CN112557834B - Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum - Google Patents
Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum Download PDFInfo
- Publication number
- CN112557834B CN112557834B CN202011116736.7A CN202011116736A CN112557834B CN 112557834 B CN112557834 B CN 112557834B CN 202011116736 A CN202011116736 A CN 202011116736A CN 112557834 B CN112557834 B CN 112557834B
- Authority
- CN
- China
- Prior art keywords
- aging
- raman spectrum
- sample
- oil
- aging stage
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention provides an oil paper insulation equipment aging diagnosis method based on Raman spectrum. According to the method, a large number of oil paper insulation aging samples are obtained through an accelerated thermal aging experiment, the polymerization degree of the insulation paper of the aging samples is measured, the corresponding relation between the aging stage and the polymerization degree of the oil paper insulation aging samples is established, the insulation sample to be tested is quickly and effectively classified and regressed in the aging stage through KNN analysis of the Raman spectrum of the insulation oil to be tested and the Raman spectrum of the aging oil sample, and therefore real-time diagnosis of the aging condition of the oil paper insulation equipment is achieved. The diagnosis method is efficient and time-saving, saves the pretreatment work of chromatography, does not need to train a model, and can realize non-contact continuous measurement; the influence of environmental factors is small, and online monitoring is easy to realize.
Description
Technical Field
The invention belongs to the field of power equipment safety state evaluation, and particularly relates to an oil paper insulation equipment aging diagnosis method based on Raman spectrum.
Background
The oil-paper composite insulation is an important and excellent insulation combination mode and is widely used for power equipment such as large transformers, sleeves, transformers, capacitors, cables and the like. In the operation process of the oil paper insulation power equipment, the oil paper insulation of the oil paper insulation power equipment bears various external stress effects of heat, electricity, machinery, chemistry and the like for a long time, so that the self insulation and the mechanical property are gradually reduced, the equipment aging problem occurs, equipment faults can be caused along with the increase of the aging degree, and the reliable operation of a power system is influenced. Therefore, the aging degree of the oil paper insulating material is accurately diagnosed, the insulation aging state of the power equipment is mastered in time, a basis can be provided for the insulation state and the whole life cycle management of the oil-immersed power transmission and transformation equipment, and the safe operation of a power grid is ensured.
Up to now, the measurement of the degree of polymerization of the oiled paper insulation equipment is the index for determining the degree of aging which is the highest in international recognition. However, in the actual operation process of the equipment, the insulation paper cannot be directly taken out, so that the measurement of the polymerization degree can only be used as a judgment standard in the experimental process, but not as a judgment index in the actual operation process. Therefore, the diagnosis of the aging stage in the actual operation process of the oil-paper insulation equipment is easier to realize by using the detection analysis of the insulation oil which can be directly taken out as a judgment index.
At present, methods for detecting the aging state of oil mainly comprise High Performance Liquid Chromatography (HPLC), gas Chromatography (GC) and the like, but the methods can only carry out measurement on limited samples, and more pretreatment work is needed, so that the time is long. Raman spectroscopy has been widely used in the fields of substance composition analysis and condition diagnosis. Compared with the prior art, the Raman spectrum technology is applied to the analysis of the furfural content in the transformer oil and has the advantages that the Raman spectrum detection is non-contact measurement, and continuous measurement can be realized; the transformer aging trace features have Raman activity and can be subjected to Raman detection; is not greatly influenced by environmental factors and is easy to realize on-line monitoring.
Therefore, if the aging degree of the oil-paper insulation equipment in the actual operation process can be effectively and reliably diagnosed according to the Raman spectrum of the insulation oil, a great problem in the field of evaluation of the safety state of the power equipment can be solved.
Disclosure of Invention
In view of the above, in order to solve the problems in the prior art, the present invention aims to provide a method for diagnosing aging of an oil-paper insulation device based on raman spectroscopy of insulation oil.
