CN111257242A - High-spectrum identification method for pollutant components of insulator - Google Patents
High-spectrum identification method for pollutant components of insulator Download PDFInfo
- Publication number
- CN111257242A CN111257242A CN202010122812.9A CN202010122812A CN111257242A CN 111257242 A CN111257242 A CN 111257242A CN 202010122812 A CN202010122812 A CN 202010122812A CN 111257242 A CN111257242 A CN 111257242A
- Authority
- CN
- China
- Prior art keywords
- sample
- components
- pollution
- component
- reflectivity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N2021/1765—Method using an image detector and processing of image signal
- G01N2021/177—Detector of the video camera type
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a hyperspectral recognition method for insulator filthy components, which comprises the following steps: simulating and detecting artificial pollution of the insulator; acquiring hyperspectral images of different samples, selecting a first-class pollution area as a sensitive area ROI, analyzing by taking a reflectivity spectrum of the sensitive area ROI as an input quantity principal component, calculating load factors of the first three principal components, substituting the load factors into a sample set spectrum, calculating to obtain scores of the first three principal components, and constructing to obtain three-dimensional distribution characteristic spaces of different pollution components; acquiring a hyperspectral image of a sample to be detected to extract a sample surface spectrum, and substituting the hyperspectral image into principal component score calculation formulas of different pollution components respectively to obtain the scores of the first three principal components of the sample to be detected; and respectively carrying out scoring and three-dimensional distribution characteristic space clustering analysis on the first three main components of the sample to be detected, and realizing the identification of the pollution components according to clustering results.
Description
Technical Field
The invention belongs to the technical field of on-line monitoring of power equipment, and particularly relates to a hyperspectral identification method for insulator contamination components.
Background
The insulator plays an important role in mechanical support and electrical isolation of a high-voltage power transmission and distribution line, has excellent stain resistance in an initial state, but the stain resistance is obviously reduced due to the aging effect of long-term natural environment and mechanical stress, so that pollution flashover is caused, and the safety of a power grid is seriously damaged. Research shows that the pollution flashover voltage is not only related to the pollution degree, but also has a significant relation with pollution components, so that the effective identification of the pollution components on the surface of the insulator is significant.
The existing pollution component analysis means are offline detection, ion detection is carried out in a laboratory through material analysis equipment (such as XRD, EDS and the like) after a tower is manually arranged on the tower to take down a pollution sample on the surface of a flange, and the compound components are presumed in an ion pairing mode. In order to keep the pollution distribution information, part of researchers take the insulator back to a laboratory for analysis after quitting the running state, and the pollution distribution characteristics are kept by extracting pollution samples in different areas on the insulator for analysis.
In summary, the development conditions of the analysis means of the pollution components of the insulator at the present stage are complex, and an effective means applied to live detection under the actual working condition is lacked. Based on the current situation, the invention provides a hyperspectral identification method of the pollutant components of the insulator based on principal component analysis and spatial clustering, which can be used for acquiring non-contact type pollutant state information through a hyperspectral camera and then realizing charged detection of the pollutant components by matching with the hyperspectral identification method.
The above information disclosed in this background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides the hyperspectral recognition method for the insulator contamination components, which can be used for acquiring information of a non-contact contamination state through a hyperspectral camera, realizing the efficient recognition of the contamination components and solving the defects that the analysis means of the insulator contamination components at the present stage is complex in development conditions and lacks of effective means which can be used for large-scale detection under actual working conditions.
