A method and apparatus for identifying defects in power cables

Abstract

The application relates to the technical field of power cable fault diagnosis, and particularly provides a power cable defect identification method and device, which comprises the following steps: acquiring terahertz frequency domain imaging of a power cable and second-order cumulants corresponding to local discharge parameters of the power cable; inputting the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the local discharge parameters of the power cable into a pre-constructed power cable defect identification model to acquire a power cable defect category output by the pre-constructed power cable defect identification model; and the technical scheme provided by the application has high sensitivity and high resolution capability in cable defect identification. In particular, the scheme has low requirements on detection environment conditions and cable surface conditions, is simple and convenient to use, does not have any ionizing radiation, and can detect multilayer structures and layered defects which are difficult to detect by other technologies.

Description

Technical Field

[0001] This invention relates to the field of power cable fault diagnosis technology, and specifically to a method and apparatus for identifying power cable defects. Background Technology

[0002] Power cables are a crucial component of power system networks. With the continuous construction and expansion of power grids, the power load is growing rapidly. To meet the needs of urban beautification and power quality requirements, the length of high-voltage power cable lines is increasing rapidly. During normal grid operation, due to factors such as installation techniques and laying environment, as well as external forces and usage conditions, insulation defects and even insulation breakdown can occur, leading to safety accidents. Therefore, cable defects are a high-incidence area for power cable insulation accidents and a key point for accident prevention. Ensuring their safety and reliability is of great significance to guaranteeing the stability of the entire transmission line.

[0003] Partial discharge in power cables is often accompanied by physical and chemical phenomena, such as the emission of ultrasonic waves or electrical pulses, as well as electromagnetic radiation, light, temperature, and gases. These phenomena can be measured to indirectly monitor partial discharge signals and infer the severity and potential location of the partial discharge. Partial discharge is often associated with cable defects; therefore, partial discharge detection is also a method for cable defect detection. Extensive research has been conducted both domestically and internationally on partial discharge detection methods. The pulse current detection method is currently the most widely used. When a partial discharge occurs, sensors that detect impedance or current can detect the apparent discharge quantity of the measured object. Because the pulse current calculation result is the apparent discharge quantity and is very close to the actual discharge quantity, it is generally considered the most sensitive partial discharge detection method. However, partial discharge detection methods cannot detect information such as defect size and shape.

[0004] Currently, widely used non-destructive testing (NDT) technologies include ultrasonic testing, radiographic testing, and infrared thermography. In cable defect detection, radiographic testing offers direct and effective advantages, exhibiting high sensitivity and reliability for volumetric deformation defects. However, it struggles with common delamination defects due to lower sensitivity. Furthermore, high-energy radiation poses ionizing radiation hazards, complicates the operation process, and significantly affects the resolution of materials with inhomogeneous internal structures. Ultrasonic testing provides intuitive visuals and is fast, but it suffers from significant attenuation in multilayered materials, resulting in a low signal-to-noise ratio for the received echo signal. This method is also complex, requiring the use of a coupling agent and different probe specifications for different types of defects. Additionally, ultrasonic testing requires a certain degree of surface smoothness, necessitating the filling of the gap between the probe and the surface with the coupling agent. Therefore, it is difficult to implement on insulated equipment with surface contamination after operation. Moreover, ultrasonic testing is not very intuitive in defect identification, making qualitative and quantitative analysis difficult. These characteristics limit the application scope of ultrasonic testing. Infrared thermography can accurately detect common delamination defects within materials with high efficiency and intuitive image-based results. However, this method is very limited in detecting defects in thermal insulation materials. Its reliance on the material's thermal conductivity severely restricts its application, and it is easily affected by environmental factors such as sunlight, strong winds, and humidity, which can cause changes in the surface temperature of the insulation material and thus affect the test results. Furthermore, infrared detection judges the quality of insulation materials by heating, making it unsuitable for detecting early-stage internal defects in insulation materials that have not yet shown a significant temperature rise. Therefore, there is a need to find a more practical and valuable detection method to fill the gap in the field of non-destructive testing. Summary of the Invention

[0005] To overcome the above-mentioned shortcomings, the present invention proposes a method and apparatus for identifying defects in power cables.

