CN112666084A - SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection - Google Patents
SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection Download PDFInfo
- Publication number
- CN112666084A CN112666084A CN202110004035.2A CN202110004035A CN112666084A CN 112666084 A CN112666084 A CN 112666084A CN 202110004035 A CN202110004035 A CN 202110004035A CN 112666084 A CN112666084 A CN 112666084A
- Authority
- CN
- China
- Prior art keywords
- peak
- raman
- decomposition
- raman spectrum
- wavelet
- 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
Landscapes
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention relates to SF based on Raman spectrum6A noise reduction method for detecting decomposition characteristic components belongs to the technical field of power equipment fault diagnosis. The method comprises the following steps: s1, by comparing the measured SF6Performing wavelet transformation on the Raman signals of the decomposed components, and searching ridge lines to identify the position and intensity of a spectral peak; s2, acquiring half-peak width based on different forms of the signal and the noise at different decomposition scales; s3, performing least square fitting according to the obtained peak value and half peak width; and S4, checking the denoising effect. The method provided by the invention can effectively reconstruct the Raman signal submerged by strong background noise under the condition of low signal-to-noise ratio, thereby greatly improving the time resolution of the detection system. Furthermore, based on Raman spectroscopy on SF6And the detection precision of the decomposed components is higher, the accuracy is higher, the GIS equipment can be better monitored, and the insulation hidden danger in the equipment can be prevented.
Description
Technical Field
The invention belongs to the technical field of power equipment fault diagnosis, and relates to SF (sulfur hexafluoride) based on Raman spectrum6A noise reduction method for decomposition feature component detection.
Background
SF6And accurate detection of the decomposed gas products thereof are one of the keys for realizing effective diagnosis of the state of the GIS equipment. At present, SF6The detection method of the decomposed gas has the problems of cross interference, poor stability and the like, the Raman spectrum has the advantages of simultaneous detection of multi-component gas, no need of sample gas pretreatment, strong anti-interference capability and the like, and the SF based on the Raman spectrum6And the detection of the decomposed gas has better application prospect. In order to improve the time resolution of the raman spectrum detection system, it is often necessary to use a short sampling integration time, and in this case, the useful raman signal with the molecular structure vibration spectrum may be completely submerged in noise, which seriously affects the further analysis of the signal, so that it is necessary to perform noise elimination processing on the measured spectrum signal. And the traditional Raman spectrum signal denoising algorithm comprises a moving window, an average smoothing method, empirical mode decomposition and the like. Although the noise cancellation of the raman spectrum signal can be achieved to some extent, there are also respective problems. For example: the moving window average smoothing method is based on the difference between the statistical characteristics of the signal and the noise, the basic assumption is that the noise is zero-mean noise, the purpose of improving the signal-to-noise ratio is achieved by averaging the original signal, and the method is the most common method for eliminating the noise. By selecting a smooth window with the width of odd number 2 omega +1, moving the window from left to right by taking the central wavelength point k as a reference point, and replacing the measured values corresponding to the central wavelength point by the average value of all the measurements in the coverage area of the window until the smoothing of all the points is completed, wherein the width of the smooth window is odd number 2 omega +1The width omega of the smoothing window of the method influences the smoothing result, the width is too small, the smoothing effect is not good, and the width is too large, so that the characteristic peak information is smoothed out, and the spectrum distortion is caused; and there is a boundary problem. For a smooth window of width 2 ω +1, ω points at each of the left and right ends of the spectrum cannot be processed. The traditional denoising method is suitable for the condition of low noise intensity and has poor effect on Raman signals submerged in a strong noise background.
Disclosure of Invention
In view of the above, the present invention provides an SF based on Raman spectroscopy6A noise reduction method for decomposition feature component detection.
In order to achieve the purpose, the invention provides the following technical scheme:
SF based on Raman spectrum6A method of noise reduction for decomposition feature component detection, the method comprising the steps of:
s1, by comparing the measured SF6Performing wavelet transformation on the Raman signals of the decomposed components, and searching ridge lines to identify the position and intensity of a spectral peak;
s2, acquiring half-peak width based on different forms of the signal and the noise at different decomposition scales;
s3, performing least square fitting according to the obtained peak value and half peak width;
and S4, checking the denoising effect.