In order to achieve the above object, the present invention specifically adopts the following technical solutions.
A method for diagnosing aging of oiled paper insulation equipment based on Raman spectroscopy is characterized by comprising the following steps:
step A: establishing a corresponding relation between the aging stage and the polymerization degree of the oiled paper insulation aging sample: obtaining an oil paper insulation aging sample through an oil paper insulation thermal aging experiment, periodically interrupting heating according to set sampling time, and measuring the polymerization degree after cooling;
and B: acquiring Raman spectrum data of the aged oil sample in the known aging stage in the step A, carrying out standardization processing on the data, screening information wave bands according to the Raman spectrum data of the aged oil sample with each wave band number and a correlation coefficient corresponding to aging time, taking the Raman spectrum data of the aged oil sample corresponding to the screened information wave band as a sample Raman spectrum, and establishing a sample Raman spectrum database 1;
and C: acquiring Raman spectrum data of the insulating oil at an unknown aging stage, carrying out standardization processing on the data, calculating the similarity between the Raman spectrum of the insulating oil sample and the Raman spectrum of the sample in the database 1 according to the Pearson correlation coefficient, and selecting the Raman spectra of m known aging stage aging oil samples which are closest to the Raman spectrum of the insulating oil at the unknown aging stage as a database 2 according to the similarity;
step D: calculating the similarity between the Raman spectrum of the insulating oil sample at the unknown aging stage in the step C and the Raman spectrum of the sample in the database 2 according to the Euclidean distance, and selecting the Raman spectrums of p known aging oil samples at the unknown aging stage, which are closest to the Raman spectrum of the insulating oil sample at the unknown aging stage, as a database 3 according to the similarity;
step E: and judging the aging stage and the polymerization degree of the insulating oil sample at the unknown aging stage according to the value of the polymerization degree corresponding to the Raman spectrum of the sample in the database 3 and the rule of the KNN algorithm.
The present invention further includes the following preferred embodiments.
The step B comprises the following steps:
step B1: acquiring Raman spectrum data of the aged oil sample in each aging stage;
and step B2: the Raman spectrum of the aging oil sample at each aging stage is subjected to standardization treatment, and the standardization treatment is carried out according to the following formula
Wherein x is*Is a normalized value, xminIs the minimum value of the spectral data, xmaxIs the maximum of the spectral data
The value x is the original value of the spectral data;
and step B3: screening the characteristic quantity by calculating a correlation coefficient:
the characteristics are screened by adopting a method for calculating the correlation coefficient of the Raman intensity and the aging time at each wavenumber, and the calculation method of the correlation coefficient adopts the following calculation formula of the Pearson correlation coefficient:
in the formula, rho is a Pearson correlation coefficient between Raman spectrum data of various wave bands of the aged oil sample in the known aging stage and corresponding aging time, x (y) is Raman spectrum intensity x (i) of a certain wave number when the aging time is y, n is the number of the wave bands, and y (i) is the aging time corresponding to the ith wave band;
selecting a wave band with the absolute value of the correlation coefficient larger than or equal to 0.5 as an information wave band for subsequent analysis; and establishes a sample raman spectrum database 1.
The step C comprises the following steps:
step C1: acquiring Raman spectrum data of the insulating oil sample at an unknown aging stage;
and C2: carrying out standardization processing on the acquired Raman spectrum data;
and C3: selecting standardized Raman spectrum data corresponding to the information wave band screened in the step B;
and C4: calculating the correlation coefficient rho of the Raman spectrum of the insulating oil sample at the unknown aging stage selected by C3 and the sample spectrum in the sample Raman spectrum database 1 established in the step B according to the Pearson correlation coefficientp：
Where rhopFor the Pearson correlation coefficient, y, between the Raman spectrum of the insulating oil sample at the unknown aging stage and the spectrum of the sample in database 1 established in step BTestIs a characteristic value, y, of the Raman spectrum of the insulating oil sample at an unknown aging stageRefIs a characteristic value, n, of the spectrum of the sample in the database 1bB, the number of the information wave bands obtained in the step B is the total wave band number corresponding to the sample spectrum in the sample Raman spectrum database 1;
and C5, selecting m wave band Raman spectrums closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity to construct a database 2 of the Raman spectrums of the aging oil samples in the known aging stage.
The upper limit of m is 10.
Said m is preferably 8.
The step D comprises the following steps:
step D1: calculating the similarity rho between the Raman spectrum of the insulating oil in each unknown aging stage after the standardization treatment in the step C and the aging oil sample in the m wave band known aging stages obtained in the step C according to the following formulae：
In the formula, yTestIs a characteristic value, y, of the Raman spectrum of the insulating oil sample at an unknown aging stageref2Is known asAging the characteristic value of the oil sample in an aging stage;
step D2: and selecting the Raman spectrums of the aged oil samples in the known aging stage of p wave bands closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity, taking the Raman spectrums of the aged oil samples in the aging stage of the p wave bands as samples, and establishing a database 3.
The upper limit of p is 10.
Said p is preferably 5.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention overcomes the problem that the aging degree of the oil paper insulation power equipment is difficult to diagnose in real time because the insulation paper cannot be directly taken out for detection in the actual operation process, determines the aging degree of the aged sample by the index of the oil paper polymerization degree with the highest international recognition degree by establishing the corresponding relation between the aging stage and the polymerization degree of the oil paper insulation aged sample, and can quickly and effectively realize the real-time diagnosis of the integral aging state of the on-site oil paper insulation power equipment by the Raman spectrum KNN algorithm analysis of the aged oil sample and the insulation oil sample to be detected.
2. According to the invention, a large number of oil paper insulation aging samples are obtained through an accelerated thermal aging experiment, and the polymerization degree of the insulation paper of the aging samples is measured, so that the corresponding relation between the aging stage and the polymerization degree of the oil paper insulation aging samples is established, and a large number of effective samples can be obtained in a short time.
3. In order to improve the accuracy of the KNN algorithm, the invention selects k nearest neighbor samples in the KNN algorithm
For treating
The process is divided into two parts: first the pearson correlation coefficient is taken as the first criterion and then the euclidean measure is taken as the second criterion. Meanwhile, in order to reduce random errors, 10 times of interactive verification is adopted to calculate the prediction precision. The reliability of the KNN algorithm is higher.
4. The diagnosis method is efficient and time-saving, omits the pretreatment work of chromatography, does not need to train the model, quickly and effectively classifies and regresses the aging stage of the insulation sample to be measured, and can realize non-contact continuous measurement; the influence of environmental factors is small, and online monitoring is easy to realize.
Drawings
FIG. 1 is a 12 sets of raw Raman spectra;
FIG. 2a is a correlation coefficient of each band;
FIG. 2b is a Raman spectrum after screening;
FIG. 3a is the classification accuracy for different values of m and n;
FIG. 3b shows the regression error for different values of m and n.
FIG. 3c is the classification result;
FIG. 3d shows the predicted results.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only used to illustrate the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
The invention discloses an oil paper insulation equipment aging diagnosis method based on Raman spectrum, which comprises the following steps:
step A: establishing a corresponding relation between the aging stage and the polymerization degree of the oiled paper insulation aging sample: obtaining an oil paper insulation aging sample through an oil paper insulation heat aging experiment, periodically interrupting heating according to set sampling time, and measuring polymerization degree after cooling;
and B: acquiring Raman spectrum data of the aged oil sample in the known aging stage in the step A, carrying out standardization processing on the data, screening information wave bands according to the Raman spectrum data of the aged oil sample with each wave band number and a correlation coefficient corresponding to aging time, taking the Raman spectrum data of the aged oil sample corresponding to the screened information wave band as a sample Raman spectrum, and establishing a sample Raman spectrum database 1;
b1, acquiring Raman spectrum data of the aged oil sample in each aging stage;
step B2, carrying out standardization treatment on the obtained Raman spectra of the aged oil samples in each aging stage, wherein the standardization treatment is carried out according to the following formula
Wherein x*Is a normalized value, xminIs the minimum value of the spectral data, xmaxIs the maximum of the spectral data
The value x is the original value of the spectrum data;
the raman spectra of the insulating oil after normalization on each aging day are shown in figure 1,
and B3, screening characteristic quantity through calculation of correlation coefficient:
screening characteristics by adopting a method for calculating the correlation coefficient of the Raman intensity and the aging time at each wave number, wherein the calculation method of the correlation coefficient adopts the following calculation formula of the Pearson correlation coefficient:
in the formula, rho is a Pearson correlation coefficient between Raman spectrum data of various wave bands of the aged oil sample in the known aging stage and corresponding aging time, x (y) is Raman spectrum intensity x (i) of a certain wave number when the aging time is y, n is the number of the wave bands, and y (i) is the aging time corresponding to the ith wave band;
the correlation coefficient for each wave number is shown in figure 2a,
selecting a wave band with the absolute value of the correlation coefficient being greater than or equal to 0.5 as an information wave band for subsequent analysis; and a sample raman spectrum database 1 is established, and the screened raman spectrum is shown as a shaded part in fig. 2 b.
And C: acquiring Raman spectrum data of the insulating oil at an unknown aging stage, carrying out standardization processing on the data, calculating the similarity between the Raman spectrum of the insulating oil sample and the Raman spectrum of the sample in the database 1 according to the Pearson correlation coefficient, and selecting the Raman spectra of m known aging stage aging oil samples which are closest to the Raman spectrum of the insulating oil at the unknown aging stage as a database 2 according to the similarity;
the step C comprises the following steps:
step C1, acquiring Raman spectrum data of the insulating oil sample at an unknown aging stage;
step C2, carrying out standardization processing on the obtained Raman spectrum data;
c3, selecting standardized Raman spectrum data corresponding to the information wave band screened in the step B;
step C4, calculating the correlation coefficient rho of the Raman spectrum of the insulating oil sample at the unknown aging stage selected by the step C3 and the sample spectrum in the sample Raman spectrum database 1 established in the step B according to the Pearson correlation coefficientp：
Where ρ ispFor the Pearson correlation coefficient, y, between the Raman spectrum of the insulating oil sample at the unknown aging stage and the spectrum of the sample in database 1 established in step BTestIs a characteristic value, y, of the Raman spectrum of the insulating oil sample at an unknown aging stageRefIs a characteristic value, n, of the spectrum of the sample in the database 1bB, the number of the information wave bands obtained in the step B is the total wave band number corresponding to the sample spectrum in the sample Raman spectrum database 1;
and C5, selecting m wave band Raman spectrums closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity to construct a database 2 of the Raman spectrums of the aging oil samples in the known aging stage.
The upper limit of m is 10.
In the present invention, m is preferably 8.
Step D: calculating the similarity between the Raman spectrum of the insulating oil sample at the unknown aging stage in the step C and the Raman spectrum of the sample in the database 2 according to the Euclidean distance, and selecting the Raman spectrums of p known aging oil samples at the unknown aging stage, which are closest to the Raman spectrum of the insulating oil sample at the unknown aging stage, as a database 3 according to the similarity;
the step D comprises the following steps:
step D1, calculating the similarity rho between the Raman spectrum of the insulating oil in each unknown aging stage after the standardization treatment in the step C and the aging oil sample samples in the m wave band known aging stages obtained in the step C according to the following formulae：
In the formula, yTestIs a characteristic value, y, of Raman spectrum of an insulating oil sample at an unknown aging stageref2The characteristic value of the aged oil sample in the known aging stage is obtained;
and D2, selecting the Raman spectrums of the aged oil sample in the known aging stage of p wave bands closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity, taking the Raman spectrums of the aged oil sample in the aging stage of the p wave bands as samples, and establishing a database 3.
The upper limit of p is 10.
In the present invention, p is preferably 5.
The selection process of m and p values in step C and step D is shown in fig. 3a and fig. 3 b.
Step E: and judging the aging stage and the polymerization degree of the insulating oil sample at the unknown aging stage according to the value of the polymerization degree corresponding to the Raman spectrum of the sample in the database 3 and the rule of the KNN algorithm. FIG. 3c shows the effect of the aging stage of the randomly selected 23 samples, and FIG. 3d shows the effect of the polymerization degree prediction of the randomly selected 23 samples
The technical solution of the present invention is described in detail by a specific embodiment.
Establishing a corresponding relation between the aging stage and the polymerization degree of the oiled paper insulation aging sample:
1. obtaining the aging sample of the insulation of the oil paper by the experiment of the thermal aging of the insulation of the oil paper (Chongqing university transmission and distribution equipment and system safety and new technology national key laboratory)
In the experiment, cramey 25 # naphthenic mineral oil was used as the insulating oil, and kraft paper 0.2mm thick was used as the insulating paper. Drying oil and paper in vacuum oven at 90 deg.C for 48 hr, wherein the water content of oil and paper is controlled to 10mg/kg and below 0.5%. The temperature of the vacuum box was adjusted to 60 ℃ and the pattern was immersed in insulating oil for 24 hours. The mass ratio of the oil paper is 10.
The oiled paper sample was transferred to a heating tank, which was evacuated and charged with nitrogen. The heated cans of the oilpaper samples were placed in a constant temperature oil bath pan stabilized at 130 ℃ and subjected to accelerated heat aging for 25 days.
2. The insulating oil and the insulating paper were periodically taken out, and the degree of polymerization of the insulating paper was measured.
12 aging samples aged for 2 days, 4 days, 6 days, 8 days, 10 days, 12 days, 14 days, 16 days, 18 days, 20 days, 23 days and 25 days are collected. In order to better reflect the classification effect, the aged samples are classified into 12 types according to sampling time.
The degree of polymerization of the 12-type aged insulation paper was measured.
And (3) detecting the Raman spectrum of the aged sample:
and carrying out Raman spectrum detection on the 12 known aging samples with different aging degrees to obtain Raman spectra of the samples. To reduce the effect of errors in the spectral measurements, the mean of 5 replicates per sample was used as sample spectral data in this study. The raw spectral data obtained are shown in fig. 1.
Data processing and screening features
The spectrum signal has baseline interference, which is generated by fluorescent substances, impurities in oil, fluorescence of insulating oil and equipment generated in the aging process of the oil paper insulation, and the baseline interference has great influence on the extraction of the spectrum characteristic quantity. Therefore, the first step of data preprocessing is to adopt a cubic spline function to remove a base line, then adopt a five-point cubic smoothing algorithm to reduce spectral noise, and finally carry out normalization processing on the data.
The original spectral signal of the insulating oil is 1023 data points, and the high data dimensionality brings huge computation load, is difficult to optimize the algorithm and even causes dimensionality disaster. In the invention, after data preprocessing such as noise removal, baseline correction, normalization and the like is carried out, a method for calculating the correlation coefficient of Raman intensity and aging time at each wave number is adopted to screen characteristics so as to establish a diagnosis model of the oil paper insulation aging stage. The screening results are shown in FIG. 2.
