CN105842596B - A kind of high sensitivity power cable local defect diagnostic method - Google Patents

Abstract

The invention discloses a high-sensitivity power cable local defect diagnosis method. The method is based on the power cable reflection coefficient spectrum in the frequency domain, uses non-destructive test technology to measure and analyze the power cable reflection coefficient spectrum, and combines gate technology and modern digital The signal processing technology realizes the location of local defects in the power cable, and then compares the wave velocity of the tested cable obtained by the real or imaginary part of the converted reflection coefficient spectrum with the wave velocity of the new cable, and combines the converted reflection coefficient spectrum The difference between the amplitude and the corresponding model fitting curve can be used to evaluate the severity of local defects. Compared with the prior art, the present invention can not only realize localization and multi-defect identification of weakly changing local defects, but also evaluate the severity of local defects of the power cable.

Description

A High Sensitivity Diagnosis Method for Local Defects of Power Cables

technical field

The invention relates to defect diagnosis technology during power cable operation, and more specifically relates to a diagnosis method for local defects of power cables.

Background technique

Cross-linked polyethylene (XLPE) power cables are widely used in my country's electric power industry because of their reliable electrical and mechanical properties. In the early 1990s, my country carried out a large number of urban network transformation work, and a large number of cables began to be applied to urban power grids. However, due to the influence of the manufacturing process and long-term operating conditions, power cables often suffer from moisture, overheating, extrusion, excessive bending, etc., resulting in water trees, deformation of the insulation medium, loose copper shielding, etc. Local defects, if the local defects of the power cable are not dealt with, the local defects of the power cable will develop rapidly under the action of the strong electric field, and eventually lead to the failure of the power cable insulation, thus bringing a lot of work to the maintenance and replacement of the power cable. Result in a lot of manpower and financial waste. In addition, since the power cables are buried underground and concentrated in central urban areas, it is also a difficult job to replace the entire cable. From the perspective of cost saving, if the local defects of power cables can be diagnosed and repaired, a lot of manpower and financial resources can be saved.

The relatively mature Time Domain Reflectometry (TDR) can use pulse signals to diagnose power cable faults. The application number is 201310063871.3, and the Chinese invention patent named "Time Domain Reflectometry System and Method" provides a method using time domain The reflection method is a method and system for determining conductor abnormalities, but this method is still unable to diagnose defects with a weak degree of change, and the resolution of TDR in multi-defect identification and near-end defect identification is not high enough. Partial discharge diagnosis of power cables is a method to detect local defects of cables, but due to the small amplitude of partial discharge signals, serious attenuation, difficulty in signal separation and difficulty in multi-fault identification, it is difficult to realize partial discharge diagnosis in actual measurement. Defects are diagnosed. At the same time, the partial discharge off-line test experiment is often a destructive experiment, which itself will have a certain impact on the structure of the cable. The existing power cable detection and diagnosis technology still stays in the fault diagnosis, lacks the diagnosis of local defects with weak changes in cable structure parameters, and cannot troubleshoot the early defects of cables. Therefore, it is of great significance to develop new technologies for local defect diagnosis of cables .

Contents of the invention

The purpose of the present invention is to provide a high-sensitivity power cable local defect diagnosis method based on the power cable reflection coefficient spectrum, which can accurately locate and evaluate the local defects of the cable, and can also perform a good inspection of the middle joint of the power cable. distinguish. It has the advantages of advanced technology, easy operation, low cost and high sensitivity.