Optionally, step S1 specifically includes:
using MATLAB software for measured SF6Performing spectral peak identification based on ridge line extraction on a Raman spectrum curve of a decomposition component, and performing spectral peak identification by adopting Mexian Hat module continuous wavelet transform in software, wherein the definition of the wavelet is as shown in a formula (1):
wherein a is a scaling factor, b is a translation factor,
to represent
Conjugation of (a) Wf(a, b) is a second derivative proportional to f (x) convolved with a gaussian probability density function with standard deviation a;
SF6the Raman spectrum of the decomposed components is approximate to a Gaussian function form, the second derivative of the decomposed components is necessarily expressed as a local maximum value at the position of a spectral peak, the local maximum values of the same spectral peak on different scales are close, the local maximum value is enlarged along with the enlargement of the wavelet scale, the shape of a ridge is expressed through a three-dimensional graph of the wavelet coefficient, and a ridge line in a wavelet coefficient matrix is searched out;
method for extracting wavelet ridge line according to modulus information6Wavelet ridge lines extracted by Raman spectrum of the decomposed components are searched for ridge lines in a mode that a maximum value with a small size spreads to a large size; selecting a ridge line smoothly connected to the maximum scale from the obtained ridge lines, setting a threshold value by using a threshold value method, removing the part with weaker intensity, and leaving the maximum scale point of the part as the identified peak point; the spectral peaks extracted by the threshold method have false peaks caused by noise with larger intensity, and partial false peaks are removed by adopting a mode of referring to an average signal; when wavelet transform is carried out on signals at the same time, SF is subjected to6The Raman spectrum of the decomposition component adopts a boundary element replication method to carry out boundary continuation.
Optionally, step S2 specifically includes:
the half-peak width detection refers to monitoring the width of a half-peak height position, and detecting SF6A single peak of the Raman spectrum of the decomposed component, the local maximum value increasing faster with increasing scale when the scale is smaller, until the rulerMatching the degree with the spectral peak width to estimate the spectral peak width of a single peak;
for SF6And (3) when the overlapped peaks in the Raman spectrum of the decomposition components are enlarged to a certain degree by setting the scale, selecting the first or second maximum value meeting the requirement as the half-peak width.
Optionally, step S3 specifically includes:
based on extracted SF6Decomposing the position, intensity and half-peak width information of the peak in the Raman spectrum of the component, and using least square method to the SF6Reconstructing the Raman spectrum of the decomposed component;
the specific fitting steps are as follows:
s1, determining the number N of the Raman peak Gaussian fitting function as m according to the number m of the peak points determined in the peak detection and peak width detection steps;
s2, identifying the Peak position Peak [ i ] (i is 1, 2, …, N) and detecting the Half width Half width [ i ] (i is 1, 2, …, N) of each Peak according to the spectrum Peak, and creating a spectrum curve fitting expression
S3, constructing a normal equation set, and solving each coefficient to obtain the reconstructed SF6Raman spectral representation of the decomposed component.
Optionally, step S4 specifically includes:
SF establishment using Gaussview software6And its decomposition characteristic gas SO2、H2S、SOF2And SO2F2The molecular model is based on the first principle density functional theory basis, the molecules are optimized to a stable structure with the lowest energy by using Gaussian software, and the B3LYP method and the triple splitting valence-bond basis function 6-311G are selected to express the molecular orbit based on the DFT theory;
meanwhile, adding a polarization basis to add a d function to the heavy atoms, an f function to the transition metal and a p function to H;
finally, adding a diffusion function to 6-311G (df, p) to obtain 6-311G + (df, p);
calculating the vibration frequency of the molecule and the corresponding Raman activity of each vibration mode by the algorithm and the basis function to obtain a simulation signal, and finally reconstructing the reconstructed SF6And comparing the Raman spectrum expression of the decomposition component with the simulation signal to detect the denoising effect.
The invention has the beneficial effects that: the spectral denoising method based on feature extraction provided by the invention is used for measuring SF6Performing wavelet transformation on the Raman signals of the decomposed components, searching ridge lines to perform spectral peak position and intensity detection, then acquiring half-peak width based on different forms of the signals and noise in different decomposition scales, and finally performing least square fitting according to the acquired peak value and half-peak width, thereby realizing signal-noise separation of the Raman signals under the background of strong noise.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 shows Raman spectrum-based SF according to the present invention6Decomposing a noise reduction method flow chart of characteristic component detection;
FIG. 2 is a graph of SF for a calculated response6A flow chart of raman spectra of the decomposed components;
FIG. 3 is SO2A pattern of gas molecules;
FIG. 4 shows SO2Raman signature graph of the molecule;
FIG. 5 shows SO2Raw signal plot of gas raman spectrum without treatment;
FIG. 6 is a diagram of SO denoised using an algorithm2Gas raman spectroscopy.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Please refer to fig. 1-6, which illustrate a raman spectroscopy based SF6A noise reduction method for decomposition feature component detection.
As shown in FIG. 1, first, for the experimentally measured SF6Performing data processing on the Raman data of the decomposed components, namely executing a denoising algorithm introduced herein, and performing wavelet transformation on the original signal to search a ridge line, so as to obtain a peak position and a peak value of the Raman signal; then, the half-width of the Raman peak is obtained. And then using the collected peak position and peak informationRedrawing Raman spectrum signals by least square fitting to obtain SF6And decomposing the denoised signals of the components.