[0006] Firstly, a method for identifying defects in power cables is provided, the method comprising:

[0007] Obtain terahertz frequency domain imaging of power cables and the second-order cumulants corresponding to the partial discharge parameters of power cables;

[0008] The terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable are input into a pre-built power cable defect identification model to obtain the power cable defect category output by the pre-built power cable defect identification model.

[0009] The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable.

[0010] Preferably, acquiring terahertz frequency domain imaging of power cables includes:

[0011] The complex reflection coefficient of the power cable is determined based on the terahertz time-domain signal reflected back from the power cable and the terahertz time-domain signal reflected back from the reference sample.

[0012] The absorption coefficient of a power cable is determined based on its complex reflection coefficient.

[0013] The real part of the complex refractive index of a power cable is determined based on its absorption coefficient.

[0014] A two-dimensional matrix is ​​constructed using the absorption coefficient and the real part of the complex refractive index of the power cable, and a grayscale image is generated based on the two-dimensional matrix.

[0015] The grayscale image is used as a terahertz frequency domain image of the power cable.

[0016] Furthermore, the formula for calculating the complex reflection coefficient of the power cable is as follows:

[0017]

[0018] In the above formula, R s E is the complex reflection coefficient of the power cable. sam E is the terahertz time-domain signal reflected back from the power cable. ref The terahertz time-domain signal reflected back from the reference sample.

[0019] Furthermore, the absorption coefficient of the power cable is calculated using the following formula:

[0020]

[0021] In the above formula, M s The absorption coefficient of the power cable. The angle is the reflection angle of the terahertz wave.

[0022] Furthermore, the formula for calculating the real part of the complex refractive index of the power cable is as follows:

[0023]

[0024] In the above formula, n is the real part of the complex refractive index of the power cable, Re[M s [M] represents the absorption coefficient of the power cable. s The real part.

[0025] Preferably, the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable includes: the second-order equivalent duration coefficient and the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0026] Furthermore, the second-order cumulative quantities corresponding to the partial discharge parameters of the power cable are obtained, including:

[0027] Extract the time-domain waveform signal corresponding to the partial discharge parameters of the power cable;

[0028] Convert the time-domain waveform signal into a frequency-domain waveform signal;

[0029] Calculate the expected time value and expected frequency value of the time-domain waveform signal and the expected frequency value of the frequency-domain waveform signal in the discrete domain, respectively;

[0030] Based on the expected time value and expected frequency value, determine the time standard deviation in the time domain and the frequency standard deviation in the frequency domain of a single pulse signal;

[0031] Calculate the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain for a single pulse signal, respectively.

[0032] The second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain is used as the second-order equivalent duration coefficient corresponding to the partial discharge parameters of the power cable, and the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain is used as the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0033] Furthermore, the mathematical expression for the time-domain waveform signal corresponding to the partial discharge parameters of the power cable is as follows:

[0034]

[0035] In the above formula, g(t) j Let be the time-domain waveform signal at the j-th point of the pulse, m be the total number of points constituting the pulse, and t be the time-domain waveform signal at the j-th point of the pulse. j b is the time corresponding to the j-th point of the pulse. j Let be the time-domain pulse value at the j-th point of the pulse.

[0036] Furthermore, the formulas for calculating the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain of a single pulse signal are as follows:

[0037]

[0038] In the above formula, T0 i f is the expected time value of the time-domain waveform signal in the discrete domain. j Let F0 be the terahertz frequency at the j-th point of the pulse. i Let g(f) be the expected frequency value of the frequency domain waveform signal in the discrete domain. jLet be the frequency domain waveform signal corresponding to the time domain waveform signal at the j-th point of the pulse, T be the second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain, and F be the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain.