The high-sensitivity power cable local defect diagnosis method provided by the present invention comprises the following steps:

1. Data measurement of the cable to be tested

Transmit a low-voltage linear frequency modulation, pseudo-random or high-frequency narrow pulse signal V _i to the power cable under test, measure the reflected signal V _r reflected from the end of the cable, and pass Find the amplitude |Γ _d (f)| of the reflection coefficient spectrum of the cable and the real part Real(Γ _d (f)) of the reflection coefficient spectrum and the imaginary part Imag(Γ _d (f)) of the reflection coefficient spectrum, where f is the signal frequency;

2. Data processing of the cable to be tested

Using the conversion function f→t′, the measured frequency domain signal f is converted into a t′ domain signal, and then the converted Real(Γ _d (t′)) or Imag(Γ _d (t′)) is fast Fourier leaf (FFT) transform _or discrete Fourier ₍ DFT) transform, record the frequency point f ₀ ′ with the highest energy; Multiply with the kaiser (Kaiser) window function, and perform fast Fourier transform or discrete Fourier transform to obtain the preliminary processing data _KΓd ;

Do the discrete Fourier transform from 0 Hz to the frequency point f ₀ ′ on the obtained preliminary processing data KΓ _d , map the transformation result to the original distance diagnostic map D ₀ , and add a distance window to the original distance diagnostic map D ₀ , Obtain the distance diagnostic map D, wherein the mapping distance at 0 Hz is 0 meters, and the mapping distance at the frequency point f ₀ ′ is the length l of the power cable;

3. Reference cable data measurement and processing

Use the non-measured phase power cable in the three-phase system as the reference power cable, and then use the same method as step 1 to find the amplitude |Γ _d1 (f)| of the reflection coefficient spectrum of the reference power cable and the real part of the reflection coefficient spectrum Real (Γ _d1 (f)) and the imaginary part Imag (Γ _d1 (f)) of the reflection coefficient spectrum, and carry out the processing in the same way in step 2, obtain the distance diagnostic map E after processing;

4. Determination of the location of local defects in power cables

Comparing the distance diagnostic map D and the processed distance diagnostic map E, using the gate technology to eliminate the distortion points at the same position in the distance diagnostic map D and the processed distance diagnostic map E, to obtain the local defect distance diagnostic map F, from the local The distortion point found in the defect distance diagnostic map F is the local defect position of the power cable;

5. Judgment of local defects of power cables

Inverse discrete Fourier transform or inverse fast Fourier transform is performed on the obtained local defect distance diagnostic map to obtain the transformed t′-domain reflection coefficient spectrum curve, and use the conversion function t′→f to convert the t′-domain reflection coefficient spectrum Convert to Real(Γ _d2 (f)) or Imag(Γ _d2 (f)), and then find each maximum point or minimum value in Real(Γ _d2 (f)) or Imag(Γ _d2 (f)) The frequency f _k corresponding to the point, using the formula Find the average wave velocity of the power cable under test, and compare v _c with the wave velocity v ₀ of a new power cable of the same specification. If the difference between v _c and v ₀ is 1% or more, it means that there is a local defect in the test cable, where N is The number of maximum or minimum points in Real(Γ _d2 ) or Imag(Γ _d2 ).

The present invention can also take the following methods to further evaluate the severity of the local defect of the tested power cable:

Establishment of Mathematical Model of Spectrum Modulus of Cable Reflection Coefficient

abs(Γ _d2 (f))＝a·e ^b·f +c·e ^d·f

Use the established mathematical model to perform data fitting on the reflection coefficient spectrum measured by the measured power cable using the gate technology |Γ _d2 (f)|, and find the corresponding quasi-parameter at the higher frequency extreme point f _k Combined value, and the defect degree error calculated by the following formula for evaluation:

error＝std((|Γ _d2 (f _k )|-abs(Γ _d2 (f _k )))/abs(Γ _d2 (f _k )))×100%

When the error<5%, the partial defect of the cable is relatively light. At this time, the power cable under test may have a small amount of loose copper shielding layer or the insulation medium has defects such as knife marks and scratches; when 5%≤error≤10%, the power cable The degree of defect is moderate. At this time, the power cable under test may have defects such as damp insulation medium or large-area copper shielding layer damage. When error>10%, the cable defect is very serious. At this time, the power cable under test may have severe deformation of the insulation medium, etc. defect. The std in the formula is the standard deviation.

In the above-mentioned technical scheme of the present invention, the β value in the described kaiser window function should satisfy:

Where α is the side lobe attenuation multiple.

In the above-mentioned technical scheme of the present invention, the original distance diagnostic atlas D ₀ preferably adopts the following formula to add distance window processing:

Where s is the window length, and the value is not greater than the spatial resolution in the distance diagnostic atlas D ₀ , and D is the distance diagnostic atlas obtained after processing.

In the above-mentioned technical scheme of the present invention, described in the step 5 with the wave velocity _v of the new cable of specification can be by formula Obtained by calculation, where L and C are the inductance and capacitance per unit length of a new cable of the same specification; it can also be obtained by time domain reflection method.

Since the position of the middle joint of the power cable will also have distortion points during the measurement process, which will affect the judgment of the defect position, in order to eliminate the influence of the middle joint of the power cable on the judgment of the defect position, the present invention introduces a reference power cable for comparison in step 3 of the diagnosis method , that is, the non-measured phase power cable in the three-phase system is used as the reference power cable, and then the amplitude |Γ _d1 (f)| of the reflection coefficient spectrum of the reference power cable and the real part of the reflection coefficient spectrum Real(Γ _d1 (f)) and the imaginary part Imag(Γ _d1 (f)) of the reflection coefficient spectrum, and perform the same processing in step 2 to obtain the processed distance diagnostic map E.

Usually, each phase of the power cable will make an intermediate joint at the same position, in order to eliminate the impact of the intermediate joint on the judgment of the local defect position of the cable, in step 4 of the diagnostic method, the present invention adopts a comparison distance diagnosis map D and a distance diagnosis The method of map E uses the gate technology to eliminate the distortion points at the same position in the distance diagnostic map D as the distance diagnostic map E, and obtains the local defect distance diagnostic map F, the distortion points found from the local defect distance diagnostic map F, the distortion The point is the local defect location.

In the local defect diagnosis of power cables, there are long power cables, severe signal attenuation, and a narrow signal frequency band required; short power cables, small signal attenuation, and a wide signal frequency band required. At the same time, when the signal is not attenuated to zero, the wider the frequency band of the signal, the higher the positioning accuracy and sensitivity.

The method for diagnosing local defects of power cables provided by the present invention has the following outstanding advantages and beneficial technical effects:

1. The defect diagnosis method of the present invention adopts a low-voltage measurement method instead of a destructive test measurement method, which not only does not cause damage to the cable insulation, but also reduces the weight and volume of the measurement equipment.

2. The present invention uses the method of processing frequency domain information to make local cable defects easier to display, and on this basis, through the combination with Kaiser window function, the sensitivity of local cable defect location is greatly improved.

3. Compared with TDR, the present invention has higher identification ability for local defects with weaker changes, identification accuracy and identification sensitivity for multiple defects, and its identification ability for near-end defects is also stronger.

4. The present invention adopts the method of adding a distance window to the distance diagnostic atlas to make the distortion points easier to display, which greatly improves the positioning accuracy.

5. The present invention adopts the comparison of the distance diagnostic atlas between the cable to be tested and the reference cable and the gating gate technology to find out the distortion point at the same position and eliminate it, so as to eliminate the influence of the cable intermediate joint on the judgment of the local defect position of the cable .

6. The present invention preliminarily evaluates the severity of local defects of the cable by calculating the wave velocity of the cable. On this basis, it further evaluates the severity of the local defects in combination with the analog value fitting curve of the reflection coefficient spectrum after gating, and can accurately evaluate the severity of the cable. Severity of local defects.

7. The cable defect diagnosis method of the present invention is not limited to the voltage level of the cable to be tested. The traditional cable defect diagnosis method needs to select different test equipment according to different voltage levels.

Description of drawings

Fig. 1 is a schematic flow chart of the power cable defect diagnosis method of the present invention.