SF to be reconstructed6And comparing the Raman spectrum expression of the decomposition component with the simulation signal to verify the denoising effect. SF establishment using Gaussview software6And its decomposition of characteristic gases and calculation of the responsive SF6A flow chart of raman spectroscopy of the decomposed components is shown in fig. 2.
With SO2The molecular configuration diagram obtained by simulation software for gas is shown in FIG. 3, SO2The raman characteristic of the molecule is shown in figure 4. SO (SO)2The raw signal of the gas Raman spectrum without processing is shown in FIG. 5, and SO is denoised by the algorithm2The gas raman spectrum is shown in figure 6,
finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (5)
1. SF based on Raman spectrum6A noise reduction method for decomposition feature component detection is characterized in that: the method comprises the following steps:
s1, by comparing the measured SF6Performing wavelet transformation on the Raman signals of the decomposed components, and searching ridge lines to identify the position and intensity of a spectral peak;
s2, acquiring half-peak width based on different forms of the signal and the noise at different decomposition scales;
s3, performing least square fitting according to the obtained peak value and half peak width;
and S4, checking the denoising effect.
2. SF according to claim 1 based on Raman spectroscopy6Noise reduction for decomposition feature component detectionThe method is characterized in that: the step S1 specifically includes:
using MATLAB software for measured SF6Performing spectral peak identification based on ridge line extraction on a Raman spectrum curve of a decomposition component, and performing spectral peak identification by adopting Mexian Hat module continuous wavelet transform in software, wherein the definition of the wavelet is as shown in a formula (1):
wherein a is a scaling factor, b is a translation factor,
to represent
Conjugation of (a) Wf(a, b) is a second derivative proportional to f (x) convolved with a gaussian probability density function with standard deviation a;
SF6the Raman spectrum of the decomposed components is approximate to a Gaussian function form, the second derivative of the decomposed components is necessarily expressed as a local maximum value at the position of a spectral peak, the local maximum values of the same spectral peak on different scales are close, the local maximum value is enlarged along with the enlargement of the wavelet scale, the shape of a ridge is expressed through a three-dimensional graph of the wavelet coefficient, and a ridge line in a wavelet coefficient matrix is searched out;
method for extracting wavelet ridge line according to modulus information6The Raman spectrum of the decomposed components extracts the wavelet ridges, going to the large scale through the maxima having small sizesSearching ridges in a cun-spreading mode, selecting ridges smoothly connected to the largest scale from the obtained ridges, setting a threshold value by using a threshold value method, removing the part with weaker strength, and leaving the largest scale point of the part as an identified peak point; the spectral peaks extracted by the threshold method have false peaks caused by noise with larger intensity, and partial false peaks are removed by adopting a mode of referring to an average signal; when wavelet transform is carried out on signals at the same time, SF is subjected to6The Raman spectrum of the decomposition component adopts a boundary element replication method to carry out boundary continuation.
3. SF according to claim 1 based on Raman spectroscopy6A noise reduction method for decomposition feature component detection is characterized in that: the step S2 specifically includes:
the half-peak width detection refers to monitoring the width of a half-peak height position, and detecting SF6Decomposing a single peak of the Raman spectrum of the component, wherein when the scale is smaller, the local maximum value is increased rapidly along with the increase of the scale until the scale is matched with the spectrum peak width so as to estimate the spectrum peak width of the single peak;
for SF6And (3) when the overlapped peaks in the Raman spectrum of the decomposition components are enlarged to a certain degree by setting the scale, selecting the first or second maximum value meeting the requirement as the half-peak width.
4. SF according to claim 1 based on Raman spectroscopy6A noise reduction method for decomposition feature component detection is characterized in that: the step S3 specifically includes:
based on extracted SF6Decomposing the position, intensity and half-peak width information of the peak in the Raman spectrum of the component, and using least square method to the SF6Reconstructing the Raman spectrum of the decomposed component;
the specific fitting steps are as follows:
s1, determining the number N of the Raman peak Gaussian fitting function as m according to the number m of the peak points determined in the peak detection and peak width detection steps;
s2, establishing a spectral curve fitting expression according to Peak position Peak [ i ] obtained by spectral Peak identification and Half-width of each Peak [ i ] obtained by Half-width detection;
s3, constructing a normal equation set, and solving each coefficient to obtain the reconstructed SF6Raman spectral representation of the decomposed component.