The invention aims to realize the hyperspectral recognition method of the polluted components of the insulator by the following technical scheme, which comprises the following steps:
in the first step, insulator artificial pollution simulation detection is carried out, wherein a high-temperature vulcanized silicone rubber sheet is used as a base material, a pollution component substance is selected as a detection object, the surface of the base material is quantitatively coated with the pollution component substance, the base material is divided into m types of samples according to different components, and the number of each type of sample is n sample sets;
in the second step, hyperspectral images of different samples are obtained, a first class of pollution area is selected as a sensitive area ROI, the ROI reflectivity spectrum of the sensitive area ROI is used as an input quantity main component for analysis, load factors of the first three main components are calculated and substituted into the sample set spectrumWherein, tnjThe reflectance value of the Nth sample under the J wave band is calculated to obtain the scores of the first three principal components and the three-dimensional distribution characteristic space of different pollution components is constructedj is the number of wave bands, n is the number of samples, PCn 1-PCn 3 respectively represent the scores of the first three principal components of the nth sample, and the rest types of pollution samples are carried out according to the same steps to construct m three-dimensional distribution characteristic spaces, wherein the principal component score formula is as follows:each sample gave x ═ PC1PC2PC3]N samples are constructed to obtain a feature spaceObtaining m characteristic spaces by the same method, wherein K is a load factor vector, and t is the reflectivity intensity under different wave bands; (ii) a
In the third step, a hyperspectral image of the sample to be detected is obtained to extract a sample surface spectrum, and the hyperspectral image is substituted into principal component score calculation formulas of different pollution components respectively to obtain the first three principal component scores x ═ PC of the sample to be detected1PC2PC3];
In the fourth step, the first three principal components of the sample to be tested are respectively scored as x ═ PC1PC2PC3]And three-dimensionally distributed feature spaceAnd (4) clustering analysis, namely realizing filth component identification according to a clustering result.
In the method, in the first step, the surface of a base material is quantitatively coated with a filthy component substance to obtain the mass of the filthy component in unit area, the mass of each filthy component is calculated according to the area of the base material, and the sample surface is uniformly coated with the filthy component after being stirred based on distilled water and the filthy component.
In the method, in the second step, hyperspectral images of different samples are obtained through black and white correction and filtering smoothing.
In the method, in the second step, a standard polytetrafluoroethylene white board with the reflectivity of 99 percent is collected asCalibrating white images, closing a lens cover with the reflectivity of 0% to acquire dark correction images, and realizing the reflectivity correction of the acquired hyperspectral data through black and white correction according to a correction formulaWherein I is the corrected hyperspectral image reflectivity0The hyperspectral image reflectivity before correction is shown, B is the total black corrected image reflectivity, and W is the total white corrected image reflectivity.
In the method, the filtering smoothing comprises Fourier transform, convolution filtering, mathematical morphology filtering or Savitzky-Golay smoothing.
In the second step, the principal component analysis adopts a covariance matrix or a correlation coefficient matrix as a transformation matrix to perform data dimension reduction on the original spectral data, and a PC is selected according to the eigenvalue of the transformation matrix1,PC2,PC3Characteristic value of λ1,λ2,λ3The corresponding feature vector isLoad factor vector k for each type of sampleijThe calculation formula is as follows:j is the number of bands.
In the method, in the third step, the spectrum T 'of the sample to be detected is extracted from the main component score of the sample to be detected'To be measured=[t′1,…,t′j]Calculating the principal component score x using the principal component score formulas corresponding to the m classes of substances, respectivelym=[PC′1PC′2PC′3]。
In the fourth step, the method calculates x in the dirty component identification according to the clustering resultm=[PC1PC2PC3]Andsum of euclidean distances between:
min (S) with minimum sum of Euclidean distances to sample space1,…,Sm) The sample component (2) is of the fouling class.
In the method, the filth components comprise NaCl and CaSO4、SiO2、CuSO4、CaCl2Or Fe2O3。
Compared with the prior art, the invention has the following advantages:
the method can overcome the defects that the development conditions of the analysis means of the pollution components of the insulator are complex in the prior stage and effective means which can be applied to large-scale detection under the actual working condition is lacked, non-contact pollution state information acquisition is carried out through a hyperspectral camera, and the pollution component identification method is matched to realize high-efficiency identification of the pollution components, so that the non-contact detection of the pollution components on the surface of the insulator can be realized, and the method has the capability of developing the field live inspection of the actual working condition; the distribution characteristic of the filthy components on the surface of the insulator can be effectively reserved; the identification of the pollution components can be effectively realized, and the pollution flashover accident of the power transmission line caused by different pollution components can be effectively avoided.
Drawings
Various other advantages and benefits of the present invention will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. It is obvious that the drawings described below are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. Also, like parts are designated by like reference numerals throughout the drawings.
In the drawings:
FIG. 1 is a schematic flow chart of a hyperspectral identification method of insulator contamination components according to an embodiment of the invention;
fig. 2 is a feature space distribution diagram of eight types of pollution components according to an embodiment of the present invention.
The invention is further explained below with reference to the figures and examples.
Detailed Description
Specific embodiments of the present invention will be described in more detail below with reference to fig. 1 to 2. While specific embodiments of the invention are shown in the drawings, it should be understood that the invention may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
It should be noted that certain terms are used throughout the description and claims to refer to particular components. As one skilled in the art will appreciate, various names may be used to refer to a component. This specification and claims do not intend to distinguish between components that differ in name but not function. In the following description and in the claims, the terms "include" and "comprise" are used in an open-ended fashion, and thus should be interpreted to mean "include, but not limited to. The description which follows is a preferred embodiment of the invention, but is made for the purpose of illustrating the general principles of the invention and not for the purpose of limiting the scope of the invention. The scope of the present invention is defined by the appended claims.
For the purpose of facilitating understanding of the embodiments of the present invention, the following description will be made by taking specific embodiments as examples with reference to the accompanying drawings, and the drawings are not to be construed as limiting the embodiments of the present invention.
In order to better understand, the hyperspectral recognition method for the insulator contamination components comprises the following steps:
in the first step S1, performing artificial pollution simulation detection on the insulator, wherein a high-temperature vulcanized silicone rubber sheet is used as a base material, a pollution component substance is selected as a detection object, the surface of the base material is quantitatively coated with the pollution component substance, the base material is divided into m types of samples according to different components, and the number of each type of sample is n;
in the second step S2, hyperspectral images of different samples are obtained, a first class of filth area is selected as a sensitive area ROI, the reflectivity spectrum of the sensitive area ROI is used as the input quantity principal component for analysis, the load factors of the first three principal components are calculated and substituted into the spectrum of the sample setWherein, tnjThe reflectance value of the Nth sample under the J wave band is calculated to obtain the scores of the first three principal components and the three-dimensional distribution characteristic space of different pollution components is constructedj is the number of wave bands, n is the number of samples, PCn 1-PCn 3 respectively represent the scores of the first three principal components of the nth sample, and the rest types of pollution samples are carried out according to the same steps to construct m three-dimensional distribution characteristic spaces, wherein the principal component score formula is as follows:each sample gave x ═ PC1PC2PC3]N samples are constructed to obtain a feature spaceObtaining m characteristic spaces by the same method, wherein K is a load factor vector, and t is the reflectivity intensity under different wave bands;
in the third step S3, a hyperspectral image of the sample to be tested is obtained to extract a sample surface spectrum, and the hyperspectral image is substituted into principal component score calculation formulas of different contamination components respectively to obtain the first three principal component scores x ═ PC of the sample to be tested1PC2PC3];
In the fourth step S4, the first three principal components of the sample are scored as x ═ PC1PC2PC3]And three-dimensionally distributed feature spaceAnd (4) clustering analysis, namely realizing filth component identification according to a clustering result.
In a preferred embodiment of the method, in the first step S1, the mass of the contaminant component per unit area is obtained by quantitatively coating the contaminant component on the surface of the base material, and the mass of each contaminant component is calculated from the area of the base material and uniformly coated on the surface of the sample after being stirred with the contaminant component based on distilled water.
In a preferred embodiment of the method, in a second step S2, hyperspectral images of different samples are acquired via black and white correction and smoothing by filtering.
In a preferred embodiment of the method, in the second step S2, a standard teflon whiteboard with a reflectivity of 99% is collected as a white calibration image, then a lens cover with a reflectivity of 0% is closed to collect a dark correction image, the reflectivity correction of the collected hyperspectral data is realized by black and white correction, and the correction formula is thatWherein I is the corrected hyperspectral image reflectivity0The hyperspectral image reflectivity before correction is shown, B is the total black corrected image reflectivity, and W is the total white corrected image reflectivity.
In a preferred embodiment of the method, the smoothing by filtering comprises fourier transform, convolution filtering, mathematical morphology filtering or Savitzky-Golay smoothing.
In a preferred embodiment of the method, in the second step S2, the principal component analysis uses the covariance matrix or the correlation coefficient matrix as a transformation matrix to perform data dimension reduction on the original spectrum data, and selects the PC according to the eigenvalue of the transformation matrix1,PC2,PC3Characteristic value of λ1,λ2,λ3The corresponding feature vector is Load of each type of sampleFactor vector kijThe calculation formula is as follows:
In a preferred embodiment of the method, in the third step S3, the spectrum T 'of the sample to be measured is extracted from the score of the principal component of the sample to be measured'To be measured=[t′1,…,t′j]Calculating the principal component score x using the principal component score formulas corresponding to the m classes of substances, respectivelym=[PC'1PC'2PC'3]。
In a preferred embodiment of the method, in the fourth step S4, in the recognition of the filth component based on the clustering result, x is calculatedm=[PC1PC2PC3]Andsum of euclidean distances between:
min (S) with minimum sum of Euclidean distances to sample space1,…,Sm) The sample component (2) is of the fouling class.
In a preferred embodiment of said method, the fouling component comprises NaCl, CaSO4、SiO2、CuSO4、CaCl2Or Fe2Oa。
To further describe the present invention, it is further described below with reference to the accompanying drawings.
In this embodiment:
as shown in fig. 1, a hyperspectral identification method of insulator contamination components based on principal component analysis and spatial clustering is characterized by comprising the following steps:
s1: carrying out an artificial pollution simulation experiment, taking a high-temperature vulcanized silicone rubber sheet composite insulator umbrella skirt manufacturing material as a base material, selecting a typical pollution component substance as a detection object, coating corresponding pollution on the surface of the base material according to a quantitative coating method, and dividing the pollution into m types of samples according to different components, wherein the number of each type of sample is n.
S2: acquiring hyperspectral images of different pollution samples, selecting a first class of pollution area as an inductive area region of interest (ROI), performing principal component analysis (principal component analysis, PCA) by taking a reflectivity spectrum of the ROI area as an input quantity, calculating load factors of the first three principal components, substituting the load factors into a sample set spectrumj is the number of wave bands, n is the number of samples, the scores of the first three principal components are calculated and constructed to obtain the three-dimensional distribution characteristic space of different pollution componentsn is the number of samples of the type of contamination. The other types of pollution samples are carried out according to the same steps, m three-dimensional distribution feature spaces can be constructed, as shown in fig. 2, feature spaces of 8 common pollution substances are established in fig. 2. The spatial distribution in three-dimensional coordinates of the feature space constituted by the first three PC scores of the different samples, i.e. PC1 PC2 PC3, fig. 2 is an example of the results made on 80 samples.
S3: acquiring a hyperspectral image of a sample to be detected, extracting a surface spectrum of an object to be detected, respectively substituting the surface spectrum into principal component score calculation formulas of different pollution components, and obtaining the first three principal component scores x ═ PC (personal computer) of the sample to be detected1PC2PC3]。
S4: respectively scoring principal components of the sample to be measured with x ═ PC1PC2PC3]And the above characteristic spacePerforming cluster analysis, and identifying filth components according to the cluster result by calculating xm=[PC1PC2PC3]Andoldham's series betweenSum of distances:
considering the minimum sum of Euclidean distances to the sample space, namely min (S)1,…,Sm) The sample component (2) is of the fouling class.
In one embodiment, the identification method comprises the steps of:
s1: carrying out an artificial pollution simulation experiment, taking a high-temperature vulcanized silicone rubber sheet composite insulator umbrella skirt manufacturing material as a base material, selecting a typical pollution component substance as a detection object, coating corresponding pollution on the surface of the base material according to a quantitative coating method, and dividing the pollution into m types of samples according to different components, wherein the number of each type of sample is n.