Fig. 2 is the comparison distance diagnostic atlas of the measured power cable and the reference power cable drawn in Example 1 of the present invention, wherein the solid line represents the distance diagnostic atlas of the tested power cable, and the dotted line represents the distance diagnostic atlas of the reference power cable.

FIG. 3 is a comparison diagram of the reflection coefficient amplitude spectrum drawn in Embodiment 1 of the present invention after passing through the gate technology. The dotted line represents the reflection coefficient amplitude spectrum of the power cable under test after using the gate technology, and the solid line represents the curve fitted according to the mathematical model of the reflection coefficient spectrum modulus.

Fig. 4 is a diagnostic map of local defect distance drawn in Example 2 of the present invention.

Detailed ways

The present invention is specifically described below by way of examples. It is necessary to point out that this embodiment is only used to further illustrate the present invention, and should not be construed as limiting the protection scope of the present invention. Those skilled in the art can make some non-essential improvements and adjustments based on the content of the present invention above.

Example 1

The flow chart of the method for diagnosing local defects in power cables of this embodiment is shown in Figure 1, and the comparison distance diagnostic map of the tested power cable and the reference power cable is shown in Figure 2, which is for the 100m ZR-YJV22-8.7/15 power cable Local defect diagnosis, the defect type is loose copper shielding layer defect. The diagnostic method includes the following steps:

(1) Data measurement of the cable to be tested

Transmit a low-voltage pseudo-random signal V _i to the 100m ZR-YJV22-8.7/15 power cable under test, measure the reflected signal V _r reflected from the end of the cable, and pass Calculate the amplitude |Γ _d (f)| of the reflection coefficient spectrum of the cable, the real part Real(Γ _d (f)) of the reflection coefficient spectrum and the imaginary part Imag(Γ _d (f)) of the reflection coefficient spectrum.

(2) Data processing of the cable to be tested

Using the conversion function f→t′, the measured frequency domain signal is converted into a t′ domain signal, and then the converted Real(Γ _d (t′)) is subjected to fast Fourier (FFT) transformation, and the maximum energy is recorded frequency point f ₀ ′; then multiply the converted Real(Γ _d (t′)) with the Kaiser (Kaiser) window and perform fast Fourier (FFT) transformation to obtain the preliminary processing data KΓ _d , the kaiser window The value of β is 6;

Do the DFT transformation from 0 Hz to the frequency point f ₀ ′ on KΓ _d , and map the transformation result to the original distance diagnostic map D ₀ , where the mapping distance at 0 Hz is 0 meters, and the mapping distance at f ₀ ′ is the cable length of 100 meters .

The obtained original distance diagnostic map is subjected to windowing processing with a distance window length s of 5 meters, and the processing method is as follows:

(3) Reference cable data measurement and processing

Use the non-measured phase cable in the three-phase system as the reference cable, and then use the same method as step (1) to obtain the amplitude |Γ _d1 (f)| of the reflection coefficient spectrum of the reference power cable and the real part of the reflection coefficient spectrum Real (Γ _d1 (f)) and the imaginary part Imag(Γ _d1 (f)) of the reflection coefficient spectrum, and perform the processing under the same conditions in step (2), to obtain the processed distance diagnostic map E.

(4) Determination of the location of local defects in power cables

Comparing the distance diagnostic map D and the distance diagnostic map E, as shown in Figure 2, using the gate technology to eliminate the distortion points at the same position as the distance diagnostic map E in the distance diagnostic map D, and obtain the local defect distance diagnostic map F. From Figure 2, it can be found that there is an obvious distortion point at the position of 70 meters, which is exactly the set local defect point.

(5) Severity assessment of cable local defects

The inverse fast Fourier transform (IFFT) is performed on the obtained local defect distance diagnosis map to obtain the transformed reflection coefficient spectrum t′ domain curve, which is transformed into Real(Γ _d2 (f)) using the conversion function t′→f. Find the frequency f _k corresponding to each maximum point in Real(Γ _d2 (f)), using the formula Calculate the average wave velocity v _c of the tested power cable = 1.7247×10 ⁸ m/s, and measure the wave velocity v ₀ of the new cable = 1.7261×10 ⁸ m/s. Where N is the number of maximum or minimum points in Real(Γ _d2 ) or Imag(Γ _d2 ). The difference between v _c and v ₀ is 0.81%<1%, indicating that the local defect of the tested cable is relatively light.

In order to further evaluate the severity of the local defects of the tested cable, a mathematical model of the spectral modulus of the cable reflection coefficient is established:

abs(Γ _d2 (f))＝a·e ^b·f +c·e ^d·f

Use the established mathematical model to perform data fitting on the reflection coefficient spectrum measured by the measured cable using the gate technology |Γ _d2 (f)|, and the fitting result is As shown in Figure 3, the fitting value corresponding to the extreme point f _k of the higher frequency is obtained and calculated by the formula

error＝std((|Γ _d2 (f _k )|-abs(Γ _d2 (f _k )))/abs(Γ _d2 (f _k )))×100%

Calculate the standard deviation of the relative difference between |Γ _d2 (f)| and the fitting curve error=1.16%<5%, indicating that the local defect of the tested power cable is relatively light, which confirms that the tested power cable has copper shielding at this time Defects of loose layers.

Example 2

The process flow of the power cable local defect diagnosis method of this embodiment is shown in Figure 1, and the comparison distance diagnosis map of the tested power cable and the reference power cable is shown in Figure 4, for the local defect diagnosis of 1500m 10kV XLPE power cable, the defect type Also severe deformation defects of the insulating medium, including the following steps:

(1) Data measurement of the cable to be tested

Transmit low-voltage high-frequency narrow pulse V _i to the 1500m ZR-YJV22-8.7/15 power cable under test, measure the reflected signal V _r reflected from the end of the cable, and pass Calculate the amplitude |Γ _d (f)| of the reflection coefficient spectrum of the cable, the real part Real(Γ _d (f)) of the reflection coefficient spectrum and the imaginary part Imag(Γ _d (f)) of the reflection coefficient spectrum.

(2) Data processing of the cable to be tested

Using the conversion function f→t′, the measured frequency domain signal is converted into a t′ domain signal, and then the converted Imag(Γ _d (t′)) is subjected to discrete Fourier (DFT) transformation, and the maximum energy frequency point f ₀ ′; then the transformed Imag(Γ _d (t′)) is multiplied by the kaiser (Kaiser) window, and the discrete Fourier (DFT) transform is carried out to the preliminary processing data KΓ _d , the kaiser window The value of β is 8;

Perform DFT transformation from 0 Hz to frequency point f ₀ ′ on KΓ _d , and map the transformation result to the original distance diagnostic map D ₀ , where the mapping distance at 0 Hz is 0 meters, and the mapping distance at f ₀ ′ is the cable length of 1500 meters.

The obtained original distance diagnostic map is subjected to windowing processing with a distance window length s of 30 meters, and the processing method is as follows:

(3) Reference cable data measurement and processing

Use the non-tested phase cable in the three-phase system as the reference cable, and then use the same method as step (1) to obtain the amplitude |Γ _d1 (f)| of the reflection coefficient spectrum of the reference power cable and the real part of the reflection coefficient spectrum Real (Γ _d1 (f)) and the imaginary part Imag(Γ _d1 (f)) of the reflection coefficient spectrum, and perform the processing under the same conditions in step (2), to obtain the processed distance diagnostic map E.

(4) Determination of the location of local defects in power cables

Compare the distance diagnostic atlas D and distance diagnostic atlas E, and use the gate technology to eliminate the distortion points at the same position in the distance diagnostic atlas D and distance diagnostic atlas E, and obtain the local defect distance diagnostic atlas F, as shown in Figure 3. From Figure 3, it can be found that there are obvious distortion points at 500 meters and 1200 meters, which are exactly the set local defect points.

(4) Severity assessment of cable local defects

Inverse discrete Fourier transform (IDFT) is performed on the obtained local defect distance diagnostic map to obtain the transformed reflection coefficient spectrum t′ domain curve, which is transformed into Imag(Γ _d2 (f)) using the transfer function t′→f. Find the frequency f _k corresponding to each minimum point in Imag(Γ _d2 (f)), use the formula Calculate the average wave velocity v ₀ of the tested power cable = 1.7217×10 ⁸ m/s, and measure the new cable wave velocity v ₀ = 1.7261×10 ⁸ m/s. Where N is the number of maximum or minimum points in Imag(Γ _d2 ). The difference between v _c and v ₀ is 2.55%>1%, indicating that the local defects of the tested cable are relatively serious.

In order to further evaluate the severity of the local defects of the tested cable, a mathematical model of the spectral modulus of the cable reflection coefficient is established:

abs(Γ _d2 (f))＝a·e ^b·f +c·e ^d·f

Use the established mathematical model to perform data fitting on the reflection coefficient spectrum measured by the measured cable using the gate technology |Γ _d2 (f)|, and the fitting result is Calculate the fitting value corresponding to the extreme point f _k of the higher frequency and use the formula

error＝std((|Γ _d2 (f _k )|-abs(Γ _d2 (f _k )))/abs(Γ _d2 (f _k )))×100%

Calculate the standard deviation of the relative difference between |Γ _d2 (f)| and the fitting curve error=16.14%>10%, indicating that the local defects of the tested power cable are relatively serious, which confirms that the tested power cable has an insulating medium at this time Severely deformed serious fault.

Claims (9)

1. A high-sensitivity power cable local defect diagnosis method, is characterized in that: comprise the following steps: (1) Data measurement of the cable to be tested Transmit a low-voltage linear frequency modulation, pseudo-random or high-frequency narrow pulse signal V _i to the power cable under test, measure the reflected signal V _r reflected from the end of the cable, and pass Find the amplitude |Γ _d (f)| of the reflection coefficient spectrum of the cable and the real part Real(Γ _d (f)) of the reflection coefficient spectrum and the imaginary part Imag(Γ _d (f)) of the reflection coefficient spectrum, where f is the signal frequency; (2) Data processing of the cable to be tested Using the conversion function f→t′, the measured frequency domain signal f is converted into a t′ domain signal, and then the converted Real(Γ _d (t′)) or Imag(Γ _d (t′)) is fast Fourier leaf transform or discrete Fourier transform, record the frequency point f ₀ ′ with the highest energy; then multiply the converted Real(Γ _d (t′)) or Imag(Γ _d (t′)) with the Kaiser window function And perform Fast Fourier Transform or Discrete Fourier Transform to obtain preliminary processing data _KΓd ; Do the discrete Fourier transform from 0 Hz to the frequency point f ₀ ′ on the obtained preliminary processing data KΓ _d , map the transformation result to the original distance diagnostic map D ₀ , and add a distance window to the original distance diagnostic map D ₀ , Obtain the distance diagnostic map D, wherein the mapping distance at 0 Hz is 0 meters, and the mapping distance at the frequency point f ₀ ′ is the length l of the power cable; (3) Reference cable data measurement and processing Use the non-measured phase power cable in the three-phase system as the reference power cable, and then use the same method as step (1) to obtain the amplitude |Γ _d1 (f)| of the reflection coefficient spectrum of the reference power cable and the real value of the reflection coefficient spectrum Part Real (Γ _d1 (f)) and the imaginary part Imag (Γ _d1 (f)) of the reflection coefficient spectrum, and carry out the processing in the same way in the step (2), obtain the distance diagnostic map E after processing; (4) Determination of the location of local defects in power cables Comparing the distance diagnostic map D and the processed distance diagnostic map E, using the gate technology to eliminate the distortion points at the same position in the distance diagnostic map D and the processed distance diagnostic map E, to obtain the local defect distance diagnostic map F, from the local The distortion point found in the defect distance diagnostic map F is the local defect position of the power cable; (5) Judgment of local defects of power cables Inverse discrete Fourier transform or inverse fast Fourier transform is performed on the obtained local defect distance diagnostic map to obtain the transformed t′-domain reflection coefficient spectrum curve, and use the conversion function t′→f to convert the t′-domain reflection coefficient spectrum Convert to Real(Γ _d2 (f)) or Imag(Γ _d2 (f)), and then find each maximum point or minimum value in Real(Γ _d2 (f)) or Imag(Γ _d2 (f)) The frequency f _k corresponding to the point, using the formula Find the average wave velocity of the power cable under test, and compare v _c with the wave velocity v ₀ of a new power cable of the same specification. If the difference between v _c and v ₀ is 1% or more, it means that there is a local defect in the test cable, where N is The number of maximum or minimum points in Real(Γ _d2 ) or Imag(Γ _d2 ). 2. The method for diagnosing local defects of a high-sensitivity power cable according to claim 1, wherein the severity of the local defect of the tested power cable is further evaluated by the following method: Establishment of Mathematical Model of Spectrum Modulus of Cable Reflection Coefficient abs(Γ _d2 (f))＝a·e ^b·f +c·e ^d·f Use the established mathematical model to perform data fitting on the reflection coefficient spectrum measured by the measured power cable using the gate technology |Γ _d2 (f)|, and find the corresponding quasi-parameter at the higher frequency extreme point f _k Combined value, and the defect degree error calculated by the following formula for evaluation: error＝std((|Γ _d2 (f _k )|-abs(Γ _d2 (f _k )))/abs(Γ _d2 (f _k )))×100% When the error<5%, the local defect of the power cable is relatively light; when 5%≤error≤10%, the local defect of the power cable is moderate; when the error>10%, the local defect of the power cable is very serious, and std in the formula is the standard Difference. 3. according to claim 1 and 2 described high-sensitivity power cable local defect diagnosis methods, it is characterized in that, the β value in the described kaiser window function satisfies: Where α is the side lobe attenuation multiple. 4. according to claim 1 and 2 described high-sensitivity power cable local defect diagnostic method, it is characterized in that, original distance diagnostic atlas D ₀ adopts following formula to carry out adding distance window processing: Where s is the window length, and the value is not greater than the spatial resolution in the distance diagnostic atlas D ₀ , and D is the distance diagnostic atlas obtained after processing. 5. the high-sensitivity power cable local defect diagnosis method according to claim 3, is characterized in that, original distance diagnosis atlas D ₀ adopts following formula to carry out adding distance window processing: Where s is the window length, and the value is not greater than the spatial resolution in the distance diagnostic atlas D ₀ , and D is the distance diagnostic atlas obtained after processing. 6. The method for diagnosing local defects of high-sensitivity power cables according to claim 1 or 2, characterized in that: in step (5), the new cable wave velocity _v of the same specification is given by the formula Obtained by calculation, where L and C are the inductance and capacitance per unit length of a new cable of the same specification, or measured by time domain reflection method. 7. The high-sensitivity power cable local defect diagnosis method according to claim 3 is characterized in that: in the step (5), the new cable wave velocity _v of the same specification is given by the formula Obtained by calculation, where L and C are the inductance and capacitance per unit length of a new cable of the same specification, or measured by time domain reflection method. 8. The high-sensitivity power cable local defect diagnosis method according to claim 4 is characterized in that: in step (5), the new cable wave velocity _v of the same specification is given by the formula Obtained by calculation, where L and C are the inductance and capacitance per unit length of a new cable of the same specification, or measured by time domain reflection method. 9. The high-sensitivity power cable local defect diagnosis method according to claim 5 is characterized in that: in the step (5), the new cable wave velocity _v of the same specification is given by the formula Obtained by calculation, where L and C are the inductance and capacitance per unit length of a new cable of the same specification, or measured by time domain reflection method.