[0039] Preferably, the process of obtaining the pre-built power cable defect identification model includes:

[0040] The historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable are used as the input layer training data of the neural network model, and the cable defect category corresponding to the acquisition time of the historical partial discharge parameters of the power cable is used as the output layer training data of the neural network model. The neural network model is trained to obtain the pre-constructed power cable defect identification model.

[0041] Secondly, a power cable defect identification device is provided, the power cable defect identification device comprising:

[0042] The first acquisition module is used to acquire the terahertz frequency domain imaging of the power cable and the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable.

[0043] The second acquisition module is used to input the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable into a pre-built power cable defect identification model, and to obtain the power cable defect category output by the pre-built power cable defect identification model.

[0044] The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable.

[0045] Thirdly, a storage medium is provided, the storage medium including a stored program, wherein, when the program is executed, the device where the storage medium is located is controlled to perform the power cable defect identification method.

[0046] Fourthly, a processor is provided for running a program, wherein the program executes the power cable defect identification method during runtime.

[0047] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects:

[0048] This invention relates to the field of power cable fault diagnosis technology, specifically providing a method and apparatus for power cable defect identification. The method includes: acquiring terahertz frequency domain imaging of the power cable and the second-order cumulative quantities corresponding to the partial discharge parameters of the power cable; inputting the terahertz frequency domain imaging of the power cable and the second-order cumulative quantities corresponding to the partial discharge parameters of the power cable into a pre-constructed power cable defect identification model, and obtaining the power cable defect category output by the pre-constructed power cable defect identification model; wherein the pre-constructed power cable defect identification model is trained based on historical terahertz frequency domain imaging of the power cable and the second-order cumulative quantities corresponding to historical partial discharge parameters of the power cable. The technical solution provided by this invention, through the acquisition, processing, and analysis of partial discharge signals, can effectively detect defects and degradation of the insulation layer of power cables, providing a basis for evaluating the insulation performance of cables. This invention combines terahertz technology with partial discharge technology, exhibiting high sensitivity and high resolution in cable defect identification. In particular, it has low requirements for detection environmental conditions and cable surface conditions, the equipment is simple and convenient to use, there is no ionizing radiation, and it can detect multi-layered structures and layered defects that are difficult to detect by other technologies. Attached Figure Description

[0049] Figure 1 This is a schematic flowchart of the main steps of the power cable defect identification method according to an embodiment of the present invention;

[0050] Figure 2 This is a main structural block diagram of the power cable defect identification device according to an embodiment of the present invention. Detailed Implementation

[0051] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Terahertz (THz) waves generally refer to electromagnetic waves with frequencies between 0.1 and 10 THz and wavelengths between 0.03 and 3 mm in the electromagnetic spectrum, falling between microwaves and infrared waves. In terms of frequency, this band belongs to the far-infrared band, which is the transition zone from macroscopic electronics to microscopic optoelectronics. Due to the special characteristics of the terahertz band, it has many unique physical properties. (1) Penetration: Many non-metallic and non-polar materials, such as ceramics, foams, plastics, and rubber, lack corresponding vibration modes and have low scattering rates in the terahertz band, so terahertz waves have good penetration over them. (2) Safety: Terahertz waves have low photon energy, only 10-3 eV, which is one millionth of the energy of X-ray photons. They will not cause harmful ionization and damage the detected substances, thus having safe and green detection characteristics. (3) High resolution: The short wavelength characteristics of terahertz waves give them higher spatial resolution in imaging applications, or greater depth of field while maintaining the same spatial resolution. (4) Coherence: Terahertz coherent measurement technology can directly measure the amplitude and phase of the field, thus facilitating the extraction of optical parameters such as refractive index and absorptivity of the test object. Compared with other detection methods, terahertz waves can pass through materials that are impenetrable by visible light, infrared waves, and even ultrasound, such as foam, ceramics, plastics, polymer composites, and magnetic materials. Terahertz time-domain spectroscopy (THz-TDS) detection technology has advantages such as rapid response, excellent detection accuracy and repeatability, no need to contact the sample surface, quantitative analysis of defect location and size, and high defect resolution, making terahertz technology well-suited for non-destructive testing of internal defects in cable materials.

[0054] Power cable defects are characterized by randomness, ambiguity, and uncertainty. When a defect occurs, it is related to multiple pieces of information, or a single piece of information may reflect the occurrence of multiple defects, and different fault states may coexist. Currently, relying on a single sensor cannot comprehensively identify power cable defects. To address this, this invention utilizes the differences and complementarities of information collected by multiple sensors to compensate for the shortcomings of detection based on single information. This invention combines terahertz technology and partial discharge technology, leveraging their complementary advantages. Simultaneously, a deep learning-based defect identification method projects low-dimensional network node terahertz information and partial discharge information into a high-dimensional space using a deep neural network. Based on optimization objectives such as keeping data of the same type of defect as close as possible and data of different types of defects as far apart as possible, weight calculations and reconstruction transformations are performed in the high-dimensional space to obtain feature indicators that can identify different defects, thereby improving the separability between different defects. This leads to the construction of a defect identification classifier with good recognition performance to achieve cable defect identification.

[0055] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of a power cable defect identification method according to an embodiment of the present invention. Figure 1 As shown, the power cable defect identification method in this embodiment of the invention mainly includes the following steps:

[0056] Step S101: Obtain the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable;

[0057] Step S102: Input the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable into the pre-built power cable defect identification model, and obtain the power cable defect category output by the pre-built power cable defect identification model;

[0058] The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable.

[0059] Specifically, acquiring terahertz frequency domain imaging of power cables includes:

[0060] The complex reflection coefficient of the power cable is determined based on the terahertz time-domain signal reflected back from the power cable and the terahertz time-domain signal reflected back from the reference sample.

[0061] The absorption coefficient of a power cable is determined based on its complex reflection coefficient.

[0062] The real part of the complex refractive index of a power cable is determined based on its absorption coefficient.

[0063] A two-dimensional matrix is ​​constructed using the absorption coefficient and the real part of the complex refractive index of the power cable, and a grayscale image is generated based on the two-dimensional matrix.

[0064] The grayscale image is used as a terahertz frequency domain image of the power cable.

[0065] In one embodiment, the complex reflection coefficient of the power cable is calculated as follows:

[0066]

[0067] In the above formula, R s E is the complex reflection coefficient of the power cable. sam E is the terahertz time-domain signal reflected back from the power cable. ref The terahertz time-domain signal reflected back from the reference sample.

[0068] In one embodiment, the absorption coefficient of the power cable is calculated as follows:

[0069]

[0070] In the above formula, Ms The absorption coefficient of the power cable. The angle is the reflection angle of the terahertz wave.

[0071] In one embodiment, the real part of the complex refractive index of the power cable is calculated as follows:

[0072]

[0073] In the above formula, n is the real part of the complex refractive index of the power cable, Re[M s [M] represents the absorption coefficient of the power cable. s The real part.

[0074] In this embodiment, the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable includes: the second-order equivalent duration coefficient and the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0075] In one embodiment, obtaining the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable includes:

[0076] Extract the time-domain waveform signal corresponding to the partial discharge parameters of the power cable;

[0077] Convert the time-domain waveform signal into a frequency-domain waveform signal;

[0078] Calculate the expected time value and expected frequency value of the time-domain waveform signal and the expected frequency value of the frequency-domain waveform signal in the discrete domain, respectively;

[0079] Based on the expected time value and expected frequency value, determine the time standard deviation in the time domain and the frequency standard deviation in the frequency domain of a single pulse signal;

[0080] Calculate the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain for a single pulse signal, respectively.

[0081] The second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain is used as the second-order equivalent duration coefficient corresponding to the partial discharge parameters of the power cable, and the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain is used as the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0082] The mathematical expression for the time-domain waveform signal corresponding to the partial discharge parameters of the power cable is as follows:

[0083]

[0084] In the above formula, g(t) j Let be the time-domain waveform signal at the j-th point of the pulse, m be the total number of points constituting the pulse, and t be the time-domain waveform signal at the j-th point of the pulse. jb is the time corresponding to the j-th point of the pulse. j Let be the time-domain pulse value at the j-th point of the pulse.

[0085] The formulas for calculating the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain of a single pulse signal are as follows:

[0086]

[0087] In the above formula, T0 i f is the expected time value of the time-domain waveform signal in the discrete domain. j Let F0 be the terahertz frequency at the j-th point of the pulse. i Let g(f) be the expected frequency value of the frequency domain waveform signal in the discrete domain. j Let be the frequency domain waveform signal corresponding to the time domain waveform signal at the j-th point of the pulse, T be the second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain, and F be the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain.

[0088] In this embodiment, the process of obtaining the pre-built power cable defect identification model includes:

[0089] The historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable are used as the input layer training data of the neural network model, and the cable defect category corresponding to the acquisition time of the historical partial discharge parameters of the power cable is used as the output layer training data of the neural network model. The neural network model is trained to obtain the pre-constructed power cable defect identification model.

[0090] In one embodiment, the cable defect category can be air gap, crack, or debonding.

[0091] In an optimal implementation, the cable defect identification method based on the above-mentioned combination of terahertz technology and partial discharge technology, combined with terahertz time-domain spectroscopy (THz-TDS), partial discharge detection technology and neural network technology, can achieve high sensitivity and high resolution identification of cable defects to meet cable inspection needs.

[0092] The method includes the following steps:

[0093] Select various cable samples. Specifically, the cable samples can be 10-35kV urban cables with a length between 100-300mm.

[0094] A terahertz time-domain spectroscopy system was used to acquire and analyze the spectra and images of various cable samples. Specifically, the measurement was performed using a reflectance method. Based on the formula mentioned above, the optical characteristic parameters (refractive index and absorption coefficient) of the cable samples were obtained from the measured spectral data, and a grayscale image could be obtained.

[0095] A partial discharge detector was used to collect and analyze partial discharge signals from various cable samples. Specifically, characteristic parameters of the partial discharge signal were extracted according to the formula mentioned above.

[0096] The terahertz and partial discharge data of various cable samples were processed. For specific data processing methods using deep learning and neural networks, please refer to the appendix. Figure 1 As shown, a standard feature library is established;

[0097] A neural network is built and trained using a standard feature library. Finally, the built neural network is used to identify the type of feature parameters and defects to be identified.

[0098] Based on the same inventive concept, this invention provides a power cable defect identification device, such as... Figure 2 As shown, the power cable defect identification device includes:

[0099] The first acquisition module is used to acquire the terahertz frequency domain imaging of the power cable and the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable.

[0100] The second acquisition module is used to input the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable into a pre-built power cable defect identification model, and to obtain the power cable defect category output by the pre-built power cable defect identification model.

[0101] The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable.

[0102] Preferably, acquiring terahertz frequency domain imaging of power cables includes:

[0103] The complex reflection coefficient of the power cable is determined based on the terahertz time-domain signal reflected back from the power cable and the terahertz time-domain signal reflected back from the reference sample.

[0104] The absorption coefficient of a power cable is determined based on its complex reflection coefficient.

[0105] The real part of the complex refractive index of a power cable is determined based on its absorption coefficient.

[0106] A two-dimensional matrix is ​​constructed using the absorption coefficient and the real part of the complex refractive index of the power cable, and a grayscale image is generated based on the two-dimensional matrix.

[0107] The grayscale image is used as a terahertz frequency domain image of the power cable.

[0108] Furthermore, the formula for calculating the complex reflection coefficient of the power cable is as follows:

[0109]

[0110] In the above formula, R s E is the complex reflection coefficient of the power cable. sam E is the terahertz time-domain signal reflected back from the power cable. ref The terahertz time-domain signal reflected back from the reference sample.

[0111] Furthermore, the absorption coefficient of the power cable is calculated using the following formula:

[0112]

[0113] In the above formula, M s The absorption coefficient of the power cable. The angle is the reflection angle of the terahertz wave.

[0114] Furthermore, the formula for calculating the real part of the complex refractive index of the power cable is as follows:

[0115]

[0116] In the above formula, n is the real part of the complex refractive index of the power cable, Re[M s [M] represents the absorption coefficient of the power cable. s The real part.

[0117] Preferably, the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable includes: the second-order equivalent duration coefficient and the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0118] Furthermore, the second-order cumulative quantities corresponding to the partial discharge parameters of the power cable are obtained, including:

[0119] Extract the time-domain waveform signal corresponding to the partial discharge parameters of the power cable;

[0120] Convert the time-domain waveform signal into a frequency-domain waveform signal;

[0121] Calculate the expected time value and expected frequency value of the time-domain waveform signal and the expected frequency value of the frequency-domain waveform signal in the discrete domain, respectively;

[0122] Based on the expected time value and expected frequency value, determine the time standard deviation in the time domain and the frequency standard deviation in the frequency domain of a single pulse signal;

[0123] Calculate the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain for a single pulse signal, respectively.

[0124] The second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain is used as the second-order equivalent duration coefficient corresponding to the partial discharge parameters of the power cable, and the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain is used as the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

[0125] Furthermore, the mathematical expression for the time-domain waveform signal corresponding to the partial discharge parameters of the power cable is as follows:

[0126]

[0127] In the above formula, g(t) j Let be the time-domain waveform signal at the j-th point of the pulse, m be the total number of points constituting the pulse, and t be the time-domain waveform signal at the j-th point of the pulse. j b is the time corresponding to the j-th point of the pulse. j Let be the time-domain pulse value at the j-th point of the pulse.

[0128] Furthermore, the formulas for calculating the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain of a single pulse signal are as follows:

[0129]

[0130] In the above formula, T0 i f is the expected time value of the time-domain waveform signal in the discrete domain. j Let F0 be the terahertz frequency at the j-th point of the pulse. i Let g(f) be the expected frequency value of the frequency domain waveform signal in the discrete domain. j Let be the frequency domain waveform signal corresponding to the time domain waveform signal at the j-th point of the pulse, T be the second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain, and F be the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain.

[0131] Preferably, the process of obtaining the pre-built power cable defect identification model includes:

[0132] The historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable are used as the input layer training data of the neural network model, and the cable defect category corresponding to the acquisition time of the historical partial discharge parameters of the power cable is used as the output layer training data of the neural network model. The neural network model is trained to obtain the pre-constructed power cable defect identification model.

[0133] Furthermore, the present invention provides a storage medium comprising a stored program, wherein the program, when running, controls the device containing the storage medium to execute the power cable defect identification method.

[0134] Furthermore, the present invention provides a processor for running a program, wherein the program executes the power cable defect identification method during runtime.