S2: acquiring hyperspectral images of different pollution samples, selecting a first class of pollution area as an inductive area region of interest (ROI), carrying out principal component analysis (principal component analysis, RCA) by taking a reflectivity spectrum of the ROI as an input quantity, calculating load factors of the first three principal components, substituting the load factors into a sample set spectrumj is the number of wave bands, n is the number of samples, the scores of the first three principal components are calculated and constructed to obtain the three-dimensional distribution characteristic space of different pollution componentsn is the number of samples of the type of contamination. And performing the other categories of pollution samples according to the same steps, and constructing m three-dimensional distribution characteristic spaces.
S3: acquiring a hyperspectral image of a sample to be detected, extracting a surface spectrum of an object to be detected, respectively substituting the surface spectrum into principal component score calculation formulas of different pollution components, and obtaining the first three principal component scores x ═ PC (personal computer) of the sample to be detected1PC2PC3]。
S4: respectively scoring principal components of the sample to be measured with x ═ PC1PC2PC3]And the above characteristic spaceAnd performing clustering analysis, and realizing filth component identification according to a clustering result.
In some forms, wherein: in step S1, the artificial pollution simulation experiment includes: 1, selecting typical pollutant components including: NaCl, CaSO4,SiO2,CuSO4,CaCl2,Fe2O3Etc.; 2 with reference to IEC 60507: 2013 obtaining the quality of the filth component in unit area; 3, calculating the mass of each filthy component according to the area of the substrate material, and measuring by using a one-thousandth balance; 4, taking a proper amount of distilled water, stirring the distilled water and the components, and then uniformly brushing the mixture on the surface of the sample.
In some forms, wherein: in step S2, the hyperspectral image acquisition method includes black and white correction and filtering smoothing. The black and white correction method comprises the following steps: firstly, a standard polytetrafluoroethylene white board with the reflectivity of 99% is collected as a white calibration image, then a lens cover with the reflectivity of 0% is closed to collect a dark correction image, the reflectivity correction of the collected hyperspectral data can be realized through black and white correction, and the correction algorithm is shown as the formula (1):in the formula I0The hyperspectral image reflectivity before correction is shown, B is the total black corrected image reflectivity, and W is the total white corrected image reflectivity. Filtering smoothing methods include, but are not limited to, fourier transform, convolution filtering, mathematical morphology filtering, Savitzky-Golay smoothing methods, and the like.
In some forms, wherein: in step S2, Principal Component Analysis (PCA) may use the covariance matrix or the correlation coefficient matrix as a transformation matrix to perform data dimension reduction on the original spectral data, and select a PC according to the eigenvalue of the transformation matrix1,PC2,PC3Characteristic value of λ1,λ2,λ3The corresponding feature vector isLoad factor vector k for each type of sampleijThe calculation formula is as follows:
In some forms, wherein: in step S2, the principal component score calculation formula is as follows:
x ═ PC can be obtained for each sample1PC2PC3]N samples can be constructed to obtain a feature spaceThe same approach can yield m feature spaces.
In some forms, wherein: in the step S3, the method for calculating the principal component score of the sample to be measured extracts the spectrum T 'of the sample to be measured'To be measured=[t′1,…,t′j]Calculating the principal component score x using the principal component score calculation formulas corresponding to the m classes of substances, respectivelym=[PC′1PC'2PC'3]。
In some forms, wherein: in the step S4, the main component scores of the samples to be tested are clustered and analyzed, and the dirty component identification is realized according to the clustering result, wherein the specific method is to calculate xm=[PC1PC2PC3]Andsum of euclidean distances between:
considering the minimum sum of Euclidean distances to the sample space, namely min (S)1,…,Sm) The sample component (2) is of the fouling class.
Although the embodiments of the present invention have been described above with reference to the accompanying drawings, the present invention is not limited to the above-described embodiments and application fields, and the above-described embodiments are illustrative, instructive, and not restrictive. Those skilled in the art, having the benefit of this disclosure, may effect numerous modifications thereto without departing from the scope of the invention as defined by the appended claims.
Claims (9)
1. A hyperspectral identification method for insulator contamination components comprises the following steps:
in the first step (S1), insulator artificial pollution simulation detection is carried out, wherein high-temperature vulcanized silicone rubber sheets are used as base materials, pollution component substances are selected as detection objects, the surface of the base materials is quantitatively coated with the pollution component substances, the base materials are divided into m types of samples according to different components, and the number of each type of sample is n sample sets;
in the second step (S2), hyperspectral images of different samples are obtained, a first class of filth area is selected as a sensitive area ROI, the reflectivity spectrum of the sensitive area ROI is used as an input quantity main component for analysis, load factors of the first three main components are calculated and substituted into the sample set spectrumWherein, tnjThe reflectance value of the Nth sample under the J wave band is calculated to obtain the scores of the first three principal components and the three-dimensional distribution characteristic space of different pollution components is constructedj is the number of wave bands, n is the number of samples, PCn 1-PCn 3 respectively represent the scores of the first three principal components of the nth sample, and the rest types of pollution samples are carried out according to the same steps to construct m three-dimensional distribution characteristic spaces, wherein the principal component score formula is as follows:each sample gave x=[PC1PC2PC3]N samples are constructed to obtain a feature spaceObtaining m characteristic spaces by the same method, wherein K is a load factor vector, and t is the reflectivity intensity under different wave bands;
in the third step (S3), a hyperspectral image of the sample to be tested is obtained to extract a sample surface spectrum, and the hyperspectral image is substituted into principal component score calculation formulas of different contamination components, so as to obtain the first three principal component scores x ═ PC of the sample to be tested1PC2PC3];
2. The method according to claim 1, wherein preferably, in the first step (S1), the mass of the contaminant component per unit area is obtained by quantitatively coating the contaminant component on the surface of the base material, and the mass of each contaminant component is calculated based on the area of the base material and is uniformly coated on the surface of the sample after being stirred with the contaminant component based on distilled water.
3. The method according to claim 1, wherein in a second step (S2), hyperspectral images of different samples are acquired via black and white correction and filter smoothing.
4. The method of claim 1, wherein in the second step (S2), a standard Teflon white board with a reflectivity of 99% is collected as a white calibration image, then a lens cover with a reflectivity of 0% is closed to collect a dark correction image, and the reflectivity correction of the collected hyperspectral data is realized by black and white correction, wherein the correction formula is thatWherein I is the corrected hyperspectral image reflectivity0The hyperspectral image reflectivity before correction is shown, B is the total black corrected image reflectivity, and W is the total white corrected image reflectivity.
5. The method of claim 3, wherein filtering smoothing comprises Fourier transform, convolution filtering, mathematical morphology filtering, or Savitzky-Golay smoothing.
6. The method as claimed in claim 1, wherein in the second step (S2), the principal component analysis performs data dimension reduction on the original spectral data using the covariance matrix or the correlation coefficient matrix as a transformation matrix, and the PC is selected according to the eigenvalue size of the transformation matrix1,PC2,PC3Characteristic value of λ1,λ2,λ3The corresponding feature vector isLoad factor vector k for each type of sampleijThe calculation formula is as follows:
7. The method according to claim 1, wherein in the third step (S3), the spectrum T 'of the sample to be tested is extracted from the score of the principal components of the sample to be tested'To be measured=[t′1,…,t′j]Calculating the principal component score x using the principal component score formulas corresponding to the m classes of substances, respectivelym=[PC′1PC′2PC′3]。
8. The method according to claim 7, wherein in the fourth step (S4), in implementing the filth component recognition based on the clustering result, x is calculatedm=[PC1PC2PC3]Andsum of euclidean distances between:
min (S) with minimum sum of Euclidean distances to sample space1,…,Sm) The sample component (2) is of the fouling class.
9. The method of claim 1, wherein the fouling component comprises NaCl, CaSO4、SiO2、CuSO4、CaCl2Or Fe2O3。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010122812.9A CN111257242A (en) | 2020-02-27 | 2020-02-27 | High-spectrum identification method for pollutant components of insulator |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010122812.9A CN111257242A (en) | 2020-02-27 | 2020-02-27 | High-spectrum identification method for pollutant components of insulator |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111257242A true CN111257242A (en) | 2020-06-09 |
Family
ID=70947537
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010122812.9A Pending CN111257242A (en) | 2020-02-27 | 2020-02-27 | High-spectrum identification method for pollutant components of insulator |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111257242A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113655072A (en) * | 2021-08-20 | 2021-11-16 | 福建中烟工业有限责任公司 | Method, apparatus and computer readable medium for detecting contaminants on a surface of a sample |
WO2023284104A1 (en) * | 2021-07-14 | 2023-01-19 | 南方电网科学研究院有限责任公司 | Method and apparatus for calculating algae coverage of insulator surface, and storage medium |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102841063A (en) * | 2012-08-30 | 2012-12-26 | 浙江工业大学 | Method for tracing and identifying charcoal based on spectrum technology |
CN104849287A (en) * | 2015-06-10 | 2015-08-19 | 国家电网公司 | Composite insulator contamination degree non-contact detection method |
CN106596579A (en) * | 2016-11-15 | 2017-04-26 | 同济大学 | Insulator contamination condition detection method based on multispectral image information fusion |
CN109685099A (en) * | 2018-11-12 | 2019-04-26 | 江苏大学 | A kind of apple variety discriminating conduct of the preferred fuzzy clustering of spectral band |
CN109799245A (en) * | 2019-03-29 | 2019-05-24 | 云南电网有限责任公司电力科学研究院 | A kind of insulator contamination degree non-contact detection method and device |
CN110082310A (en) * | 2019-05-30 | 2019-08-02 | 海南大学 | A kind of near infrared band EO-1 hyperion diagnostic method of rubber tree LTN content |
CN110132890A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | According to the method and device of the unmanned culinary cuisine operation of food materials optimizing components |
CN110188735A (en) * | 2019-06-10 | 2019-08-30 | 中国农业科学院深圳农业基因组研究所 | A kind of instruction plant recognition methods based on EO-1 hyperion |
CN110261405A (en) * | 2019-07-31 | 2019-09-20 | 西南交通大学 | Insulator contamination ingredient recognition methods based on micro- hyperspectral technique |
CN110376214A (en) * | 2019-08-13 | 2019-10-25 | 西南交通大学 | Insulator dirty degree non-contact detection method based on hyperspectral technique |
CN110530876A (en) * | 2019-09-04 | 2019-12-03 | 西南交通大学 | Insulator dirty degree development prediction method based on shot and long term Memory Neural Networks |
-
2020
- 2020-02-27 CN CN202010122812.9A patent/CN111257242A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102841063A (en) * | 2012-08-30 | 2012-12-26 | 浙江工业大学 | Method for tracing and identifying charcoal based on spectrum technology |
CN104849287A (en) * | 2015-06-10 | 2015-08-19 | 国家电网公司 | Composite insulator contamination degree non-contact detection method |
CN106596579A (en) * | 2016-11-15 | 2017-04-26 | 同济大学 | Insulator contamination condition detection method based on multispectral image information fusion |
CN109685099A (en) * | 2018-11-12 | 2019-04-26 | 江苏大学 | A kind of apple variety discriminating conduct of the preferred fuzzy clustering of spectral band |
CN109799245A (en) * | 2019-03-29 | 2019-05-24 | 云南电网有限责任公司电力科学研究院 | A kind of insulator contamination degree non-contact detection method and device |
CN110132890A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | According to the method and device of the unmanned culinary cuisine operation of food materials optimizing components |
CN110082310A (en) * | 2019-05-30 | 2019-08-02 | 海南大学 | A kind of near infrared band EO-1 hyperion diagnostic method of rubber tree LTN content |
CN110188735A (en) * | 2019-06-10 | 2019-08-30 | 中国农业科学院深圳农业基因组研究所 | A kind of instruction plant recognition methods based on EO-1 hyperion |
CN110261405A (en) * | 2019-07-31 | 2019-09-20 | 西南交通大学 | Insulator contamination ingredient recognition methods based on micro- hyperspectral technique |
CN110376214A (en) * | 2019-08-13 | 2019-10-25 | 西南交通大学 | Insulator dirty degree non-contact detection method based on hyperspectral technique |
CN110530876A (en) * | 2019-09-04 | 2019-12-03 | 西南交通大学 | Insulator dirty degree development prediction method based on shot and long term Memory Neural Networks |
Non-Patent Citations (2)
Title |
---|
周利军 等: "《高电压技术实验》", 31 October 2011 * |
邱彦 等: "基于高光谱技术的绝缘子污秽等级检测方法", 《高电压技术》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023284104A1 (en) * | 2021-07-14 | 2023-01-19 | 南方电网科学研究院有限责任公司 | Method and apparatus for calculating algae coverage of insulator surface, and storage medium |
CN113655072A (en) * | 2021-08-20 | 2021-11-16 | 福建中烟工业有限责任公司 | Method, apparatus and computer readable medium for detecting contaminants on a surface of a sample |
CN113655072B (en) * | 2021-08-20 | 2023-12-26 | 福建中烟工业有限责任公司 | Method, apparatus and computer readable medium for detecting contaminants on a sample surface |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1814044B1 (en) | Paint pigment identification method, identification system, identification program, and recording medium thereof | |
CN107643267B (en) | A kind of lossless comprehensive recognition methods of ancient wall pigment based on visible spectrum imaging | |
Converse et al. | Discrimination of whole from broken corn kernels with image analysis | |
US7027627B2 (en) | Medical decision support system and method | |
CN111257242A (en) | High-spectrum identification method for pollutant components of insulator | |
CN104572538B (en) | A kind of Chinese medicine tongue image color correction method based on K PLS regression models | |
US20120002852A1 (en) | Advanced Digital Pathology and Provisions for Remote Diagnostics | |
CN110874576B (en) | Pedestrian re-identification method based on typical correlation analysis fusion characteristics | |
JP2023503564A (en) | Method and apparatus for identifying effect pigments in target coatings | |
CN101477025B (en) | Fast evaluation method for collection exhibition materials based on image processing | |
CN108960413A (en) | A kind of depth convolutional neural networks method applied to screw surface defects detection | |
CN111896497B (en) | Spectral data correction method based on predicted value | |
CN115908950B (en) | Rapid medical hyperspectral image classification method based on similarity tangent mapping | |
CN116519710A (en) | Method and system for detecting surface pollution state of composite insulator | |
CN114994069A (en) | Hyperspectrum-based method and system for detecting contamination components and content of insulator | |
CN111750993B (en) | Open measurement environment spectrum measurement method based on imaging condition correction | |
TW202238527A (en) | Air pollution detection method based on frequency spectrum image generating a detection result relevant to the air pollution level of the to-be-detected environmental frequency spectrum image | |
JP4460988B2 (en) | How to differentiate stratum corneum | |
WO2014053520A1 (en) | Targeting cell nuclei for the automation of raman spectroscopy in cytology | |
CN108181480A (en) | A kind of identification device and method of paraffin odor grade | |
Liu et al. | 3D Rotation-Invariant description from tensor operation on spherical HOG field | |
CN110956605A (en) | Cervical cancer cytology detection algorithm based on protein antibody reagent | |
CN110111029A (en) | A kind of morphological method for identifying red deer, roe deer and sika deer hair | |
Limsiroratana et al. | Computer-aided system for microscopic images: application to breast cancer nuclei counting | |
CN109406413A (en) | Method based on mountain balsam phenol content in high light spectrum image-forming technology prediction fructus lycii |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |