Method for improving time-domain recovered waveform characteristics of cable defects based on low-frequency stitching

Abstract

The application discloses a method for improving the time domain recovery waveform features of cable defects based on low-frequency stitching, and adopts a frequency domain reflection method to test the head end reflection coefficient Γ f (ω) of a power cable s (ω) of the power cable; extracts the cable head end reflection coefficient Γ s (ω) in a low-frequency range from the obtained head end reflection coefficient spectrum Γ low (ω); performs frequency domain stitching on the cable head end reflection coefficient Γ low (ω) in the low-frequency range and the head end reflection coefficient Γ f (ω) and expands the same to obtain a second head end reflection coefficient spectrum Γ(ω) in a full-frequency range; and obtains the characteristic time domain diagnostic waveform of the power cable based on the full-frequency head end reflection coefficient and a virtual incident waveform; the above method can effectively suppress the baseline oscillation of the waveform, and further realizes accurate diagnosis of the power cable defects, including polarity judgment and type identification.

Description

Technical Field

[0001] The present invention belongs to power cable defect diagnosis technology, and relates to power cable defect time domain recovery waveform polarity judgment and defect type identification diagnosis technology. More specifically, it relates to a method for improving the time-frequency domain recovery waveform characteristics of cable defects based on low-frequency stitching and its application in cable defect diagnosis. Background Art

[0002] Power cables are increasingly being used due to their attractive aesthetics and minimal floor space. However, due to poor manufacturing processes, irregular installation, and direct burial in humid underground cable trenches, long-term electrical, thermal, and mechanical stress can lead to localized moisture, corrosion, and bending in the cables, resulting in localized insulation degradation and cable defects. If cable defects are not detected promptly and accurately, the electric field forces can cause cable failures, impacting the safe and stable operation of the power system and leading to widespread unplanned power outages. Furthermore, cable replacement can result in significant economic losses.

[0003] Local defects in the cable, such as partial damage or damp joints, can cause changes in the local physical parameters of the power cable, which in turn changes the local characteristic impedance and creates impedance discontinuities. When high-frequency electromagnetic waves propagate through the cable and encounter impedance discontinuities, they generate reflected waves carrying the characteristic information of the defect. Therefore, the propagation distance of the reflected wave can be used to locate cable defects, and the shape of the reflected wave can be used to determine the polarity and type of the defect. For example, local damage will increase the impedance, resulting in a positive reflected wave, while damp joints will decrease the impedance, resulting in a negative reflected wave.

[0004] Time domain reflectometry (TDR) is a classic method for locating and diagnosing cable faults. Patent application number CN201920136039.4 discloses a cable fault location device based on time domain pulse reflectometry. This device determines the amplitude of a transmitted pulse signal based on the characteristics of the cable being tested, and identifies the fault type based on the amplitude and phase relationship between the reflected and transmitted pulse signals. While this device can locate cable faults and identify their type, it cannot locate subtle defects in the cable or determine the impedance changes associated with these defects. Frequency domain reflectometry (FDR) can locate cable defects by analyzing the reflection signal characteristics generated by the linear sweep signal injected at the impedance discontinuity point at the cable head end. Patent application document CN201911125092.5 discloses a method for locating and diagnosing moisture in the intermediate joints of power cables based on input impedance spectrum. First, the input impedance spectrum of the cable head end is tested, and then the real or imaginary part of the input impedance spectrum is processed by discrete Fourier transform to obtain the positioning spectrum function of the cable under test, thereby locating the cable defects. Afterwards, the input impedance spectrum of the normal cable is discrete Fourier transformed through simulation and testing, and the positioning spectrum function of the cable under test is compared with the spectrum function of the normal cable, thereby diagnosing the degree of moisture in the intermediate joints of the cable under test. Although this method can locate and diagnose moisture in the cable intermediate joints, it is difficult to identify changes in the cable characteristic impedance caused by other defects. At the same time, the original parameters of the cable under test are also required, which are often difficult to obtain on site.

