acquisition module 606 is further configured to:

[0216] For any preset first test point on the insulation layer, the depolarization current data is obtained. The depolarization current data is obtained by performing a polarization-depolarization current test on the insulation layer from the preset first test point.

[0217] Extract the de-trapping current component from the depolarization current data;

[0218] By performing staged fitting of the detrapping current component, the trap charge density and trap energy level depth of the insulating layer at the preset first test site are determined, and the trap charge density and trap energy level depth are determined as the trap characteristics of the insulating layer.

[0219] In an exemplary embodiment, the feature acquisition module 606 is further configured to:

[0220] The space charge distribution of the insulating layer at a preset second test site is obtained by performing an electroacoustic pulse test on the insulating layer from the preset second test site.

[0221] Based on the space charge distribution, the level of opposite polarity charge accumulation and the internal electric field distortion rate of the insulating layer at the preset second test site are determined, and the level of opposite polarity charge accumulation and the internal electric field distortion rate are determined as the space charge distribution characteristics of the insulating layer.

[0222] In an exemplary embodiment, the feature acquisition module 606 is further configured to:

[0223] Based on the bubble distribution, the test area is located on the insulation layer, and the bubble distribution density in the test area is less than the preset density.

[0224] Infrared spectroscopy analysis of the test area was performed to detect the intensity of the characteristic absorption peak of the preset cross-linking byproducts, and the intensity of the characteristic absorption peak was determined as the characteristic of the cross-linking byproducts of the insulation layer.

[0225] In one exemplary embodiment, the evaluation module 608 is further configured to:

[0226] The bubble distribution, trap characteristics, space charge distribution characteristics, and cross-linking byproduct characteristics are input into a trained comprehensive evaluation model for degassing effect to evaluate the degassing effect of the cable under test, and the degassing uniformity index and comprehensive degassing grade are obtained.

[0227] The degassing effect evaluation device also includes a training module, which is used for:

[0228] The training time-domain reflectance signal is obtained by performing a terahertz time-domain spectral scan on the training cable, which includes a training insulation layer containing air bubbles.

[0229] The distribution of training bubbles in the training insulating layer is determined based on the training time-domain reflection signal;

[0230] Acquire training trap characteristics, training space charge distribution characteristics, and training crosslinking byproduct characteristics of the training insulating layer;

[0231] By inputting the training bubble distribution, training trap features, training space charge distribution features, and training crosslinking byproduct features into the machine learning model to be trained, the degassing effect of the training cable is evaluated, and the training degassing uniformity index and training comprehensive degassing level are obtained.

[0232] The degassing uniformity index label and the comprehensive degassing level label of the training insulation layer are obtained by destructive testing of the training cable.

[0233] Based on the differences between the training degassing uniformity index and the degassing uniformity index label, as well as the differences between the training comprehensive degassing level and the comprehensive degassing level label, the machine learning model to be trained is iteratively optimized to obtain a well-trained comprehensive evaluation model for degassing effect.

[0234] Each module in the aforementioned degassing effect evaluation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0235] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a degassing effect evaluation method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0236] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0237] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0238] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0239] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0240] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0241] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0242] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0243] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for evaluating degassing effect, characterized in that, The method includes: A time-domain reflectance signal is acquired by performing a terahertz time-domain spectral scan on the cable under test, which includes an insulation layer. If bubbles are detected in the insulating layer based on the time-domain reflection signal, the distribution of bubbles in the insulating layer is determined based on the time-domain reflection signal. Acquire the trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics of the insulating layer; The degassing effect of the cable under test is evaluated based on the bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics.

2. The method according to claim 1, characterized in that, Determining the bubble distribution in the insulating layer based on the time-domain reflection signal includes: The propagation time interval and reflection intensity of the terahertz pulse in the insulating medium can be identified from the time-domain reflection signal. Based on the propagation time interval and the reflection intensity, locate the surface position of at least one bubble in the insulating layer; The bubble distribution in the insulating layer is generated based on the respective bubble surface positions.

3. The method according to claim 1, characterized in that, The acquisition of the trap features of the insulating layer includes: For any preset first test point on the insulating layer, depolarization current data is obtained. The depolarization current data is obtained by performing polarization-depolarization current test on the insulating layer from the preset first test point. Extract the depolarization current component from the depolarization current data; By performing staged fitting on the detrapping current component, the trap charge density and trap energy level depth of the insulating layer at the preset first test site are determined, and the trap charge density and trap energy level depth are determined as the trap characteristics of the insulating layer.

4. The method according to claim 1, characterized in that, To obtain the characteristics of space charge distribution, including: The space charge distribution of the insulating layer at a preset second test site is obtained, and the space charge distribution is obtained by performing an electroacoustic pulse test on the insulating layer from the preset second test site; Based on the space charge distribution, the level of opposite polarity charge accumulation and the internal electric field distortion rate of the insulating layer at the preset second test site are determined, and the level of opposite polarity charge accumulation and the internal electric field distortion rate are determined as the space charge distribution characteristics.

5. The method according to claim 1, characterized in that, Acquire characteristics of cross-linking byproducts, including: Based on the bubble distribution, a test area is located on the insulating layer, wherein the bubble distribution density of the test area is less than a preset density. By performing infrared spectral analysis on the test area, the intensity of the characteristic absorption peak of the preset crosslinking byproduct is detected, and the intensity of the characteristic absorption peak is determined as the characteristic of the crosslinking byproduct of the insulating layer.

6. The method according to any one of claims 1 to 5, characterized in that, The evaluation of the degassing effect of the cable under test based on the bubble distribution, trap characteristics, space charge distribution characteristics, and crosslinking byproduct characteristics includes: The bubble distribution, trap features, space charge distribution features, and crosslinking byproduct features are input into a trained comprehensive degassing effect evaluation model to evaluate the degassing effect of the cable under test, and the degassing uniformity index and comprehensive degassing level are obtained. The model training process of the comprehensive evaluation model for degassing effect includes: A training time-domain reflectance signal is obtained by performing a terahertz time-domain spectral scan on a training cable, the training cable including a training insulation layer containing air bubbles; The distribution of training bubbles in the training insulating layer is determined based on the training time-domain reflection signal. Acquire the training trap characteristics, training space charge distribution characteristics, and training crosslinking byproduct characteristics of the training insulating layer; By inputting the training bubble distribution, the training trap features, the training space charge distribution features, and the training crosslinking byproduct features into the machine learning model to be trained, the degassing effect of the training cable is evaluated, and the training degassing uniformity index and the training comprehensive degassing level are obtained. Obtain the degassing uniformity index label and the comprehensive degassing level label of the training insulation layer. The degassing uniformity index label and the comprehensive degassing level label are obtained by performing destructive testing on the training cable. Based on the difference between the training degassing uniformity index and the degassing uniformity index label, and the difference between the training comprehensive degassing level and the comprehensive degassing level label, the machine learning model to be trained is iteratively optimized to obtain a well-trained comprehensive evaluation model for degassing effect.

7. A degassing effect evaluation device, characterized in that, The device includes: The signal acquisition module is used to acquire the time-domain reflection signal, which is obtained by performing a terahertz time-domain spectral scan on the cable under test, and the cable under test includes an insulation layer. A bubble distribution determination module is used to determine the bubble distribution in the insulating layer based on the time-domain reflection signal when bubbles are identified in the insulating layer based on the time-domain reflection signal. The feature acquisition module is used to acquire the trap features, space charge distribution features, and crosslinking by-product features of the insulating layer; The evaluation module is used to evaluate the degassing effect of the cable under test based on the bubble distribution, the trap characteristics, the space charge distribution characteristics, and the crosslinking byproduct characteristics.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images