CN104331701A - Insulator external discharge mode identification method based on ultraviolet map - Google Patents

Abstract

The invention relates to an insulator external discharge mode identification method based on an ultraviolet map. The method is characterized in that the method comprises the following steps: a) to begin with, carrying out non-contact detection on insulator external discharge by utilizing an ultraviolet imager and forming an ultraviolet discharge image; b) carrying out gray scale processing on the obtained ultraviolet discharge image and converting a color picture into a binary gray image; c) then, carrying out mathematical morphology filtering on the binary gray image to obtain a clear ultraviolet light spot binary gray map; d) carrying out constant-torch characteristic parameter extraction on the ultraviolet light spot binary gray map obtained after filtering, and seven constant-torch characteristic parameters are extracted; and e) inputting the seven constant-torch characteristic parameters into a neural network to carry out discharge type mode identification, and outputting the identification results. Experiment shows that the ultraviolet detection map is utilized to carry out mode recognition of insulator external discharge, and the method has the advantages of high identification degree and simple method.

Description

Based on the exterior insulator discharge mode recognition methods of uv-spectrogram

Technical field

The present invention relates to a kind of exterior insulator discharge mode recognition methods based on uv-spectrogram.

Background technology

Insulator is the insulator arrangement be most widely used in electric system, and it has a large amount of application in circuit, transformer station.Insulator can cause external discharge due to surface filth, condensation, burr etc. in operational process, and long-term external discharge can cause its decreasing insulating, even causes it badly damaged, causes the generation of power grid accident.There occurs a lot of accident due to exterior insulator electric discharge initiation in domestic electrical network, therefore strengthening the determination and analysis of exterior insulator electric discharge is the key avoiding this type of accident.It is a kind of development in recent years that ultraviolet imagery detects, and applies more electrical equipment external discharge non-contact detection method at the scene.What current application was more is day blind type ultraviolet detection method, it has that flexible operation, detecting distance are far away, noncontact, highly sensitive, fast response time, not by features such as daylight interference, be widely applied in the external discharge context of detection of the equipment such as insulator, wire.By carrying out UV detect to insulator, can whether external discharge occurred to it the detecting of aspect directly perceived, because it is easy to use, visual result, the application in electrical network is more and more extensive.

Current UV detect result presents mainly with the form of ultraviolet picture or video, it can provide the quantized values whether electric discharge and discharging light subnumber occur intuitively, but result detailed further cannot be provided, as this insulator occur be surface filth electric discharge or surperficial condensation electric discharge then cannot provide, which also limits the further analysis to Site Detection result.Therefore, obtaining accurately on UV detect basis, be further analyzed its detected image, obtaining testing result specifically, is deeply utilize the method, effectively analyzes the basis of testing result.

Summary of the invention

The object of this invention is to provide a kind of exterior insulator discharge mode recognition methods based on uv-spectrogram, can accurately realize utilizing ultraviolet image process and extracting parameter thus carry out the identification of discharge mode.

Based on an exterior insulator discharge mode recognition methods for uv-spectrogram, its special feature is, comprises the steps:

A, first, utilizes ultraviolet imager to carry out non-contact detection to exterior insulator electric discharge, and forms EUV discharge image;

B, secondly, carries out gray proces by the EUV discharge image of acquisition, colour picture is converted into two-value gray level image;

C, then, carries out mathematical morphology filter to two-value gray level image, obtains ultraviolet hot spot two-value gray images clearly;

D, carries out constant torch characteristic parameter extraction according to the ultraviolet hot spot two-value gray images after filtering, extracts 7 constant torch characteristic parameters;

7 constant torch characteristic parameter input neural networks are carried out electric discharge type pattern-recognition by e, export recognition result.

In step b when colour picture being converted into two-value gray scale picture, first by original image digitizing, form image digitization matrix, be then " 1 " by the white portion assignment in matrix, remainder assignment is " 0 ", formation two-value gray level image.

In characteristic parameter extraction stage in steps d, utilize the constant torch feature of image, extract 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation as characteristic parameter amount.

Further, steps d specifically comprises the steps:

Carry out constant torch characteristic parameter extraction according to the ultraviolet hot spot collection of illustrative plates after filtering, extract 7 constant torch characteristic parameters;

Concrete, two dimension (p+q) the rank square of digital picture f (x, y) is defined as:

$m_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{x}^p{y}^{q}f(x,y);$ (formula 1)

The wherein character matrix of f (x, y) for being formed after image digitazation, x characterize image digitazation after each point horizontal ordinate, y characterize each point ordinate; (p, q)=0,1,2, sue for peace and to carry out on all volume coordinate x, the y of image, corresponding center square is defined as:

${\mathrm{\μ}}_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{(x-\stackrel{\‾}{x})}^p{(y-\stackrel{\‾}{y})}^{q}f(x,y);$ (formula 2)

Wherein:

$\stackrel{\‾}{x}=\frac{m_{10}}{m_{00}},\stackrel{\‾}{y}=\frac{m_{01}}{m_{00}};$

M in formula ₁₀can be tried to achieve by p=1, q=0 in formula (1); m ₀₀can be tried to achieve by p=0, q=0 in formula (1); m ₀₁can be tried to achieve by p=0, q=1 in formula (1);

Normalization (p+q) center, rank square is defined as:

${\mathrm{\η}}_{\mathrm{pq}}=\frac{{\mathrm{\μ}}_{\mathrm{pq}}}{{{\mathrm{\μ}}^{\mathrm{\γ}}}_{00}};$

Wherein p, q=0,1,2 ...,

$\mathrm{\γ}=\frac{p+q}{2}+1;$

Wherein p+q=2,3 ...;

Select to 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation, its computing formula is:

(1)φ ₁＝η ₂₀+η ₀₂；

(2)φ ₂＝(η ₂₀-η ₀₂) ²+4η ₁₁ ²；

(3)φ ₃＝(η ₃₀-3η ₁₂) ²+(3η ₂₁-η ₀₃) ²；

(4)φ ₄＝(η ₃₀+η ₁₂) ²+(η ₂₁+η ₀₃) ²；

φ ₅＝(η ₃₀-3η ₁₂)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]

(5) +(3η ₂₁-η ₀₃)(η ₂₁+η ₀₃)[3(n ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]；

φ ₆＝(η ₂₀-η ₀₂)[(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]

(6) +4η ₁₁(η ₃₀+η ₁₂)(η ₂₁+η ₀₃)；

φ ₇＝(3η ₂₁-η ₀₃)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]+

(7) (3η ₁₂-η ₃₀)(η ₂₁+η ₀₃)[3(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]。

Further, steps d specifically comprises the steps: 7 constant torch characteristic parameter input neural networks to carry out electric discharge type pattern-recognition, judges whether electric discharge belongs to surface filth electric discharge, the electric discharge of surperficial condensation, surface spikes electric discharge.

Further, the concrete process utilizing neural network to carry out pattern-recognition is: the uv-spectrogram obtaining insulator typical defect external discharge first in the lab, extract the constant torch feature operator of typical defect external discharge ultraviolet image, neural network is utilized to train typical defect feature operator, morphogenesis characters database, secondly for the electric discharge ultraviolet image of UNKNOWN TYPE, after obtaining its constant torch feature operator, input neural network, carry out pattern-recognition, export recognition result, output recognition result is contaminant flashover, condensation electric discharge or burr electric discharge.

Through probationary certificate, the inventive method utilizes UV detect collection of illustrative plates to carry out the pattern-recognition of exterior insulator electric discharge, has that identification degree is high, the simple feature of method.

Accompanying drawing explanation

Accompanying drawing 1 is for the present invention is based on the schematic flow sheet of the exterior insulator discharge mode recognition methods of uv-spectrogram.

Embodiment

As shown in Figure 1, the invention provides a kind of exterior insulator discharge mode recognition methods based on uv-spectrogram, the ultraviolet image that first Site Detection obtains by the method carries out gradation conversion, colour picture is converted into two-value gray level image; Then mathematical morphology filter is carried out to two-value gray level image, obtain ultraviolet hot spot two-value gray images clearly; Constant torch characteristic parameter extraction is carried out to ultraviolet hot spot collection of illustrative plates, extracts 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation as characteristic parameter; Characteristic parameter input neural network is carried out electric discharge type pattern-recognition.The method utilizes UV detect collection of illustrative plates to carry out the pattern-recognition of exterior insulator electric discharge, has that identification degree is high, the simple feature of method.This mode identification method based on image parameter solves the difficult point that current UV detect result cannot carry out electric discharge type pattern-recognition.

Embodiment 1:

Below in conjunction with accompanying drawing, the technical scheme in the present invention is clearly and completely described.

Figure 1 shows that the schematic flow sheet of the GIS partial discharge mode identification method that the present invention is based on ultrasound examination, described method comprises the steps:

Step 1: utilize ultraviolet imager to carry out UV detect to insulator, detects and obtains its EUV discharge original image, and described original image is the coloured image comprising discharge signal and other signals.

Step 2: colored original image is carried out gray proces, is converted into two-value gray level image by coloured image.

Concrete, first by original image digitizing, form image digitization matrix f (x, y), be then " 1 " by the white portion assignment in matrix, remainder assignment is " 0 ", formation two-value gray level image.

Step 3: carry out mathematical morphology filter to two-value gray level image, obtains ultraviolet hot spot two-value gray images clearly.Image mathematical morphology filter is conventional Image denoising algorithm, is existing ripe algorithm, repeats no more herein.

Step 4: carry out constant torch characteristic parameter extraction according to the ultraviolet hot spot collection of illustrative plates after filtering, extracts 7 constant torch characteristic parameters.

Concrete, two dimension (p+q) the rank square of digital picture f (x, y) is defined as:

$m_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{x}^p{y}^{q}f(x,y);$

Wherein (p, q)=0,1,2, sue for peace and to carry out on all volume coordinate x, the y of image.Corresponding center square is defined as:

${\mathrm{\μ}}_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{(x-\stackrel{\‾}{x})}^p{(y-\stackrel{\‾}{y})}^{q}f(x,y);$

Wherein:

$\stackrel{\‾}{x}=\frac{m_{10}}{m_{00}},\stackrel{\‾}{y}=\frac{m_{01}}{m_{00}};$

Normalization (p+q) center, rank square is defined as:

${\mathrm{\η}}_{\mathrm{pq}}=\frac{{\mathrm{\μ}}_{\mathrm{pq}}}{{{\mathrm{\μ}}^{\mathrm{\γ}}}_{00}};$

Wherein p, q=0,1,2 ...,

$\mathrm{\γ}=\frac{p+q}{2}+1;$

Wherein p+q=2,3 ...,

Select to 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation, its computing formula is:

1.φ ₁＝η ₂₀+η ₀₂；

2.φ ₂＝(η ₂₀-η ₀₂) ²+4η ₁₁ ²；

3.φ ₃＝(η ₃₀-3η ₁₂) ²+(3η ₂₁-η ₀₃) ²；

4.φ ₄＝(η ₃₀+η ₁₂) ²+(η ₂₁+η ₀₃) ²；

φ ₅＝(η ₃₀-3η ₁₂)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]

5.+(3η ₂₁-η ₀₃)(η ₂₁+η ₀₃)[3(n ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]；

φ ₆＝(η ₂₀-η ₀₂)[(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]

6.+4η ₁₁(η ₃₀+η ₁₂)(η ₂₁+η ₀₃)；

φ ₇＝(3η ₂₁-η ₀₃)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]+

7.(3η ₁₂-η ₃₀)(η ₂₁+η ₀₃)[3(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]；

These 7 constant torch operators characterize the shape facility of uv-spectrogram, and its value does not change with the translation of image, convergent-divergent, mirror image and rotation, and different electric discharge types has different shape facilities.

Further, specifically comprise the steps:

Carry out constant torch characteristic parameter extraction according to the ultraviolet hot spot collection of illustrative plates after filtering, extract 7 constant torch characteristic parameters;

Concrete, two dimension (p+q) the rank square of digital picture f (x, y) is defined as:

$m_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{x}^p{y}^{q}f(x,y);$ (formula 1)

The wherein character matrix of f (x, y) for being formed after image digitazation, x characterize image digitazation after each point horizontal ordinate, y characterize each point ordinate.(p, q)=0,1,2, sue for peace and to carry out on all volume coordinate x, the y of image, corresponding center square is defined as:

${\mathrm{\μ}}_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{(x-\stackrel{\‾}{x})}^p{(y-\stackrel{\‾}{y})}^{q}f(x,y);$ (formula 2)

Wherein:

$\stackrel{\‾}{x}=\frac{m_{10}}{m_{00}},\stackrel{\‾}{y}=\frac{m_{01}}{m_{00}};$

M in formula ₁₀can be tried to achieve by p=1, q=0 in formula (1); m ₀₀can be tried to achieve by p=0, q=0 in formula (1); m ₀₁can be tried to achieve by p=0, q=1 in formula (1).

Normalization (p+q) center, rank square is defined as:

${\mathrm{\η}}_{\mathrm{pq}}=\frac{{\mathrm{\μ}}_{\mathrm{pq}}}{{{\mathrm{\μ}}^{\mathrm{\γ}}}_{00}};$

Wherein p, q=0,1,2 ...,

$\mathrm{\γ}=\frac{p+q}{2}+1;$

Wherein p+q=2,3 ...;

Select to 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation, its computing formula is:

(1)φ ₁＝η ₂₀+η ₀₂；

(2)φ ₂＝(η ₂₀-η ₀₂) ²+4η ₁₁ ²；

(3)φ ₃＝(η ₃₀-3η ₁₂) ²+(3η ₂₁-η ₀₃) ²；

(4)φ ₄＝(η ₃₀+η ₁₂) ²+(η ₂₁+η ₀₃) ²；

φ ₅＝(η ₃₀-3η ₁₂)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]

(5) +(3η ₂₁-η ₀₃)(η ₂₁+η ₀₃)[3(n ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]；

φ ₆＝(η ₂₀-η ₀₂)[(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]

(6) +4η ₁₁(η ₃₀+η ₁₂)(η ₂₁+η ₀₃)；

φ ₇＝(3η ₂₁-η ₀₃)(η ₃₀+η ₁₂)[(η ₃₀+η ₁₂) ²-3(η ₂₁+η ₀₃) ²]+

(7) (3η ₁₂-η ₃₀)(η ₂₁+η ₀₃)[3(η ₃₀+η ₁₂) ²-(η ₂₁+η ₀₃) ²]。

Each letter all can be derived by formula 1 and formula 2 above, and the implication of letter is 0,1,2,3 etc.

Step 5: 7 constant torch characteristic parameter input neural networks are carried out electric discharge type pattern-recognition, judges whether electric discharge belongs to surface filth electric discharge, the electric discharge of surperficial condensation and surface spikes electric discharge.

Concrete, the process utilizing neural network to carry out pattern-recognition is: the uv-spectrogram obtaining insulator typical defect external discharge first in the lab, extract the constant torch feature operator of typical defect external discharge ultraviolet image, neural network is utilized to train typical defect feature operator, morphogenesis characters database.Secondly for the electric discharge ultraviolet image of UNKNOWN TYPE, after obtaining its constant torch feature operator, input neural network, carries out pattern-recognition, exports recognition result.

Step 6: output recognition result is contaminant flashover, condensation electric discharge or burr electric discharge.

The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, anyly belongs to those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claim.

Claims (6)

1., based on an exterior insulator discharge mode recognition methods for uv-spectrogram, it is characterized in that, comprise the steps:

A, first, utilizes ultraviolet imager to carry out non-contact detection to exterior insulator electric discharge, and forms EUV discharge image;

B, secondly, carries out gray proces by the EUV discharge image of acquisition, colour picture is converted into two-value gray level image;

C, then, carries out mathematical morphology filter to two-value gray level image, obtains ultraviolet hot spot two-value gray images clearly;

D, carries out constant torch characteristic parameter extraction according to the ultraviolet hot spot two-value gray images after filtering, extracts 7 constant torch characteristic parameters;

7 constant torch characteristic parameter input neural networks are carried out electric discharge type pattern-recognition by e, export recognition result.

2. as claimed in claim 1 based on the exterior insulator discharge mode recognition methods of uv-spectrogram, it is characterized in that: in step b when colour picture being converted into two-value gray scale picture, first by original image digitizing, form image digitization matrix, then be " 1 " by the white portion assignment in matrix, remainder assignment is " 0 ", forms two-value gray level image.

3. as claimed in claim 1 based on the exterior insulator discharge mode recognition methods of uv-spectrogram, it is characterized in that: in characteristic parameter extraction stage in steps d, utilize the constant torch feature of image, extract 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation as characteristic parameter amount.