[0005] In fact, the characteristic impedance changes of power cable defects of different types and degrees are different, which leads to the incident signal pulse generating cable time domain pulse waveforms of different shapes at the local defects of the cable. Therefore, a virtual incident signal can be input at the incident end of the cable, and the time domain pulse waveform at the cable defect can be obtained by convolving it with the zero-state response of the reflection coefficient at the head end of the power cable. The characteristic impedance changes of the local defects of the power cable are judged based on the time domain pulse waveform, and the different defect types of the power cable are preliminarily identified. However, this method is greatly affected by the lack of low-frequency band data of the reflection coefficient at the head end of the power cable. The greater the proportion of defect data in the reflection coefficient frequency band data, the more severe the baseline oscillation of the obtained power cable time domain recovery waveform; such as Figure 1 As shown, when the lower limit frequency f low When the frequency is higher, the proportion of missing data in the low-frequency band increases, the baseline oscillation of the time domain recovery waveform becomes more serious, and it is difficult to identify the time domain signal at the defect; when the lower limit frequency is lower, the baseline oscillation is effectively suppressed, and the time domain waveform characteristics at the defect are improved.

[0006] Therefore, the ratio of missing low-frequency data to the reflection coefficient data within the effective cable test frequency range causes baseline oscillation in the time-domain recovery waveform of cable defects, drowning out the time-domain waveform characteristics of the power cable defect and making it difficult to determine the polarity of the power cable defect. When using FDR-based power cable time-domain characteristic waveforms to perform impedance transformation diagnosis of local power cable defects, how to stitch or extend the missing low-frequency bands of the power cable reflection coefficient spectrum to improve the accuracy of defect diagnosis is one of the key issues that need to be addressed in the field of cable defect diagnosis. Summary of the Invention

[0007] The purpose of the present invention is to address the deficiencies in the prior art and provide a method for improving the time domain recovery waveform characteristics of cable defects based on low-frequency stitching, which helps to improve the identification of impedance changes of different types of cable defects.

[0008] The inventive idea of ​​the present invention is: first, the head-end reflection coefficient of the full frequency domain segment is obtained based on the frequency domain reflection method, and then the head-end reflection coefficient spectrum is obtained based on the time domain reflection method, its low-frequency band is extracted and stitched to the head-end reflection coefficient obtained based on the frequency domain reflection method, and then the characteristic time domain recovery waveform with suppressed baseline oscillation is obtained based on the head-end reflection coefficient of the stitched low-frequency band, and it is used as the characteristic time domain diagnostic waveform.

[0009] Based on the above-mentioned inventive ideas, the method provided by the present invention for recovering waveform characteristics of cable defects in the time domain based on low-frequency stitching includes the following steps:

[0010] S1 uses the frequency domain reflection method to test the reflection coefficient Г of the power cable head end f (ω);

[0011] S2 uses time domain reflection method to test the reflection coefficient spectrum of the power cable head end s (ω);

[0012] S3 Extract the cable head end reflection coefficient Γ in the low frequency range from the head end reflection coefficient spectrum obtained in step S2 low (ω);

[0013] S4 is the reflection coefficient of the cable head end in the low frequency range Г low (ω) and the head-end reflection coefficient Γ obtained in step S1 f (ω) frequency domain stitching and expansion are performed to obtain the head-end reflection coefficient spectrum Γ(ω) in the full frequency range;

[0014] S5 obtains the characteristic time domain diagnostic waveform of the test power cable according to the following formula:

[0015] (1);

[0016] Where, ; FFT (·) is fast Fourier transform; IFFT (·) is inverse Fourier transform.

[0017] In the above step S1, the frequency domain reflection method test system is composed of a broadband reflection coefficient spectrum tester, a control computer and the cable under test. The broadband frequency domain reflection coefficient spectrum tester is used to inject the lower limit frequency f into the power cable under test. low , the upper limit frequency is f up The frequency sweep signal is obtained by controlling the computer low -f up Head-end reflection coefficient Г within the frequency band f (ω), where angular frequency ω=2πf, and f is the signal test frequency.

[0018] In the above step S2, the time domain reflectometry test system consists of a signal generator, an oscilloscope and the cable under test. First, the signal generator and oscilloscope are used to test the time domain incident signal y of the power cable. si (t) and the time domain reflection signal y sr (t); Then the time domain incident signal and the time domain reflected signal are normalized to obtain the normalized cable incident signal y si and the reflected signal y sr

[0019] (2);

[0020] (3);

[0021] The incident signal y is transformed by fast Fourier transform si Processing is performed to obtain the spectrum Y of the time domain incident signal si (ω), that is, Y si (ω)=FFT(y si );For the time domain reflection signal y sr Perform fast Fourier transform to obtain the spectrum Y of the time domain reflection signal sr (ω), that is, Y sr (ω)=FFT(y sr ). Calculate the spectrum Y of the time domain reflection signal sr (ω) and the spectrum Y of the time domain incident signal sr (ω) ratio, and the reflection coefficient Г of the cable head end obtained by time domain reflection method s (ω), the calculation formula is as follows:

[0022] (4).

[0023] The above-mentioned time domain incident signal is a unipolar Gaussian signal.

[0024] In the above step S3, since the head-end reflection coefficient spectrum measured based on the time domain reflectometry is full-band, it contains the reflection information of the low-frequency band of the tested cable; therefore, the reflection coefficient of the low-frequency band can be extracted and stitched into Γ f (ω). In the present invention, Г is extracted by a rectangular window. s (ω) in the low frequency band 0~f low The head-end reflection coefficient value Г within the range low (ω), and then use the interpolation method (such as cubic spline interpolation algorithm) to interpolate the low frequency band Г low (ω) data is interpolated to make Г low Frequency resolution △f of (ω) s and G f The frequency resolution △f of (ω) remains consistent.

[0025] The purpose of the above step S4 is to expand the head-end reflection coefficient to the full frequency domain, so as to obtain the time domain waveform of the test power cable. First, for the frequency band 0~f up The reflection coefficient spectrum of step S3 is low (ω) and Г obtained in step S1 f (ω) is spliced ​​to achieve the low frequency band 0~f low And the effective test frequency band f low ~f up Frequency domain stitching of the head-end reflection coefficient; at the same time, for up The reflection coefficient of the frequency band is set to 0 to obtain the head-end reflection coefficient spectrum Γ(ω) of the full frequency band.

[0026] The purpose of the above step S5 is to obtain the characteristic time domain diagnostic waveform of the test power cable. First, the virtual incident waveform y is introduced. i (t), since the Fourier spectrum calculation of the Gaussian pulse signal does not produce the Gibbs effect; at the same time, in order to make the best use of the FDR test results and have a higher distance resolution, the pulse width of the pulse signal is selected as the reciprocal value of the upper limit of the scanning frequency of the broadband reflection coefficient spectrum tester; in the preferred implementation, y i (t) Select a Gaussian pulse signal with slow rising and falling edges. Then, calculate the characteristic recovery waveform of the test power cable according to the above formula (1). The characteristic recovery waveform at this time is the characteristic time domain diagnostic waveform.

[0027] In the virtual incident waveform y i After the fast Fourier transform of (t), considering that its frequency resolution is different from that of Γ(ω), an interpolation method (such as cubic spline interpolation) is further used to make the virtual incident waveform y i Frequency resolution △f of (t) y Keep consistent with the frequency resolution △f of Г(ω).

[0028] By comparing the characteristic time domain diagnostic waveform of the power cable with the characteristic time domain recovery waveform directly recovered for low-frequency band processing, it can be found that the baseline oscillation of the reflection coefficient is effectively suppressed after low-frequency stitching, making the time domain characteristics of the local defects of the cable more obvious.

[0029] The present invention further provides the use of the characteristic time-domain diagnostic waveform of the power cable obtained by the above method, which can be applied to the diagnosis of power cable defects, including polarity judgment and type identification. According to the polarity judgment and defect type identification model of the power cable, the polarity judgment and type identification of the power cable defect are performed: the characteristic time-domain diagnostic waveform is mapped to the test cable. If the shape of the time-domain diagnostic waveform at the power cable defect is peak-up and basically consistent with the introduced virtual incident waveform, the defect polarity is positive and the defect type is "open circuit"; if the shape of the time-domain diagnostic waveform at the power cable defect is peak-down and basically consistent with the introduced virtual incident waveform, the defect polarity is negative and the defect type is "grounding" or "short circuit"; if the shape of the time-domain diagnostic waveform at the power cable defect is a peak-to-"first up and then down" feature, the defect polarity is positive and the defect type is "impedance increase"; if the shape of the time-domain diagnostic waveform at the power cable defect is a peak-to-"first down and then up" feature, the defect polarity is negative and the defect type is "impedance decrease".

[0030] Compared with the prior art, the method provided by the present invention for improving the time domain waveform characteristics of cable defects based on low-frequency stitching has the following beneficial effects:

[0031] 1. The present invention first adopts the frequency domain method to obtain the reflection coefficient of the head end of the power cable, and then uses the time domain method to obtain the reflection coefficient of the head end of the power cable. The low-frequency band of the reflection coefficient of the head end of the power cable obtained by the time domain method is then stitched to the frequency domain method to obtain the reflection coefficient of the head end of the power cable, and then expanded to the full frequency band. Finally, based on the full-band head end reflection coefficient and the introduced virtual incident waveform, the characteristic time domain diagnostic waveform of the power cable is obtained; the above method can achieve effective suppression of waveform baseline oscillation.

[0032] 2. The method provided by the present invention for restoring the waveform of cable defects in the time domain based on low-frequency stitching has the advantages of high sensitivity, good intuitiveness, simple operation and low cost.

[0033] 3. Based on the characteristic time-domain diagnostic waveform of the power cable obtained by the present invention, accurate diagnosis of power cable defects can be achieved, including polarity judgment and type identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This figure shows the impact of missing data in the low-frequency band on the time-domain restored waveform of cable characteristics.

[0035] Figure 2This is a flow chart of the low-frequency stitching method for improving the time-domain recovery of waveform characteristics of cable defects according to the present invention.

[0036] Figure 3 This is a schematic diagram of the frequency domain reflection test system of the present invention.

[0037] Figure 4 This is a schematic diagram of the time domain reflection test system of the present invention.

[0038] Figure 5 The reflection coefficient spectrum amplitude abs(Г obtained by the frequency domain reflection method test system of the present invention is the cable head end. f (ω)).

[0039] Figure 6 The reflection coefficient spectrum amplitude abs(Г obtained by the time domain reflection method test system of the present invention is the cable head end. s (ω)).

[0040] Figure 7 The present invention adopts low frequency stitching technology to obtain 0-f up The reflection coefficient spectrum amplitude abs(Г(ω)) at the cable head end within the frequency band.

[0041] Figure 8 It is the time-domain recovery waveform y2(t) of the cable defect characteristics after low-frequency stitching and the time-domain recovery waveform y1(t) of the cable characteristics before stitching. DETAILED DESCRIPTION

[0042] The following is a clear and complete description of the technical solutions of the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments derived by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts belong to the present invention.

[0043] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

[0044] Example

[0045] The research object of this embodiment is a 500m YJLV228.7 / 15-3×25XLPE power cable. The effective frequency band of the injected high-frequency signal is: 0.15MHz-6MHz, and the number of sampling points is 1601.

[0046] like Figure 2 As shown, this embodiment obtains the characteristic time-domain recovery waveform of the power cable through the following steps:

[0047] S1 uses the frequency domain reflection method to test the reflection coefficient Г of the power cable head end f (ω).

[0048] The principle of frequency domain reflectometry is as follows Figure 3 As shown in the figure, the test system used is composed of a broadband reflection coefficient spectrum tester, a control computer and the cable under test. The frequency domain reflection coefficient spectrometer is used to inject the lower limit frequency f into the YJLV228.7 / 15-3×25XLPE power cable. low is 0.15MHz, the upper limit frequency f up The frequency sweep signal is 6MHz and 5V, and the f low -f up Head-end reflection coefficient Г within the frequency band f (ω), the test results are as follows Figure 5 shown.

[0049] S2′ obtains the characteristic time domain recovery waveform of the test power cable; for this purpose, the virtual incident waveform is first introduced. Since the Fourier spectrum calculation of the Gaussian pulse signal does not produce the Gibbs effect; at the same time, in order to make the best use of the FDR test results and have a higher distance resolution, the pulse signal pulse width is selected as the upper limit of the scanning frequency f of the broadband reflection coefficient spectrum tester. max In the preferred embodiment, y i (t) Select a Gaussian pulse signal with slow rising and falling edges. i The Fourier transform of (t) is full frequency band, so in order to solve the characteristic time domain recovery waveform of the power cable, it is necessary to convert the cable head end reflection coefficient Г f (ω) is extended to the full frequency domain.

[0050] In this embodiment, the introduced virtual Gaussian incident waveform is , t is time, ω is the pulse width of the virtual incident Gaussian pulse waveform, which is taken as the reciprocal of the upper limit of the broadband reflection coefficient spectrometer scan 1 / f max .

[0051] Based on the above analysis, this step includes the following sub-steps:

[0052] S21′ performs fast Fourier transform on the virtual Gaussian incident waveform to obtain the spectrum Y of the virtual incident Gaussian signal i (ω).

[0053] (5);

[0054] S22′ uses the cubic spline interpolation method (see Li Qingyang, Wang Nengchao, Yi Dayi. Numerical Analysis [M]. Fourth Edition. Beijing: Tsinghua University Press), so that the virtual incident waveform y i Frequency resolution △f of (t) y With G f The frequency resolution Δf of (ω) (Δf is 3656.25 Hz in this embodiment) remains consistent.

[0055] S23′ is the reflection coefficient of the cable head end Г f (ω) is expanded to the full frequency band, and the reflection coefficient spectrum Г1(ω) of the full frequency band is obtained.

[0056] In this embodiment, 0~f low The reflection coefficient data of the frequency band is set to 1, which is greater than f up The reflection coefficient data of the frequency band range is set to 0, and the reflection coefficient spectrum Г1(ω) of the full frequency band range is obtained.

[0057] S24′ obtains the characteristic time domain recovery waveform of the test power cable according to formula (1):

[0058]

[0059] Where, , is the spectrum after interpolation processing in step S22′, FFT(·) represents fast Fourier transform, and IFFT(·) represents inverse Fourier transform.

[0060] In this embodiment, the characteristic time domain recovery waveform of the test power cable is mapped onto the power cable (that is, the time is multiplied by the wave velocity in the cable (here 1.62×10 8 m / s)), the result is as follows Figure 8 shown.

[0061] The method provided in this embodiment is based on low-frequency stitching to improve the time domain waveform characteristics of cable defects. Figure 2 As shown, the following steps are included:

[0062] S1 uses the frequency domain reflection method to test the reflection coefficient Г of the power cable head end f (ω).

[0063] The principle of frequency domain reflectometry is as follows Figure 3 As shown in the figure, the test system used is composed of a broadband reflection coefficient spectrum tester, a control computer and the cable under test. The frequency domain reflection coefficient spectrometer is used to inject the lower limit frequency f into the YJLV228.7 / 15-3×25XLPE power cable. low is 0.15MHz, the upper limit frequency f up The frequency sweep signal is 6MHz and 5V, and the f low-f up Head-end reflection coefficient Г within the frequency band f (ω), the test results are as follows Figure 5 shown.

[0064] S2 uses time domain reflection method to test the reflection coefficient spectrum of the power cable head end Г s (ω).

[0065] The principle of time domain reflectometry is as follows Figure 4 As shown in the figure, the test system used consists of a signal generator, an oscilloscope and the cable under test. The incident signal generated by the signal generator is a unipolar Gaussian signal (consistent with the virtual Gaussian incident waveform introduced earlier) with an amplitude of 1V. First, the time domain incident signal y of the power cable is tested using the signal generator and oscilloscope. si (t) and the time domain reflection signal y sr (t); Then the time domain incident signal and the time domain reflected signal are normalized according to the following formula to obtain the normalized cable incident signal y si and the reflected signal y sr :

[0066] (2);

[0067] (3);

[0068] The incident signal y is transformed by fast Fourier transform si Processing is performed to obtain the spectrum Y of the time domain incident signal si (ω), that is, Y si (ω)=FFT(y si );For the time domain reflection signal y sr Perform fast Fourier transform to obtain the spectrum Y of the time domain reflection signal sr (ω), that is, Y sr (ω)=FFT(y sr ). Calculate the spectrum Y of the time domain reflection signal sr (ω) and the spectrum Y of the time domain incident signal sr (ω) ratio, and the reflection coefficient Г of the cable head end obtained by time domain reflection method s (ω), the calculation formula is as follows:

[0069] (4);

[0070] The head-end reflection coefficient Г obtained by the above-mentioned time domain reflection method s (ω) Figure 6 As shown in the figure, it can be seen that the head end reflection coefficient Г s (ω) contains the low frequency part.

[0071] S3 Extract the cable head end reflection coefficient Γ in the low frequency range from the head end reflection coefficient spectrum obtained in step S2 low (ω).

[0072] Here, Г is extracted by a rectangular window s (ω) in the low frequency band 0~f low The head-end reflection coefficient value Г within the range low (ω), and then use the cubic spline interpolation algorithm to interpolate the low frequency band Г low (ω) data is interpolated to make Г low Frequency resolution △f of (ω) s and G f The frequency resolution Δf of (ω) (Δf is 3656.25 Hz in this embodiment) remains consistent.

[0073] S4 is the reflection coefficient of the cable head end in the low frequency range Г low (ω) and the head-end reflection coefficient Γ obtained in step S1 f (ω) is stitched and expanded in the frequency domain to obtain the head-end reflection coefficient spectrum Γ2(ω) in the full frequency band.

[0074] Here, Г low (ω) and Г f (ω) frequency domain splicing, and at the same time for those greater than f up The reflection coefficient of the frequency band is set to 0 to obtain the head-end reflection coefficient spectrum Γ2(ω) of the full frequency band.

[0075] The unstitched head-end reflection coefficient spectrum obtained in step S1 and the head-end reflection coefficient spectrum obtained in step S5 are in the range of 0~f up Range such as Figure 7 As shown. Figure 7 It can be seen that through the above steps, data expansion of the low-frequency band of power cables can be achieved.

[0076] S5 obtains the characteristic time domain diagnostic waveform of the test power cable according to formula (1):

[0077] ;

[0078] Where Y i (ω) is the same as in the previous step S24'.

[0079] The characteristic time domain diagnostic waveform obtained in this step is mapped onto the power cable (that is, the time coordinate is multiplied by the wave velocity in the cable (here 1.62×10 8 m / s)), the result is as follows Figure 8 shown.

[0080] By comparing the waveforms of y2(t) and y1(t), it can be found that the baseline oscillation of the reflection coefficient is effectively suppressed after low-frequency stitching, making the time domain characteristics of the local defects in the cable more obvious.

[0081] The characteristic time-domain diagnostic waveform of the power cable obtained by the method of this embodiment can be applied to power cable defect diagnosis, including polarity determination and type identification. The polarity judgment and type identification model of the power cable is used to judge the polarity and identify the defect of the power cable: the characteristic time domain diagnostic waveform is mapped onto the test cable. If the shape of the time domain diagnostic waveform at the defect position of the power cable is peak-up and basically consistent with the introduced virtual incident waveform (i.e., a Gaussian waveform in this embodiment) from the cable test end to the cable end, the polarity of the defect is positive and the defect type is "open circuit"; if the shape of the time domain diagnostic waveform at the defect position of the power cable is peak-down and basically consistent with the introduced virtual incident waveform (i.e., a Gaussian waveform in this embodiment), the polarity of the defect is negative and the defect type is "grounding" or "short circuit"; if the shape of the time domain diagnostic waveform at the defect position of the power cable has the characteristic of a peak direction of "first up and then down", the polarity of the defect is positive and the defect type is "impedance increase"; if the shape of the time domain diagnostic waveform at the defect position of the power cable has a peak direction of "first down and then up", the polarity of the defect is negative and the defect type is "impedance decrease".

[0082] The waveform of y2(t) shows that there is a defect in the power cable at 250 m, and the time domain waveform of the defect has a peak shape of "first up and then down". According to the diagnostic criteria for the polarity and type of power cable defects, it can be determined that the polarity of the defect is positive and the defect type is "impedance increase", which is consistent with the impedance change of a normal power cable joint at this location of the tested power cable.

Claims

1. A method for improving the time domain waveform characteristics of cable defects based on low-frequency stitching, characterized in that: The following steps are involved: S1 uses the frequency domain reflection method to test the reflection coefficient Г of the power cable head end f (ω); S2 uses time domain reflection method to test the reflection coefficient spectrum of the power cable head end s (ω); S3 Extract the cable head end reflection coefficient Γ in the low frequency range from the head end reflection coefficient spectrum obtained in step S2 low (ω); S4 is the reflection coefficient of the cable head end in the low frequency range Г low (ω) and the head-end reflection coefficient Γ obtained in step S1 f (ω) frequency domain stitching and expansion are performed to obtain the head-end reflection coefficient spectrum Γ(ω) in the full frequency range; S5 obtains the characteristic time domain diagnostic waveform of the test cable according to the following formula: （1）； Where, ; FFT (·) is the fast Fourier transform; IFFT (·) is the inverse Fourier transform, y i (t) is the introduction of virtual incident waveform.

2. The method for recovering waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 1 is characterized in that: In step S1, the frequency domain reflection method test system is composed of a broadband reflection coefficient spectrum tester, a control computer and a power cable under test; the broadband frequency domain reflection coefficient spectrum tester is used to inject a lower limit frequency f into the power cable under test. low , the upper limit frequency is f up The frequency sweep signal is obtained by controlling the computer low -f up Head-end reflection coefficient Г within the frequency band f (ω).

3. The method for recovering waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 1 is characterized in that: In step S2, the time domain reflectometry test system consists of a signal generator, an oscilloscope and a cable under test; first, the signal generator and the oscilloscope are used to test the time domain incident signal y of the power cable. si (t) and the time domain reflection signal y sr (t); Then the time domain incident signal and the time domain reflected signal are normalized to obtain the normalized cable incident signal y si and the reflected signal y sr ; （2）； （3）； The incident signal y is transformed by fast Fourier transform si Processing is performed to obtain the spectrum Y of the time domain incident signal si (ω), that is, Y si (ω)=FFT(y si );For the time domain reflection signal y sr Perform fast Fourier transform to obtain the spectrum Y of the time domain reflection signal sr (ω), that is, Y sr (ω)=FFT(y sr ) ; Calculate the spectrum Y of the time domain reflection signal sr (ω) and the spectrum Y of the time domain incident signal sr (ω) ratio, and the reflection coefficient Г of the cable head end obtained by time domain reflection method s (ω), the calculation formula is as follows: （4）。 4. The method for recovering waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 2 is characterized in that: In step S3, Г is extracted through a rectangular window. s (ω) in the low frequency band 0~f low The head end reflection coefficient value Г within the range low (ω), and then use the interpolation method to calculate the low frequency band Г low (ω) data is interpolated to make Г low Frequency resolution △f of (ω) s and G f The frequency resolution △f of (ω) remains consistent.

5. The method for restoring waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 2 is characterized in that: Step S4: for the low frequency band 0~f low , replace Г in step S3 low (ω) and Г obtained in step S1 f (ω) is spliced ​​to achieve the frequency domain stitching of the two head-end reflection coefficients; at the same time, for the frequency domain of the reflection coefficient greater than f up The reflection coefficient of the frequency band is set to 0 to obtain the head-end reflection coefficient spectrum Γ(ω) of the full frequency band.

6. The method for restoring waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 1 is characterized in that: In the above step S5, i (t) Select a Gaussian pulse signal with slow rising and falling edges.

7. The method for recovering waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 6 is characterized in that: In the virtual incident waveform y i (t) After fast Fourier transform, interpolation method is also used to make the virtual incident waveform y i Frequency resolution △f of (t) y Keep consistent with the frequency resolution △f of Г(ω).

8. The method for restoring waveform characteristics of cable defects in the time domain based on low-frequency stitching according to any one of claims 1 to 7, characterized in that: Applied to power cable defect diagnosis; power cable defect diagnosis includes polarity judgment and type identification.

9. The method for restoring waveform characteristics of cable defects in the time domain based on low-frequency stitching according to claim 8 is characterized in that: The characteristic time-domain diagnostic waveform is mapped onto the test cable. If the peak of the time-domain diagnostic waveform at the power cable defect is upward and basically consistent with the introduced virtual incident waveform, the defect polarity is positive and the defect type is "open circuit". If the peak of the time-domain diagnostic waveform at the power cable defect is downward and basically consistent with the introduced virtual incident waveform, the defect polarity is negative and the defect type is "ground" or "short circuit". If the peak of the time-domain diagnostic waveform at the power cable defect is "first up and then down", the defect polarity is positive and the defect type is "impedance increase". If the peak of the time-domain diagnostic waveform at the power cable defect is "first down and then up", the defect polarity is negative and the defect type is "impedance decrease".

Images