[0135] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0137] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0138] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for identifying defects in power cables, characterized in that, The method includes: Obtain terahertz frequency domain imaging of power cables and the second-order cumulants corresponding to the partial discharge parameters of power cables; The terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable are input into a pre-built power cable defect identification model to obtain the power cable defect category output by the pre-built power cable defect identification model. The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable. Acquiring terahertz frequency domain imaging of power cables includes: The complex reflection coefficient of the power cable is determined based on the terahertz time-domain signal reflected back from the power cable and the terahertz time-domain signal reflected back from the reference sample. The absorption coefficient of a power cable is determined based on its complex reflection coefficient. The real part of the complex refractive index of a power cable is determined based on its absorption coefficient. A two-dimensional matrix is ​​constructed using the absorption coefficient and the real part of the complex refractive index of the power cable, and a grayscale image is generated based on the two-dimensional matrix. The grayscale image is used as a terahertz frequency domain image of the power cable; The second-order cumulative quantities corresponding to the partial discharge parameters of the power cable include: the second-order equivalent duration coefficient and the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable. Obtain the second-order cumulative quantities corresponding to the partial discharge parameters of the power cable, including: Extract the time-domain waveform signal corresponding to the partial discharge parameters of the power cable; Convert the time-domain waveform signal into a frequency-domain waveform signal; Calculate the expected time value and expected frequency value of the time-domain waveform signal and the expected frequency value of the frequency-domain waveform signal in the discrete domain, respectively; Based on the expected time value and expected frequency value, determine the time standard deviation in the time domain and the frequency standard deviation in the frequency domain of a single pulse signal; Calculate the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain for a single pulse signal, respectively. The second-order cumulative amount of the time standard deviation of a single pulse signal in the time domain is used as the second-order equivalent duration coefficient corresponding to the partial discharge parameters of the power cable, and the second-order cumulative amount of the frequency standard deviation of a single pulse signal in the time domain is used as the second-order equivalent bandwidth coefficient corresponding to the partial discharge parameters of the power cable.

2. The method as described in claim 1, characterized in that, The formula for calculating the complex reflection coefficient of the power cable is as follows: In the above formula, The complex reflection coefficient of the power cable. The terahertz time-domain signal reflected back from the power cable. The terahertz time-domain signal reflected back from the reference sample.

3. The method as described in claim 2, characterized in that, The absorption coefficient of the power cable is calculated using the following formula: In the above formula, The absorption coefficient of the power cable. The angle is the reflection angle of the terahertz wave.

4. The method as described in claim 3, characterized in that, The formula for calculating the real part of the complex refractive index of the power cable is as follows: In the above formula, Let be the real part of the complex refractive index of the power cable. Absorption coefficient of power cable The real part.

5. The method as described in claim 4, characterized in that, The mathematical expression for the time-domain waveform signal corresponding to the partial discharge parameters of the power cable is as follows: In the above formula, For the pulse number j Time-domain waveform signal at each point, m The total number of points that make up the pulse. For the pulse number j The time corresponding to each point For the pulse number j The time-domain pulse value of each point.

6. The method as described in claim 5, characterized in that, The formulas for calculating the second-order cumulant of the time standard deviation in the time domain and the second-order cumulant of the frequency standard deviation in the frequency domain of a single pulse signal are as follows: In the above formula, Let the time-domain waveform signal be the expected value in the discrete domain. For the pulse number j The terahertz frequency of each point The frequency of the frequency domain waveform signal is the expected value of its frequency in the discrete domain. For the pulse number j The frequency domain waveform signal corresponding to each point in the time domain waveform signal This is the second-order cumulant of the time standard deviation of a single pulse signal in the time domain. It is the second-order cumulative quantity of the frequency standard deviation in the time domain and frequency domain of a single pulse signal.

7. The method as described in claim 1, characterized in that, The process of obtaining the pre-built power cable defect identification model includes: The historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable are used as the input layer training data of the neural network model, and the cable defect category corresponding to the acquisition time of the historical partial discharge parameters of the power cable is used as the output layer training data of the neural network model. The neural network model is trained to obtain the pre-constructed power cable defect identification model.

8. An apparatus based on the power cable defect identification method according to any one of claims 1-7, characterized in that, The device includes: The first acquisition module is used to acquire the terahertz frequency domain imaging of the power cable and the second-order cumulative quantity corresponding to the partial discharge parameters of the power cable. The second acquisition module is used to input the terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the partial discharge parameters of the power cable into a pre-built power cable defect identification model, and to obtain the power cable defect category output by the pre-built power cable defect identification model. The pre-built power cable defect identification model is obtained by training based on the historical terahertz frequency domain imaging of the power cable and the second-order cumulants corresponding to the historical partial discharge parameters of the power cable.

9. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the storage medium to perform the method described in any one of claims 1 to 7.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 7 when it runs.

Images