4., as claimed in claim 1 based on the exterior insulator discharge mode recognition methods of uv-spectrogram, it is characterized in that, steps d specifically comprises the steps:

Carry out constant torch characteristic parameter extraction according to the ultraviolet hot spot collection of illustrative plates after filtering, extract 7 constant torch characteristic parameters;

Concrete, two dimension (p+q) the rank square of digital picture f (x, y) is defined as:

$m_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{x}^p{y}^{q}f(x,y);$ (formula 1)

The wherein character matrix of f (x, y) for being formed after image digitazation, x characterize image digitazation after each point horizontal ordinate, y characterize each point ordinate; (p, q)=0,1,2, sue for peace and to carry out on all volume coordinate x, the y of image, corresponding center square is defined as:

${\mathrm{\μ}}_{\mathrm{pq}}=\underset{x}{\mathrm{\Σ}}\underset{y}{\mathrm{\Σ}}{(x-\stackrel{\‾}{x})}^p{(y-\stackrel{\‾}{y})}^{q}f(x,y);$ (formula 2)

Wherein:

$\stackrel{\‾}{x}=\frac{m_{10}}{m_{00}},\stackrel{\‾}{y}=\frac{m_{01}}{m_{00}};$

M in formula ₁₀can be tried to achieve by p=1, q=0 in formula (1); m ₀₀can be tried to achieve by p=0, q=0 in formula (1); m ₀₁can be tried to achieve by p=0, q=1 in formula (1);

Normalization (p+q) center, rank square is defined as:

${\mathrm{\η}}_{\mathrm{pq}}=\frac{{\mathrm{\μ}}_{\mathrm{pq}}}{{{\mathrm{\μ}}^{\mathrm{\γ}}}_{00}};$

Wherein p, q=0,1,2 ...,

$\mathrm{\γ}=\frac{p+q}{2}+1;$

Wherein p+q=2,3 ...;

Select to 7 to all insensitive two dimension invariant moment parameter of image translation, convergent-divergent, mirror image and rotation, its computing formula is:

(1)φ ₁＝η ₂₀+η ₀₂；

(2)φ ₂＝(η ₂₀-η ₀₂) ²+4η ₁₁ ²；

(3)φ ₃＝(η ₃₀-3η ₁₂) ²+(3η ₂₁-η ₀₃) ²；

(4)φ ₄＝(η ₃₀+η ₁₂) ²+(η ₂₁+η ₀₃) ²；

$\left(5\right)\begin{array}{c}{\mathrm{\φ}}_5=({\mathrm{\η}}_{30}-3{\mathrm{\η}}_{12})({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})[{({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})}^2-3{({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})}^2]\\ +(3{\mathrm{\η}}_{21}-{\mathrm{\η}}_{03})({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})[3{(n_{30}+{\mathrm{\η}}_{12})}^2-{({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})}^2]\end{array};$

$\left(6\right)\begin{array}{c}{\mathrm{\φ}}_6=({\mathrm{\η}}_{20}-{\mathrm{\η}}_{02})[{({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})}^2-3{({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})}^2]\\ +4{\mathrm{\η}}_{11}({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})\end{array};$

$\left(7\right)\begin{array}{c}{\mathrm{\φ}}_7=({3\mathrm{\η}}_{21}-{\mathrm{\η}}_{03})({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})[{({\mathrm{\η}}_{30}+{\mathrm{\η}}_{12})}^2-3{({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})}^2]+\\ (3{\mathrm{\η}}_{12}-{\mathrm{\η}}_{30})({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})[3{(n_{30}+{\mathrm{\η}}_{12})}^2-{({\mathrm{\η}}_{21}+{\mathrm{\η}}_{03})}^2]\end{array}.$

5. as claimed in claim 1 based on the exterior insulator discharge mode recognition methods of uv-spectrogram, it is characterized in that, steps d specifically comprises the steps: 7 constant torch characteristic parameter input neural networks to carry out electric discharge type pattern-recognition, judges whether electric discharge belongs to surface filth electric discharge, the electric discharge of surperficial condensation, surface spikes electric discharge.

6. as claimed in claim 5 based on the exterior insulator discharge mode recognition methods of uv-spectrogram, it is characterized in that, the concrete process utilizing neural network to carry out pattern-recognition is: the uv-spectrogram obtaining insulator typical defect external discharge first in the lab, extract the constant torch feature operator of typical defect external discharge ultraviolet image, neural network is utilized to train typical defect feature operator, morphogenesis characters database, secondly for the electric discharge ultraviolet image of UNKNOWN TYPE, after obtaining its constant torch feature operator, input neural network, carry out pattern-recognition, export recognition result, output recognition result is contaminant flashover, condensation electric discharge or burr electric discharge.