Method and system for power cable insulation detection and defect evaluation based on frequency-increased pulse

Abstract

The application discloses a kind of power cable insulation detection and defect evaluation method and system based on frequency-increasing pulse, it includes: obtaining the reflection signal generated by the propagation of frequency-increasing pulse in power cable;Time domain information extraction is carried out to frequency-increasing pulse and reflection signal, obtain the time difference between them, the pulse polarity of reflection signal, amplitude attenuation coefficient;Frequency domain information is extracted from frequency-increasing pulse and reflection signal, the frequency domain information extracted includes signal amplitude-frequency curve correlation coefficient, phase-frequency curve correlation coefficient, power spectral density curve correlation coefficient;Time-frequency domain information is extracted from frequency-increasing pulse and reflection signal;In time domain and frequency domain, the position of defect in power cable is calculated respectively, and the position of intermediate joint in power cable line is combined, to evaluate that defect is located in power cable body or joint;The type and severity of the defect of power cable are evaluated.The application improves the accuracy of defect positioning and evaluation.

Description

Technical Field

[0001] This invention relates to the field of power cable testing technology, and in particular to a method and system for power cable insulation testing and defect assessment based on frequency-increased pulses. Background Technology

[0002] With urbanization, power cables have been widely used in urban power grids and are maintaining a high growth rate. Although power cables are more reliable and safer than overhead lines, various types of defects can still occur during manufacturing, transportation, on-site installation, and operation, such as joint defects, moisture damage to the cable body, water treeing, electrical treeing, and buffer layer erosion. These defects pose a significant threat to power cable lines. Therefore, detecting and locating these defects and assessing their severity are urgent problems that the power sector needs to solve.

[0003] Currently, Time Domain Reflectometry (TDR) and Frequency Domain Reflectometry (FDR) are commonly used to detect and locate local defects in power cables. Both methods are based on the impedance mismatch between the defect location and the normal power cable. The difference lies in that TDR injects a pulse signal into the power cable to directly detect and identify the reflected signal caused by the defect in the time domain, while FDR injects a sinusoidal sweep signal with a linearly increasing frequency into the power cable and assesses the defect by identifying changes in the impedance spectral coefficients.

[0004] However, both methods have their own limitations. For TDR, the amplitude and polarity of the reflected pulse waveform can be directly obtained, making it relatively convenient for defect location and identification. However, its pulse signal has fewer high-frequency components, making it difficult to detect weak defects, and it also has a certain blind zone in terms of testing distance. On the other hand, FDR provides more high-frequency information during frequency domain testing, enabling the detection of minute defects. However, frequency domain reflected signals are easily distorted by interference, and frequency domain signals are difficult to use for defect or fault identification and assessment. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for power cable insulation detection and defect assessment based on frequency-increased pulses, which compromises the advantages of TDR and FDR, and can simultaneously detect and assess defects in power cables from both the time domain and the frequency domain.

[0006] This invention discloses a method for power cable insulation detection and defect assessment based on frequency-enhanced pulses, comprising:

[0007] Step 1: Obtain the reflected signal generated by the propagation of the frequency-enhanced pulse in the power cable;

[0008] Step 2: Extract time-domain information from the frequency-enhanced pulse and the reflected signal to obtain the time difference between them, the pulse polarity of the reflected signal, and the amplitude attenuation coefficient; extract frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the correlation coefficient of the signal amplitude-frequency curve, the correlation coefficient of the phase-frequency curve, and the correlation coefficient of the power spectral density curve; extract time-frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the correlation coefficient of the time-spectrum diagram and the generalized time-frequency dimension.

[0009] Step 3: Calculate the location of the defect in the power cable in the time and frequency domains respectively, and in combination with the location of the intermediate joint of the power cable line, assess whether the defect is located in the power cable body or the joint.

[0010] Step 4: Assess the type and severity of defects in the power cables.

[0011] Furthermore, prior to step 1, the procedure also includes:

[0012] Step 01: First, determine the bandwidth BWF of the frequency enhancement pulse sequence, the single frequency enhancement amplitude Δf, and the initial frequency f0 by using the relevant parameters of the power cable under test and the preset positioning accuracy; the relevant parameters include length, cross-sectional geometric parameters, and material parameters; the cross-sectional geometric parameters include the radius, and the material parameters include the relative permittivity of the material;

[0013] Step 02: Based on step 01, generate an enhanced frequency pulse.

[0014] Further, step 01 includes:

[0015] The formula for calculating the bandwidth (BWF) of an enhanced pulse sequence is:

[0016]

[0017] Where c is the speed of light, A is the positioning accuracy, i.e., the ratio of the positioning error to the length of the cable being measured (L), and ε is the position of the cable being measured. t The comprehensive equivalent relative permittivity of the dielectric material of the power cable under test;

[0018] ε t The method for determining this depends on the cross-sectional geometric parameters and material parameters of the power cable being tested.

[0019]

[0020] Where, r i Let ε be the radius of the i-th layer in the cross-section of the power cable. i Let be the relative permittivity of the i-th layer material;

[0021] The formula for calculating the single-cycle frequency increase amplitude Δf is:

[0022]

[0023] The formula for calculating the initial frequency f0 is:

[0024] f0 = 2Δf.

[0025] Further, step 02 includes: the host computer sending a control command to the controller, the controller communicating with the programmable waveform generator to generate a frequency-increasing pulse, the expression of which is:

[0026]

[0027] Where ZP(t) is the frequency-enhanced pulse signal at time t, and θ i For the initial phase, f i Let f be the frequency of the i-th boost pulse. i =f0+i×Δf, i=0,1,2,……1 / A-1; T is the interval time of the frequency enhancement pulse, which is greater than the time it takes for the frequency enhancement pulse to return once in the power cable.

[0028] Further, step 1 includes:

[0029] The frequency-enhanced pulse generated by the programmable waveform generator is amplified by a V-fold through a wideband power amplifier. The amplified frequency-enhanced pulse signal is then input to a power divider. One of the two outputs of the power divider is directly connected to the data sampling card, and the other is connected to a wideband directional coupler. The frequency-enhanced pulse is output to an impedance adapter through the wideband directional coupler. The output impedance of the impedance adapter is adjusted to be as close as possible to the characteristic impedance of the power cable under test to reduce the loss of the frequency-enhanced pulse signal injected into the power cable. The frequency-enhanced pulse propagates in the power cable, passing through the defect and the end of the power cable. During the test, the end is suspended, and impedance mismatch generates a reflected wave. This reflected wave is transmitted through the power cable to the impedance adapter, and then through the wideband directional coupler to the data acquisition card, where the reflected signal is acquired.

[0030] Furthermore, in step 2:

[0031] The time difference between the frequency-enhanced pulse and the reflected signal is calculated using the energy EC method, i.e.:

[0032] The energy curve formulas for the frequency-enhanced pulse and the reflected signal are as follows:

[0033]

[0034] Among them, P ZP and P ZPRLet be the average power of the frequency-enhanced pulse and the reflected signal, respectively, and N be the total number of data points; the time difference t is the difference between the global minimum of the differentials of the frequency-enhanced pulse and the reflected signal. d .

[0035] Furthermore, in step 2:

[0036] Pearson cross-correlation coefficients were calculated between the extracted amplitude-frequency curve, phase-frequency curve, and power spectral density curve and the amplitude-frequency curve, phase-frequency curve, and power spectral density curve of the defect under the action of the frequency-enhanced pulse, respectively, to obtain the amplitude-frequency curve correlation coefficient ρ. af The correlation coefficient ρ of the phase frequency curve pf Correlation coefficient ρ with power spectral density pd ;

[0037] For the time-spectrum image, first convert it to grayscale. Then, calculate the correlation coefficient between the grayscale time-spectrum image and the time-spectrum images of typical defects in the power cable body defect feature fingerprint database and the power cable joint defect feature fingerprint database using the following formula: ρ = (time-spectrum image correlation coefficient ρ) tf The calculation formula is:

[0038]

[0039] Where g and h are the dimensions of the spectral image (g×h), TF is the time-spectrum diagram of the test frequency-enhanced pulse signal, and M is the time-spectrum diagram of the defect's frequency-enhanced pulse signal. To test the grayscale mean of the time-spectrum of the frequency-increasing pulse signal, The average grayscale value of the spectrum of the frequency-enhanced pulse signal for the defect;

[0040] The formula for calculating the generalized time-frequency dimension is:

[0041]

[0042] T(j)=log(ε j )

[0043]

[0044] Where J is the number of data points in the time-frequency grid, S(j) is used to describe the generalized entropy, N is the number of grids, and P... i Let q be the probability that the i-th grid covers the time-frequency information set, and q be the scaling exponent. When q = 0, D q (ε j Let q be the capacity dimension, and when q = 1, D q (ε j Let q = 2, then D is the information dimension. q (ε j ) represents the cross-dimensionality of the association.

[0045] Further, step 3 includes:

[0046] Calculate the location of the defect in the power cable in the time domain:

[0047]

[0048] Among them, l t Where is the distance from the defect to the test end, and v is the velocity of the frequency-increasing pulse wave;

[0049] Calculate the location of the defect in the power cable in the frequency domain:

[0050] Performing an inverse discrete Fourier transform (IFT) on the reflected signal, and replacing the frequency in the IFT result with distance information based on the phase difference between the frequency-enhanced pulse and the reflected signal, yields an amplitude-distance spectrum. The distance at the amplitude peak is the location of the defect. t ;

[0051] The defect distance l calculated in the time and frequency domains t By comparing and considering the location of the intermediate joints in the power cable line, an assessment is made to determine whether the defect is in the power cable itself or at the joint; if l t The defect is not located at the joint of the power cable, meaning it is within the power cable itself; if l t If the location is a joint in the power cable, assess whether there are any defects in the joint.

[0052] Further, step 4 includes:

[0053] Based on the defect location results, if the defect is located in the power cable itself, the defect feature fingerprint database of the power cable itself is called. Through a neural network model, the feature parameters are compared with the defect feature fingerprint database of the power cable itself to determine the defect type and assess the severity. The feature parameters include the pulse polarity of the reflected signal, amplitude attenuation coefficient, amplitude-frequency curve correlation coefficient, phase-frequency curve correlation coefficient, power spectral density correlation coefficient, time-frequency spectrum correlation coefficient, time-frequency capacity dimension, time-frequency information dimension, and time-frequency correlation dimension.

[0054] If the defect is at the joint of the power cable, the power cable joint defect feature fingerprint database is called. The feature parameters are compared with the power cable joint defect feature fingerprint database through a neural network model to determine the defect type and assess the severity.

[0055] This invention also discloses a power cable insulation detection and defect assessment system based on frequency-increased pulses, implementing the power cable insulation detection and defect assessment method based on frequency-increased pulses described above. The system includes a frequency-increased pulse generation module, a signal separation and injection module, a data processing module, a host computer, a controller, and a data acquisition card. The frequency-increased pulse generation module includes a programmable waveform generator and a wideband power amplifier connected in sequence. The signal separation and injection module includes a power divider, a wideband directional coupler, and a bidirectional impedance adapter connected in sequence. The bidirectional impedance adapter is connected to the cable under test. The host computer is connected to the programmable waveform generator via the controller. The wideband power amplifier is connected to the power divider. The host computer is connected to the wideband directional coupler via the data acquisition card. The data processing module is used to analyze the frequency-increased pulse signal and the reflected signal to obtain the location of defects in the power cable and to assess the type and severity of the defects.

[0056] Because of the adoption of the above technical solution, the present invention has the following advantages:

[0057] 1. The frequency-enhanced pulse test combines the advantages of time-domain reflection and frequency-domain reflection methods, providing rich frequency domain information and high sensitivity for identifying minute defects. At the same time, the frequency-enhanced pulse reflected wave can directly identify the pulse polarity in the time domain. Compared with traditional frequency-domain detection methods, it can perform pulse polarity assessment, which can greatly improve the accuracy of defect type judgment.

[0058] 2. This invention improves the accuracy of defect location by using both time-domain and frequency-domain methods for comprehensive evaluation. At the same time, depending on whether the location result is on the power cable body or at the power cable joint, different defect evaluation fingerprint databases are used to improve the accuracy of defect identification and severity assessment. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0060] Figure 1 This is a flowchart illustrating a method for detecting and assessing the insulation defects of power cables based on frequency-enhanced pulses, according to an embodiment of the present invention.

[0061] Figure 2 This is a schematic diagram of the structure of the power distribution cable according to an embodiment of the present invention;

[0062] Figure 3 This is a schematic diagram of the frequency-enhancing pulse sequence according to an embodiment of the present invention;

[0063] Figure 4This is a time-frequency spectrum diagram of an embodiment of the present invention;

[0064] Figures 5(a) and 5(b) are the amplitude-distance spectrum and amplitude-distance-time spectrum of an embodiment of the present invention, respectively;

[0065] Figure 6 This is a schematic diagram of the framework of a power cable insulation detection and defect assessment system based on frequency-increased pulses according to an embodiment of the present invention;

[0066] Figure 7 This is a schematic diagram of the test waveform of the frequency-enhanced pulse sequence on a power distribution cable according to an embodiment of the present invention. Detailed Implementation

[0067] The present invention will be further described in conjunction with the accompanying drawings and embodiments. The described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art should fall within the protection scope of the present invention.

[0068] See Figure 1 This invention provides an embodiment of a method for power cable insulation detection and defect assessment based on frequency-enhanced pulses, comprising:

[0069] S1: Acquire the reflected signal generated by the propagation of the frequency-enhanced pulse in the power cable;

[0070] S2: Extract time-domain information from the frequency-enhanced pulse and the reflected signal to obtain the time difference between them, the pulse polarity of the reflected signal, and the amplitude attenuation coefficient; extract frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the correlation coefficient of the signal amplitude-frequency curve, the correlation coefficient of the phase-frequency curve, and the correlation coefficient of the power spectral density curve; extract time-frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the time-frequency spectrum correlation coefficient and the generalized time-frequency dimension; extract the amplitude and phase of the acquired time-domain waveform signal, and the resulting time-frequency spectrum is shown below. Figure 4 As shown.

[0071] S3: Calculate the location of the defect in the power cable in the time and frequency domains respectively, and in combination with the location of the intermediate joint of the power cable line, assess whether the defect is located in the power cable body or the joint.

[0072] S4: Assess the type and severity of defects in power cables.

[0073] In an optional embodiment, before S1, the following steps are further included:

[0074] S01: First, determine the bandwidth BWF of the frequency enhancement pulse sequence, the single frequency enhancement amplitude Δf, and the initial frequency f0 by using the length parameters, cross-sectional geometric parameters, and preset positioning accuracy of the power cable under test.

[0075] S02: Based on S01, generate frequency-enhancing pulses.

[0076] In an optional embodiment, S01 includes:

[0077] The formula for calculating the bandwidth (BWF) of an enhanced pulse sequence is:

[0078]

[0079] Where c is the speed of light, A is the positioning accuracy, i.e., the ratio of the positioning error to the length of the cable being measured (L), and ε is the position of the cable being measured. t The comprehensive equivalent relative permittivity of the dielectric material of the power cable under test;

[0080] ε t The method for determining this depends on the cross-sectional geometric parameters and material parameters of the power cable being tested.

[0081]

[0082] Where, r i Let ε be the radius of the i-th layer in the cross-section of the power cable. i Let be the relative permittivity of the i-th layer material;

[0083] The formula for calculating the single-cycle frequency increase amplitude Δf is:

[0084]

[0085] The formula for calculating the initial frequency f0 is:

[0086] f0 = 2Δf.

[0087] In an optional embodiment, S02 includes: the host computer sending a control command to the controller, the controller communicating with the programmable waveform generator to generate a frequency-increasing pulse, the expression of which is:

[0088]

[0089] Where ZP(t) is the frequency-enhanced pulse signal at time t, and θ i For the initial phase, f i Let f be the frequency of the i-th boost pulse. i =f0 + i × Δf, i = 0, 1, 2, ..., 1 / A-1; T is the interval time of the frequency-increasing pulse. T is greater than the time it takes for the frequency-increasing pulse to return once in the power cable, and can be calculated as T = 10 × L / (1.6 × 10 8 (Settings). The generated frequency-enhancing pulses are as follows: Figure 3 As shown.

[0090] For example, for a typical 10kV distribution cable with a total length of 1km, the required defect location accuracy is 0.5% of the total cable length, i.e., A = 0.005.

[0091] See Figure 2 Typical values ​​are r1 = 11.5 mm, r2 = 12.7 mm, r3 = 17.2 mm, and r4 = 19.1 mm. The relative permittivity of the inner semiconducting layer is ε1 = 200, the relative permittivity of the XLPE insulating layer is ε2 = 2.5, and the relative permittivity of the outer semiconducting layer is ε3 = 200. Therefore, the applied comprehensive equivalent relative permittivity ε... t It is 4.14.

[0092] The corresponding determined frequency upsampling pulse sequence bandwidth (BWF) is 14MHz; the single upsampling amplitude (Δf) is 73.65kHz.

[0093] In an optional embodiment, S1 includes:

[0094] The frequency-enhanced pulse generated by the programmable waveform generator is amplified by a V-fold through a wideband power amplifier. The amplified frequency-enhanced pulse signal is then input to a power divider. One of the two outputs of the power divider is directly connected to the data sampling card, and the other is connected to a wideband directional coupler. The frequency-enhanced pulse is output to an impedance adapter through the wideband directional coupler. The output impedance of the impedance adapter is adjustable, integrating common power cable impedances from distribution to transmission levels for selection and adjustment. The output impedance of the impedance adapter is adjusted to be as close as possible to the characteristic impedance of the power cable under test to reduce the loss of the frequency-enhanced pulse signal injected into the power cable. The frequency-enhanced pulse propagates in the power cable, passing through defects and the end of the power cable. During the test, the end is suspended, and impedance mismatch generates a reflected wave. This reflected wave is transmitted through the power cable to the impedance adapter, and then through the wideband directional coupler to the data acquisition card, where the reflected signal is acquired.

[0095] In an optional embodiment, in S2:

[0096] The time difference between the frequency-enhanced pulse and the reflected signal is calculated using the energy EC method, i.e.:

[0097] The energy curve formulas for the frequency-enhanced pulse and the reflected signal are as follows:

[0098]

[0099] Among them, P ZP and P ZPR Let be the average power of the frequency-enhanced pulse and the reflected signal, respectively, and N be the total number of data points; the time difference t is the difference between the global minimum of the differentials of the frequency-enhanced pulse and the reflected signal. d .

[0100] In an optional embodiment, in S2:

[0101] Pearson cross-correlation coefficients were calculated between the extracted amplitude-frequency curve, phase-frequency curve, and power spectral density curve and the amplitude-frequency curve, phase-frequency curve, and power spectral density curve of the defect under the action of the frequency-enhanced pulse, respectively, to obtain the amplitude-frequency curve correlation coefficient ρ. af The correlation coefficient ρ of the phase frequency curve pf Correlation coefficient ρ with power spectral density pd ;

[0102] For the time-spectrum graph, it is first converted into a grayscale image, and then the normalized cross-correlation coefficient is applied. The correlation coefficient ρ of the time-spectrum graph is... tf The calculation formula is:

[0103]

[0104] Where g and h are the dimensions of the spectral image (g×h), TF is the time-spectrum diagram of the test frequency-enhanced pulse signal, and M is the time-spectrum diagram of the defect's frequency-enhanced pulse signal. To test the grayscale mean of the time-spectrum of the frequency-increasing pulse signal, The average grayscale value of the spectrum of the frequency-enhanced pulse signal for the defect;

[0105] The formula for calculating the generalized time-frequency dimension is:

[0106]

[0107] T(j)=log(ε j )

[0108]

[0109] Where J is the number of data points in the time-frequency grid, S(j) is used to describe the generalized entropy, N is the number of grids, and P... i Let q be the probability that the i-th grid covers the time-frequency information set, and q be the scaling exponent. When q = 0, D q (ε j Let q be the capacity dimension, and when q = 1, D q (ε j Let q = 2, then D is the information dimension. q (ε j ) represents the cross-dimensionality of the association.

[0110] In an optional embodiment, S3 includes:

[0111] Calculate the location of the defect in the power cable in the time domain:

[0112]

[0113] Among them, lt Where is the distance from the defect to the test end, and v is the velocity of the frequency-increasing pulse wave;

[0114] Calculate the location of the defect in the power cable in the frequency domain:

[0115] Performing an inverse discrete Fourier transform (IFT) on the reflected signal, and replacing the frequency in the IFT result with distance information based on the phase difference between the frequency-enhanced pulse and the reflected signal, yields the amplitude-distance spectrum, as shown in Figures 5(a) and 5(b). The distance at the peak amplitude is the location l of the defect. t ;

[0116] The defect distance l calculated in the time and frequency domains t By comparing and considering the location of the intermediate joints in the power cable line, an assessment is made to determine whether the defect is in the power cable itself or at the joint; if l t The defect is not located at the joint of the power cable, meaning it is within the power cable itself; if l t If the location is a joint in the power cable, assess whether there are any defects in the joint.

[0117] In an optional embodiment, S4 includes:

[0118] Based on the defect location results, if the defect is located within the power cable itself, the power cable defect feature fingerprint database is invoked. A neural network model compares the feature parameters with the database to determine the defect type and assess its severity. The power cable defect feature fingerprint database stores the feature parameters of typical power cable defects under frequency-enhanced pulse testing. These parameters are extracted during frequency-enhanced pulse testing of different types and severity of defects. Typical defects include insulation aging, structural deformation, and mechanical damage.

[0119] If the defect is at a power cable joint, the power cable joint defect feature fingerprint database is invoked. A neural network model compares the feature parameters with the database to determine the defect type and assess its severity. The feature parameters used include the pulse polarity of the reflected signal, amplitude attenuation coefficient, amplitude-frequency curve correlation coefficient, phase-frequency curve correlation coefficient, power spectral density correlation coefficient, time-frequency spectrum correlation coefficient, time-frequency capacity dimension, time-frequency information dimension, and time-frequency correlation dimension. The power cable joint defect feature fingerprint database stores the feature parameters of typical power cable joint defects under the action of frequency-enhanced pulses. Typical joint defects include: water immersion and structural damage.

[0120] See Figure 6This invention also provides a power cable insulation detection and defect assessment system based on frequency-increased pulses, implementing any of the aforementioned methods for power cable insulation detection and defect assessment based on frequency-increased pulses. The system includes a frequency-increased pulse generation module, a signal separation and injection module, a data processing module, a host computer, a controller, a data acquisition card, and a human-machine interface. The frequency-increased pulse generation module includes a programmable waveform generator and a wideband power amplifier connected in sequence. The signal separation and injection module includes a power divider, a wideband directional coupler, and a bidirectional impedance adapter connected in sequence. The bidirectional impedance adapter is connected to the cable under test. The host computer is connected to the programmable waveform generator via the controller. The wideband power amplifier is connected to the power divider. The host computer is connected to the wideband directional coupler via the data acquisition card. The data processing module analyzes the frequency-increased pulse signal and reflected signal to obtain the location of defects in the power cable and assess the type and severity of the defects. Users can operate the host computer through the human-machine interface.

[0121] For ease of understanding, the present invention provides a more specific embodiment:

[0122] The application of frequency-enhanced pulse sequence-based defect detection and location technology was carried out on power distribution cables. The tested power cable was model YJV8.7 / 10kV-3*240, with a length of 100m. Thermal aging was performed on the insulation at a 20-meter mark using a semi-conductive heating tape at approximately 200℃ for 5 hours. The results obtained from the frequency-enhanced pulse sequence test on this power distribution cable are as follows: Figure 7 As shown.

[0123] right Figure 7 The amplitude and phase of the acquired time-domain waveform signal are extracted to obtain the following time-frequency information: Figure 4 As shown in Figures 5(a) and 5(b), the location is 20m, which coincides with the actual location of the defect in the power cable.

[0124] 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 scope of protection of the claims of the present invention.

Claims

1. A method for detecting and evaluating defects in power cables based on frequency-enhanced pulses, characterized in that, include: Step 1: Obtain the reflected signal generated by the propagation of the frequency-enhanced pulse in the power cable; Step 2: Extract time-domain information from the frequency-enhanced pulse and the reflected signal to obtain the time difference between them, the pulse polarity of the reflected signal, and the amplitude attenuation coefficient; extract frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the correlation coefficient of the signal amplitude-frequency curve, the correlation coefficient of the phase-frequency curve, and the correlation coefficient of the power spectral density curve; extract time-frequency-domain information from the frequency-enhanced pulse and the reflected signal, including the correlation coefficient of the time-spectrum diagram and the generalized time-frequency dimension. Step 3: Calculate the location of the defect in the power cable in the time and frequency domains respectively, and in combination with the location of the intermediate joint of the power cable line, assess whether the defect is located in the power cable body or the joint. Step 4: Assess the type and severity of defects in the power cables; Before step 1, the following are also included: Step 01: First, determine the bandwidth (BWF) of the frequency-increasing pulse sequence and the single-pulse amplitude using the relevant parameters of the power cable under test and the preset positioning accuracy. and initial frequency The relevant parameters include length, cross-sectional geometry parameters, and material parameters; the cross-sectional geometry parameters include radius, and the material parameters include the relative permittivity of the material. Step 02: Based on step 01, generate an enhanced frequency pulse; Step 01 includes: The formula for calculating the bandwidth (BWF) of an enhanced pulse sequence is: Where c is the speed of light, A is the positioning accuracy (the ratio of positioning error to the length of the cable being measured), and L is the length of the power cable being measured. The comprehensive equivalent relative permittivity of the dielectric material of the power cable under test; The method for determining this depends on the cross-sectional geometric parameters and material parameters of the power cable being tested. in, Let be the radius of the i-th layer in the cross-section of the power cable. is the relative permittivity of the i-th layer material; m is the total number of layers in the power cable from conductor to metal shield; Single frequency increase amplitude The calculation formula is: initial frequency The calculation formula is: =2 ； Step 02 includes: the host computer sending a control command to the controller, the controller communicating with the programmable waveform generator to generate frequency-increasing pulses, the expression of which is: in, The frequency-incrementing pulse signal at time t, For the initial phase, Let be the frequency of the i-th boost pulse, and let its value be . , i=0,1,2,……1 / A-1; T is the interval time of the frequency enhancement pulse, which is greater than the time it takes for the frequency enhancement pulse to return once in the power cable; In step 2: Pearson cross-correlation coefficients were calculated between the extracted amplitude-frequency curves, phase-frequency curves, and power spectral density curves and the amplitude-frequency curves, phase-frequency curves, and power spectral density curves of the defect under the action of the frequency-enhanced pulse, respectively, to obtain the amplitude-frequency curve correlation coefficient. Phase frequency curve correlation coefficient Correlation coefficient with power spectral density ; For the time-spectrum image, first convert it to grayscale. Then, calculate the correlation coefficient between the grayscale time-spectrum image and the time-spectrum images of typical defects in the power cable body defect feature fingerprint database and the power cable joint defect feature fingerprint database using the following formula: (i.e., the time-spectrum image correlation coefficient). The calculation formula is: Where g and h are the dimensions of the spectral image (g×h), TF is the time-spectrum diagram of the test frequency-enhanced pulse signal, and M is the time-spectrum diagram of the defect's frequency-enhanced pulse signal. To test the grayscale mean of the time-spectrum of the frequency-increasing pulse signal, The average grayscale value of the spectrum of the frequency-enhanced pulse signal for the defect; The formula for calculating the generalized time-frequency dimension is: Where J is the number of data points in the time-frequency grid, S(j) is used to describe the generalized entropy, and N is the number of grids. Let be the probability that the i-th grid covers the time-frequency information set, and q be the scaling exponent. When q=0, Let q be the capacity dimension, and q=1. For the information dimension, when q=2, For the dimensions of the related cross-sections; Step 4 includes: Based on the defect location results, if the defect is located in the power cable itself, the defect feature fingerprint database of the power cable itself is called. Through a neural network model, the feature parameters are compared with the defect feature fingerprint database of the power cable itself to determine the defect type and assess the severity. The feature parameters include the pulse polarity of the reflected signal, amplitude attenuation coefficient, amplitude-frequency curve correlation coefficient, phase-frequency curve correlation coefficient, power spectral density correlation coefficient, time-frequency spectrum correlation coefficient, time-frequency capacity dimension, time-frequency information dimension, and time-frequency correlation dimension. If the defect is at a joint of the power cable, the power cable joint defect feature fingerprint database is called up. The feature parameters are compared with the power cable joint defect feature fingerprint database through a neural network model to determine the defect type and assess the severity. Among them, the power cable body defect feature fingerprint database is used to store the feature parameters of typical power cable body defects under the action of frequency-increased pulses, and the power cable joint defect feature fingerprint database is used to store the feature parameters of typical power cable joint defects under the action of frequency-increased pulses.

2. The method for detecting and evaluating power cable defects based on frequency-enhanced pulses according to claim 1, characterized in that, Step 1 includes: The frequency-enhanced pulse generated by the programmable waveform generator is amplified by a V-fold through a wideband power amplifier. The amplified frequency-enhanced pulse signal is then input to a power divider. One of the two outputs of the power divider is directly connected to the data sampling card, and the other is connected to a wideband directional coupler. The frequency-enhanced pulse is output to an impedance adapter through the wideband directional coupler. The output impedance of the impedance adapter is adjusted to be as close as possible to the characteristic impedance of the power cable under test to reduce the loss of the frequency-enhanced pulse signal injected into the power cable. The frequency-enhanced pulse propagates in the power cable, passing through the defect and the end of the power cable. During the test, the end is suspended, and impedance mismatch generates a reflected wave. This reflected wave is transmitted through the power cable to the impedance adapter, and then through the wideband directional coupler to the data acquisition card, where the reflected signal is acquired.

3. The method for detecting and evaluating power cable defects based on frequency-enhanced pulses according to claim 1, characterized in that, In step 2: The time difference between the frequency-enhanced pulse and the reflected signal is calculated using the energy EC method, i.e.: The energy curve formulas for the frequency-enhanced pulse and the reflected signal are as follows: in, and Let N be the average power of the frequency-enhanced pulse and the reflected signal, respectively, and N be the total number of data points. The time difference between the global minimum of the differential values ​​of the energy curves of the frequency-enhanced pulse and the reflected signal is the time difference between them. .

4. The method for detecting and evaluating power cable defects based on frequency-enhanced pulses according to claim 1, characterized in that, Step 3 includes: Calculate the location of the defect in the power cable in the time domain: in, Where is the distance from the defect to the test end, and v is the velocity of the frequency-increasing pulse. The time difference between the frequency-increasing pulse and the reflected signal; Calculate the location of the defect in the power cable in the frequency domain: Performing an inverse discrete Fourier transform (IFT) on the reflected signal, and replacing the frequency in the IFT result with distance based on the phase difference between the frequency-enhanced pulse and the reflected signal, yields an amplitude-distance spectrum. The distance at the amplitude peak indicates the location of the defect. ; Defect distance calculated in both the time and frequency domains By comparing and considering the location of the intermediate joints in the power cable line, an assessment is made to determine whether the defect lies in the power cable itself or in the joint; if The defect is not at the joint of the power cable, meaning it is within the power cable itself; if If the location is a joint in the power cable, assess whether there are any defects in the joint.

5. A power cable defect detection and evaluation system based on frequency-increased pulses, implementing the power cable defect detection and evaluation method based on frequency-increased pulses as described in any one of claims 1-4, characterized in that, The system includes a frequency-increased pulse generation module, a signal separation and injection module, a data processing module, a host computer, a controller, and a data acquisition card. The frequency-increased pulse generation module includes a programmable waveform generator and a wideband power amplifier connected in sequence. The signal separation and injection module includes a power divider, a wideband directional coupler, and a bidirectional impedance adapter connected in sequence. The bidirectional impedance adapter is connected to the cable under test. The host computer is connected to the programmable waveform generator through the controller. The wideband power amplifier is connected to the power divider. The host computer is connected to the wideband directional coupler through the data acquisition card. The data processing module is used to perform data analysis on the frequency-increased pulse signal and the reflected signal to obtain the location of defects in the power cable and to assess the type and severity of the defects.

Images