Raman spectrum determination of insulating oil sample to be detected, classification and regression through KNN algorithm
And acquiring the Raman spectrum of the oil sample to be detected, and classifying and regressing the Raman spectrum by a KNN algorithm. Euclidean metrics are a common definition of distance, which refers to the actual distance between two points. In order to improve the accuracy of the KNN algorithm, the selection process of k nearest neighbor samples in the KNN algorithm is divided into two parts. Firstly, the Pearson correlation coefficient is used as a first criterion, and 8 samples closest to the sample to be detected are selected as a new database by calculating the Pearson correlation coefficient between the sample to be detected and the sample library sample, which is called as primary screening. Then, euclidean measures are taken as a second criterion. And selecting 5 samples closest to the sample to be tested by calculating Euclidean measures between the sample to be tested and 8 samples obtained by one-time screening, and taking the class with the highest occurrence frequency in the 5 samples as the prediction class of the predicted sample.
By the method, the final aging category prediction precision can reach 87.92% at most. When m is 9,n is 5, the regression predicts a minimum RMSE of 54.28.
While the best mode for carrying out the invention has been described in detail and illustrated in the accompanying drawings, it is to be understood that the same is by way of illustration of the preferred embodiment of the invention and is not to be considered as a limitation on the scope of the invention, since it is to be understood that all changes and modifications that come within the spirit of the invention are desired to be protected by the following claims.
Claims (4)
1. The aging diagnosis method of the oiled paper insulation equipment based on the Raman spectrum is characterized by comprising the following steps:
step A: establishing a corresponding relation between the aging stage and the polymerization degree of the oiled paper insulation aging sample: obtaining an oil paper insulation aging sample through an oil paper insulation heat aging experiment, periodically interrupting heating according to set sampling time, and measuring polymerization degree after cooling;
and B: acquiring Raman spectrum data of the aged oil sample in the known aging stage in the step A, carrying out standardization processing on the data, screening information wave bands according to the Raman spectrum data of the aged oil sample with each wave band number and a correlation coefficient corresponding to aging time, taking the Raman spectrum data of the aged oil sample corresponding to the screened information wave band as a sample Raman spectrum, and establishing a sample Raman spectrum database 1;
and C: acquiring Raman spectrum data of the insulating oil at an unknown aging stage, carrying out standardization processing on the data, calculating the similarity between the Raman spectrum of the insulating oil sample and the Raman spectrum of the sample in the database 1 according to the Pearson correlation coefficient, and selecting the Raman spectra of m known aging stage aging oil samples which are closest to the Raman spectrum of the insulating oil at the unknown aging stage as a database 2 according to the similarity;
step D: calculating the similarity between the Raman spectrum of the insulating oil sample at the unknown aging stage in the step C and the Raman spectrum of the sample in the database 2 according to the Euclidean distance, and selecting the Raman spectra of p known aging oil samples at the known aging stage, which are closest to the Raman spectrum of the insulating oil sample at the unknown aging stage, as a database 3 according to the similarity;
step E: judging the aging stage and the polymerization degree of the insulating oil sample at the unknown aging stage according to the value of the polymerization degree corresponding to the Raman spectrum of the sample in the database 3 and the rule of the KNN algorithm;
the step B comprises the following steps:
step B1: acquiring Raman spectrum data of the aged oil sample at each aging stage;
and step B2: the obtained Raman spectra of the aged oil samples in each aging stage are standardized,
the normalization is processed as follows
Wherein x is*Is a normalized value, xminIs the minimum value of the spectral data, xmaxIs the maximum value of the spectrum data, and x is the original value of the spectrum data;
and step B3: screening the characteristic quantity by calculating a correlation coefficient:
screening characteristics by adopting a method for calculating the correlation coefficient of the Raman intensity and the aging time at each wave number, wherein the calculation method of the correlation coefficient adopts the following calculation formula of the Pearson correlation coefficient:
in the formula, rho is a Pearson correlation coefficient between the Raman spectrum data of each waveband of the aged oil sample in the known aging stage and the corresponding aging time, x (y) is the Raman spectrum intensity x (i) of a certain wavenumber when the aging time is y, n is the number of wavebands, and y (i) is the aging time corresponding to the ith waveband;
selecting a wave band with the absolute value of the correlation coefficient being greater than or equal to 0.5 as an information wave band for subsequent analysis; establishing a sample Raman spectrum database 1;
the step C comprises the following steps:
step C1: acquiring Raman spectrum data of an insulating oil sample at an unknown aging stage;
and step C2: carrying out standardization processing on the obtained Raman spectrum data;
and C3: selecting standardized Raman spectrum data corresponding to the information wave band screened in the step B;
and C4: calculating the correlation coefficient rho of the Raman spectrum of the insulating oil sample at the unknown aging stage selected by C3 and the sample spectrum in the sample Raman spectrum database 1 established in the step B according to the Pearson correlation coefficientp：
Where rhopFor the Pearson correlation coefficient, y, between the Raman spectrum of the insulating oil sample at the unknown aging stage and the spectrum of the sample in database 1 established in step BTestIs a characteristic value, y, of the Raman spectrum of the insulating oil sample at an unknown aging stageRefIs the characteristic value, n, of the sample spectrum in database 1bFor the information band obtained in step BThe number is the total number of wave bands corresponding to the sample spectrum in the sample Raman spectrum database 1;
and C5: selecting m wave band Raman spectrums closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity to construct a database 2;
the upper limit of m is 10;
the prediction accuracy was calculated using 10 interactive verifications.
2. The method for diagnosing aging of oiled paper insulation equipment according to claim 1, wherein the step D comprises:
step D1: calculating the similarity rho between the Raman spectrum of the insulating oil in each unknown aging stage after the standardization treatment in the step C and the aging oil sample in the m wave band known aging stages obtained in the step C according to the following formulae：
In the formula, yTestIs a characteristic value, y, of the Raman spectrum of the insulating oil sample at an unknown aging stageref2The characteristic value of the aged oil sample in the known aging stage is obtained;
step D2: and selecting the Raman spectrum of the aged oil sample in the known aging stage of p wave bands closest to the Raman spectrum of the insulating oil in the unknown aging stage according to the similarity, and establishing a database 3 by taking the Raman spectrum of the aged oil sample in the aging stage of the p wave bands as a sample.
3. The oil paper insulation equipment aging diagnosis method according to claim 2, characterized in that:
the upper limit of p is 10.
4. The oil paper insulation equipment aging diagnosis method according to claim 2, characterized in that:
said p is preferably 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011116736.7A CN112557834B (en) | 2020-10-19 | 2020-10-19 | Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011116736.7A CN112557834B (en) | 2020-10-19 | 2020-10-19 | Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112557834A CN112557834A (en) | 2021-03-26 |
| CN112557834B true CN112557834B (en) | 2022-11-01 |
Family
ID=75041384
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011116736.7A Active CN112557834B (en) | 2020-10-19 | 2020-10-19 | Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112557834B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113238129A (en) * | 2021-05-12 | 2021-08-10 | 国网安徽省电力有限公司电力科学研究院 | Transformer fault diagnosis method and system based on fluorescence analysis |
| CN113985218B (en) * | 2021-09-20 | 2024-04-12 | 重庆大学 | Fluorescent color-based oiled paper insulation aging diagnosis method |
| CN114034641A (en) * | 2021-12-08 | 2022-02-11 | 重庆大学 | Narrow-band filter-based Raman on-chip detection system and method |
| CN114965342A (en) * | 2022-02-16 | 2022-08-30 | 陕西星佑未来电气科技有限公司 | Insulation paper aging degree identification method based on improved linear discriminant analysis |
| CN114910754B (en) * | 2022-04-27 | 2024-10-22 | 青岛科技大学 | Method and equipment for evaluating aging state of insulating material of closed power equipment |
| CN118209537B (en) * | 2024-03-25 | 2024-09-13 | 哈尔滨理工大学 | Transformer oil paper insulation aging diagnosis method based on fusion spectrum and neural network |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105181671B (en) * | 2015-09-21 | 2018-06-26 | 中国科学院长春光学精密机械与物理研究所 | The method for testing nitride phosphor degree of aging |
| EP3384298A4 (en) * | 2015-12-01 | 2019-07-31 | General Electric Technology GmbH | An intelligent assessment method of main insulation condition of transformer oil paper insulation |
| CN106885978B (en) * | 2017-04-20 | 2019-04-16 | 重庆大学 | A kind of paper oil insulation Diagnosis of Aging based on insulating oil Raman spectrum wavelet-packet energy entropy |
| CN108872129A (en) * | 2018-06-11 | 2018-11-23 | 深圳供电局有限公司 | Partial least square method-based insulating paper near infrared spectrum analysis method |
| CN110031443A (en) * | 2019-04-15 | 2019-07-19 | 重庆大学 | A kind of portable oil paper insulation ageing state Raman spectrum diagnostic device and method |
| CN111537845A (en) * | 2020-04-26 | 2020-08-14 | 云南电网有限责任公司电力科学研究院 | Method for identifying aging state of oil paper insulation equipment based on Raman spectrum cluster analysis |
-
2020
- 2020-10-19 CN CN202011116736.7A patent/CN112557834B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112557834A (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112557834B (en) | Aging diagnosis method for oiled paper insulation equipment based on Raman spectrum | |
| Abu-Siada et al. | A novel fuzzy-logic approach for furan estimation in transformer oil | |
| Yang et al. | Identification of the aging stage of transformer oil-paper insulation via Raman spectroscopic characteristics | |
| CN109781658B (en) | Insulating paper type distinguishing method based on near infrared spectrum principal component analysis | |
| CN111537845A (en) | Method for identifying aging state of oil paper insulation equipment based on Raman spectrum cluster analysis | |
| Zhang et al. | Aging state assessment of transformer cellulosic paper insulation using multivariate chemical indicators | |
| CN110297167A (en) | A kind of transformer ageing state evaluation method based on Multi-source Information Fusion | |
| CN109324014B (en) | Self-adaptive near-infrared rapid prediction method for crude oil properties | |
| CN112147444B (en) | Power transformer working state monitoring method and system | |
| Blue et al. | Infrared detection of transformer insulation degradation due to accelerated thermal aging | |
| CN107941862A (en) | A kind of transformer oil Aging of Oil-paper Insulation in Oil diagnostic system based on frequency domain dielectric response | |
| CN111126497A (en) | Aging state evaluation method for solid insulating material of transformer | |
| CN103389430A (en) | Oil-immersed type transformer fault detection method based on Bayesian discrimination theory | |
| Malik et al. | UV/VIS response based fuzzy logic for health assessment of transformer oil | |
| Yang et al. | A novel recognition method of aging stage of transformer oil-paper insulation using Raman spectroscopic recurrence plots | |
| CN112255511A (en) | Transformer paper insulation aging evaluation method based on multi-element dissolution chemical characteristic parameters | |
| CN106570644B (en) | Statistical tool-based quantitative evaluation method for power transmission and transformation equipment | |
| Li et al. | Aging assessment of oil-impregnated-paper electrical equipment via near infrared spectroscopy powered by improved PCA-RBF-NN: Modelling and field practices | |
| Song et al. | Transformer aging diagnosis method based on Raman spectroscopy Wavelet Packet-SPCA feature extraction | |
| CN113238129A (en) | Transformer fault diagnosis method and system based on fluorescence analysis | |
| CN112257228A (en) | Method for predicting oil-paper insulation state of field casing based on fitting fingerprint database | |
| Raj et al. | Fuzzy logic-based spectroscopic analysis for condition assessment of distribution transformers | |
| CN114088660B (en) | Insulation paper water content evaluation method based on robust wavelength screening | |
| CN112082930A (en) | Method for diagnosing aging state of solid insulation of transformer | |
| CN114112961B (en) | Ultraviolet spectrum-based insulating oil aging state evaluation method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |