High-voltage cable typical insulation defect live detection method based on peak value comparison

Abstract

The invention belongs to the field of power equipment, and particularly relates to a high-voltage cable typical insulation defect live detection method based on peak value comparison. In field measurement of the high-voltage cable, the coupling capacitor wraps the outer side of the high-voltage cable to inject and receive signals of the high-voltage cable, and when a buffer layer or an insulating layer of the high-voltage cable is affected with damp, corroded or ablated, the volume resistivity and the relative dielectric constant will change. And impedance discontinuity points occur in a transmission line network of the cable. According to a transmission line theory, a signal can be refracted and reflected when passing through an impedance discontinuous point in the cable. A corresponding amplitude-frequency diagram can be obtained by processing a received signal. The device comprises a coupling capacitor, a signal generator, a signal collector and a computer. The cable does not need to be disassembled for maintenance, and the detection and maintenance efficiency is greatly improved.

Description

Technical Field

[0001] This invention belongs to the field of power equipment, specifically to a live detection method for typical insulation defects in high-voltage cables based on peak value comparison. Background Technology

[0002] Cross-linked polyethylene (XLPE) cables have been widely used in power systems both domestically and internationally due to their excellent performance, simple manufacturing process, and easy installation. However, with the increasing coverage of power cables, not only have the cable lines become more complex, but cables put into operation earlier are also entering their aging phase. To ensure the stable operation of cables in power systems, it is necessary to regularly inspect and evaluate the insulation condition of power cables. Current insulation condition inspection methods still have many shortcomings: repeated tests on cables with different insulation conditions lack specificity; multiple preventative tests can easily damage cables with good insulation, thus accelerating the aging of cable insulation; offline insulation condition inspection methods lack real-time capability, making it difficult to promptly grasp the insulation condition of power cables. Therefore, it is necessary to propose an online insulation condition inspection method for power cables. Summary of the Invention

[0004] The purpose of this invention is to solve the problem of difficulty in online monitoring of existing high-voltage cables, and to realize online monitoring of the insulation performance of high-voltage cables. A live detection method for typical insulation defects of high-voltage cables based on peak comparison is proposed.

[0005] The technical solution of this invention is: a live-line detection method for typical insulation defects in high-voltage cables based on peak comparison. In field measurements of high-voltage cables, a coupling capacitor is wrapped around the outside of the high-voltage cable to inject and receive signals. When the buffer layer or insulation layer of the high-voltage cable becomes damp, corroded, or ablated, the volume resistivity and relative permittivity will change, resulting in impedance discontinuities in the cable's transmission line network. According to transmission line theory, signals will undergo refraction and reflection when passing through impedance discontinuities in the cable. By processing the received signal, the corresponding amplitude-frequency diagram can be obtained.

[0006] A live-line detection device for typical insulation defects in high-voltage cables based on peak comparison includes a coupling capacitor, a signal generator, a signal acquisition unit, and a computer, wherein: A pair of coupling capacitors are used for the injection and reception of transmitted signals within the cable. The signal is transmitted from the coupling capacitor to the internal metal sheath and copper core of the cable, and then transmitted out from the receiving end through the coupling capacitor, completing one transmission process of the signal within the cable.

[0007] A signal generator is used to generate function and frequency sweep signals.

[0008] A signal acquisition device is used to receive the acquired transmission signals and transmit them to a personal computer for processing and transformation to obtain the required charts and data.

[0009] The computer is used to process the received signals, convert the time-domain signals into corresponding frequency-domain signals, extract the peak parameters of the collected information, and obtain the corresponding fault relationships of the cable after processing and comparing the parameters.

[0010] The beneficial effects of this invention are: 1. By using coupling capacitors to input and receive signals from high-voltage cables, the signal transmission results inside the cable are obtained through the coupling of the capacitors before and after transmission. This reveals signal variations during transmission through the high-voltage cable, which are then processed accordingly. 2. By extracting the corresponding amplitude-frequency diagram and peak parameters from the signal, the insulation defects of high-voltage cables can be identified. On-site personnel can perform online operations without disassembling the cable for inspection. Detection can be performed simply by placing two coupling capacitors at the ends of potentially faulty areas, greatly improving detection and repair efficiency. 3. Simultaneously, the detection range is effectively controlled by adjusting the placement of the input and receiving capacitors. This not only improves detection efficiency but also prevents excessive signal attenuation after long-distance transmission. 4. After receiving the signal, the computer processes it, converting the signal in the time domain into a signal in the frequency domain, thus obtaining the signal's frequency characteristics. The spectrum is more clearly defined than the time domain characteristics, making it easier to calculate and identify. Furthermore, software extracts the peak values ​​of the signal at different frequencies and compares them using relevant formulas to obtain the required feature values, making the results more accurate and easier to identify. 5. This method works well for power frequency high-voltage cables. Because the voltage frequency is low, the frequency of the multiple harmonics generated is also low. Effective measurement results can be obtained by observing high-frequency signals without worrying about the high-voltage signal and its harmonics having too much influence on the measurement results. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the invention.

[0013] Figure 2 This is the single-phase cable distributed parameter model used in this invention.

[0014] Figure 3 This is the frequency domain cable distribution parameter model used in this invention.

[0015] Figure 4 This is a comparison chart of the changes in K value when n=1.

[0016] Figure 5 This is a graph showing the steps of K value change when n=2.

[0017] Figure 6 This is a diagram illustrating the operational steps of the present invention.

[0018] In the diagram, cable 1, signal generator 2, signal acquisition device 3, computer 4, and transfer function calculation 5 are shown. Detailed Implementation

[0020] As a transmission line for transmitting signals, power cables can be used to apply transmission line theory to cables. Under AC signals, power cables can be equivalent to a distributed parameter model to describe the transmission characteristics of signals in power cables. The distributed parameter model plays an important theoretical role in analyzing the transmission process of voltage signals.

[0021] For cables undergoing overall insulation aging, their insulation parameters will change significantly, thus affecting the cable's voltage transfer characteristics in the frequency domain. The cable model in the frequency domain is as follows: Figure 3 As shown.

[0022] Z c (f) represents the impedance per unit length of the conductor layer in the frequency domain, Y in (f) is the admittance per unit length of the insulating layer in the frequency domain, Z s (f) represents the impedance per unit length of the metal shielding layer in the frequency domain.

[0023] According to Kirchhoff's current and voltage laws, a relationship can be established with... Figure 3 Corresponding differential equations

[0024] U c (x) and U s (x) represent the voltage between the cable conductor and ground and the voltage between the cable's metallic shield and ground, respectively. The current in the cable conductor and the metallic shield is I. c (x) and I s (x), the general solution is:

[0025] in, C1, C2, C3, and C4 are coefficients determined by the boundary conditions of the second-order differential equation. Assuming the total length of the cable is l, the voltages at both ends of the cable can be obtained as follows:

[0026] From the above formula, we can see that the output voltage U c (l) will follow the formula l and γThe frequency of voltage transmission varies accordingly. In the frequency domain, the transmission efficiency of voltage at different frequencies is closely related to the cable length, insulation impedance, and the current frequency. Whether it increases or decreases requires actual calculation at a certain frequency and cable length. This invention is based on the above relationship between cable signal transmission efficiency and frequency.

[0027] For resistors, considering the skin effect, as the frequency increases, the time-varying magnetic field induces eddy currents inside the conductor, forcing the current to concentrate towards the surface. For a round conductor, the relationship between the AC resistance R_ac and the frequency f is:

[0028] Among them, f s This is the characteristic frequency of the skin effect, which is related to the wire diameter and material. When the frequency is much higher than f... s hour, .

[0029] For a capacitor, let f be the frequency of the circuit change, and ε r (f) represents the relative permittivity of XLPE at frequency f. In practical engineering, the dielectric response of insulating materials is usually expressed using the complex relative permittivity. To describe, where the real part The imaginary part characterizes energy storage capacity (usually referring to the relative permittivity). Characterizes dielectric loss.

[0030] According to the polarization response theory of dielectric physics, polarization intensity can be described. The relationship with time t. In the frequency domain, its core relationship can be expressed as:

[0031] Where, ε s ε is the static (or very low frequency) relative permittivity. ∞ Let be the relative permittivity at optical frequencies (extremely high frequencies), and τ be the polarization relaxation time constant. For solid dielectrics like XLPE that have multiple relaxation processes, their frequency domain response can be regarded as a superposition of multiple Debye or Cole-Cole models.

[0032] For inductors, the total inductance is divided into external inductance L e and internal inductance L i (f): L(f) = L e +L i (f) The internal inductance originates from the magnetic energy stored inside the conductor. At low frequencies, the magnetic field penetrates the entire cross-section. At high frequencies: the current concentrates on the surface, and the internal magnetic field attenuates. It eventually approaches 0.

[0033] See 1-6. The following will explain in detail with specific operation examples, relevant principles and reference diagrams.

[0034] A live-line detection method and device for typical insulation defects in high-voltage cables based on peak comparison: The first step is to find a suspected fault point. Since this technology is designed to detect insulation defects in high-voltage cables, it is necessary to find a suspected fault point on the cable being tested. The discovery of a suspected fault point can be done by observing the cable's appearance, temperature, noise, or by relying on existing experience with cable faults to make a preliminary location of the defect.

[0035] The second step is to install signal coupling devices before and after the fault point. Figure 1 The diagram illustrates a simulation of the detection process of this invention, including a coupling capacitor, a signal generator, a signal acquisition unit, and a computer. Taking the measurement of a 10kV distribution cable as an example, signal receivers and coupling devices are arranged at 1m intervals around the potential fault point. In this case, a cylindrical coupling capacitor made of a 10cm wide, 0.1mm thick copper sheet is used. The cable is artificially damaged, with 1cm² cuts made on the copper shield. 2 2cm 2 3cm 2 Defective holes, such as square holes, are then added to the power distribution cable. By changing the moisture content of the cotton cloth, different degrees of moisture absorption of the cable are simulated.

[0036] The third step is to inject a sweep frequency signal. A function generator is used, with the input coupling capacitor connected directly to the probes to inject the signal. The signal used here is a 0-50MHz sinusoidal sweep frequency signal with a sweep period of 2ms. Injecting this signal from the beginning of the cable section to be tested will result in a corresponding signal being received at the end of that section.

[0037] Because the method used is capacitive coupling, the signal is coupled from the externally installed capacitor into the internal metal shielding layer and copper core, and then transmitted to the other end of the cable. During this process, due to defects in the cable, the relevant parameters of the cable will change depending on the damage to the buffer layer, semiconductive layer, and copper shield. Therefore, the presence of defects in this section can be effectively determined by the signal transmission results.

[0038] The fourth step is to receive the output signal. When the signal is transmitted from the fault point to the receiving point 1m away, an oscilloscope is connected to the signal receiving coupling capacitor to acquire the coupled 0-50MHz sinusoidal sweep frequency signal. Because the sweep period used is short, multiple sets of repetitive signals can be acquired at once. By processing the repetitive signals, errors caused by background noise and other factors that may result from insufficient sample size can be effectively avoided.

[0039] The fifth step is to perform FFT processing on the signal. After receiving the swept frequency signal, the signal is transmitted to the computer, and mathematical software is used to perform FFT (Fast Fourier Decomposition) processing on the signal. FFT is an algorithm used to quickly calculate the Discrete Fourier Transform (DFT). DFT can transform a signal from the time (or space) domain to the frequency domain, helping to analyze the frequency components of the signal. For a given signal x[n] of length N, directly calculating the DFT requires N... 2 The computational workload is large. The FFT greatly reduces the computational workload by dividing the signal into two halves—even points and odd points—compiling their DFTs separately, and then merging the results.

[0040] The sixth step involves extracting and calculating the peak parameters of the output signal. After obtaining the frequency sweep signal in the frequency domain, it is necessary to compare the differences in characteristics before and after the fault to determine whether there is a potential fault in the cable segment. When a signal passes through a cable, it is amplified at a specific frequency due to inductance and capacitance—a phenomenon known as resonance. The signal is not only affected by resonance in the cable, but the cable parameters also exhibit different characteristics at different frequencies. This means that the signal is significantly affected by frequency variations during transmission, resulting in completely different characteristics at different frequencies. These characteristics can be used to compare the cable and determine whether a defect exists.

[0041] Measurements of the cable signals revealed that the small cross-sectional area of ​​the distribution cable inevitably reduced the coupling capacitance, significantly decreasing signal transmission efficiency. This resulted in greater susceptibility to external influences, with the transmitted signal amplitude being much smaller than the input signal, and the amplitude fluctuating considerably due to noise. Therefore, a new method for defect detection is needed.

[0042] During the study of the data, it was found that although the peak amplitude of the output signal varied greatly, multiple peaks increased or decreased simultaneously. The idea of ​​using division to process them was considered, but after processing, it was found that there was still a deviation. Considering that the capacitor is affected by the frequency during signal transmission, frequency processing is also required. Since multiple peaks were extracted from the waveform, after several attempts at calculation, the following formula was proposed.

[0043]

[0044] This formula effectively transforms a widely distributed signal into a highly correlated ratio, while also taking into account the impact of different peak frequencies at different locations. Under this formula, the observability of the signal is greatly enhanced.

[0045] The table below shows the peak characteristics of the output signal when the cable is not damaged. It can be seen that the first peak of waveforms four and five differs by 18% in the untreated case, which is far beyond the acceptable range. The amplitudes of the other waveforms also differ by about 10%, which causes great difficulty in signal identification. When n=1, the K values ​​of waveforms four and five, which originally had extremely large amplitude differences, only differed by 0.625%, while the largest differences in K values ​​for other waveforms were found in waveforms three and two, at 3%. This undoubtedly facilitated the identification of signal changes.

[0046] When n=2, the K values ​​of waveforms four and five differ by 3%. At this time, the largest K value difference for other waveforms is between waveforms three and four, at 4.8%. This is a significant reduction compared to the difference between signals obtained by the original amplitude discrimination method.

[0047] The seventh step is to calculate the difference in K-value between the output signals of a cable with good performance and a cable with a suspected fault.

[0048] When n=1, record the data extracted in the above experiment, and after processing it, obtain... Figure 4 . Figure 4 The K-values ​​of the signal under different defect conditions were observed. It was found that the K-values ​​of the four groups of signals with defects were almost unchanged compared with those of intact cables. Upon magnification, it was discovered that as the defect increased, the K-value showed a very small upward trend. However, after water was injected, the signal K-value immediately changed significantly, increasing noticeably from around 8.5 to 9.5-10.

[0049] When n=2, record the data extracted in the above experiment, and after processing it, obtain... Figure 5 The bottom left corner shows the K values ​​obtained under different defect conditions. Observation revealed that as the damage gap increases, the K value shows a very small upward trend. However, after water injection, the signal K value immediately showed a more significant change, with the K value increasing significantly from around 27 to 30-32.

[0050] Cables with 1-3cm damage 2 When the defect is detected, the signal shows only a very small change in characteristics, presumably because the cable itself is already somewhat aged, and small damage at the square centimeter level is difficult to detect in such an aged cable. In the moisture simulation, the signal change is more obvious after injecting 4ml of water, and the K value increases significantly.

Claims

1. A live-line detection method for typical insulation defects in high-voltage cables based on peak value comparison, characterized in that... In the field measurement of high-voltage cables, a coupling capacitor is wrapped around the outside of the high-voltage cable to inject and receive signals. When the buffer layer or insulation layer of the high-voltage cable becomes damp, corroded, or burned, the volume resistivity and relative permittivity will change, resulting in impedance discontinuities in the cable transmission network. When the signal passes through the impedance discontinuities in the cable, refraction and reflection will occur. By processing the received signal, the corresponding amplitude-frequency diagram can be obtained.

2. A live-line detection device for typical insulation defects in high-voltage cables based on peak value comparison, characterized in that... Includes a coupling capacitor connected to the cable (1), a signal generator (2), a signal acquisition unit (3), and a computer (4) for processing the received signal, wherein: A pair of coupling capacitors are used for the injection and reception of transmitted signals within the cable. The signal is transmitted from the coupling capacitor to the internal metal sheath and copper core of the cable, and then transmitted out from the receiving end through the coupling capacitor, completing one transmission process of the signal within the cable. Signal generator (2) is used to generate function and frequency sweep signals; The signal acquisition unit (3) is used to receive the acquired transmission signal and transmit the signal to the personal computer 4 for processing and transformation to obtain the required charts and data. Computer (4) is used to process the received signal, convert the time domain signal into the corresponding frequency domain signal, extract the peak parameters of the collected information, process and compare the parameters accordingly, and obtain the corresponding fault relationship of the cable.

Images