5. SF according to claim 1 based on Raman spectroscopy6A noise reduction method for decomposition feature component detection is characterized in that: the step S4 specifically includes:
use of Gaussview software to establish SF6 and its decomposition signature gas SO2、H2S、SOF2And SO2F2The molecular model is based on the first principle density functional theory basis, the molecules are optimized to a stable structure with the lowest energy by using Gaussian software, and the B3LYP method and the triple splitting valence-bond basis function 6-311G are selected to express the molecular orbit based on the DFT theory;
meanwhile, adding a polarization basis to add a d function to the heavy atoms, an f function to the transition metal and a p function to H;
finally, adding a diffusion function to 6-311G (df, p) to obtain 6-311G + (df, p);
calculating the vibration frequency of the molecule and the corresponding Raman activity of each vibration mode by the algorithm and the basis function to obtain a simulation signal, and finally reconstructing the reconstructed SF6And comparing the Raman spectrum expression of the decomposition component with the simulation signal to detect the denoising effect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110004035.2A CN112666084A (en) | 2021-01-04 | 2021-01-04 | SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110004035.2A CN112666084A (en) | 2021-01-04 | 2021-01-04 | SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112666084A true CN112666084A (en) | 2021-04-16 |
Family
ID=75412674
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110004035.2A Pending CN112666084A (en) | 2021-01-04 | 2021-01-04 | SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112666084A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113533238A (en) * | 2021-09-15 | 2021-10-22 | 武汉敢为科技有限公司 | Method and system for detecting sulfur hexafluoride decomposition gas based on absorption spectrum |
| CN114166814A (en) * | 2021-05-25 | 2022-03-11 | 北京理工大学 | Raman spectrum peak identification method based on dual-scale correlation operation |
| CN114354571A (en) * | 2021-12-22 | 2022-04-15 | 中国科学院合肥物质科学研究院 | Easy-to-prepare chemical Raman characteristic peak identification method based on half-peak width and peak height |
| CN114965348A (en) * | 2022-07-27 | 2022-08-30 | 浙江数翰科技有限公司 | Spectrum analysis method and system based on sewage detection |
| CN117056671A (en) * | 2023-08-14 | 2023-11-14 | 上海如海光电科技有限公司 | EMD-based Raman spectrum noise reduction method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466908A (en) * | 2015-12-31 | 2016-04-06 | 安徽芯核防务装备技术股份有限公司 | Raman spectrum method for removing interference noise produced during sample bottle fixing |
| CN110032988A (en) * | 2019-04-23 | 2019-07-19 | 北京理工大学 | Uv raman spectroscopy system real-time noise-reducing Enhancement Method |
-
2021
- 2021-01-04 CN CN202110004035.2A patent/CN112666084A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466908A (en) * | 2015-12-31 | 2016-04-06 | 安徽芯核防务装备技术股份有限公司 | Raman spectrum method for removing interference noise produced during sample bottle fixing |
| CN110032988A (en) * | 2019-04-23 | 2019-07-19 | 北京理工大学 | Uv raman spectroscopy system real-time noise-reducing Enhancement Method |
Non-Patent Citations (2)
| Title |
|---|
| 张凯 等: "基于密度泛函的SF6 特征分解产物拉曼光谱研究", 重庆师范大学学报（自然科学版）, vol. 35, no. 2, pages 1 * |
| 范贤光 等: "基于特征提取的低信噪比拉曼光谱去噪方法研究", 光谱学与光谱分析, vol. 36, no. 12, pages 1 - 2 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114166814A (en) * | 2021-05-25 | 2022-03-11 | 北京理工大学 | Raman spectrum peak identification method based on dual-scale correlation operation |
| CN113533238A (en) * | 2021-09-15 | 2021-10-22 | 武汉敢为科技有限公司 | Method and system for detecting sulfur hexafluoride decomposition gas based on absorption spectrum |
| CN114354571A (en) * | 2021-12-22 | 2022-04-15 | 中国科学院合肥物质科学研究院 | Easy-to-prepare chemical Raman characteristic peak identification method based on half-peak width and peak height |
| CN114354571B (en) * | 2021-12-22 | 2024-05-28 | 中国科学院合肥物质科学研究院 | Method for identifying Raman characteristic peak of easily-toxic chemical based on half-peak width and peak height |
| CN114965348A (en) * | 2022-07-27 | 2022-08-30 | 浙江数翰科技有限公司 | Spectrum analysis method and system based on sewage detection |
| CN117056671A (en) * | 2023-08-14 | 2023-11-14 | 上海如海光电科技有限公司 | EMD-based Raman spectrum noise reduction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112666084A (en) | SF based on Raman spectrum6Noise reduction method for decomposition characteristic component detection | |
| Wang et al. | Blind source extraction of acoustic emission signals for rail cracks based on ensemble empirical mode decomposition and constrained independent component analysis | |
| Shi et al. | A novel fractional wavelet transform and its applications | |
| CN105411565A (en) | Heart rate variability feature classification method based on generalized scale wavelet entropy | |
| US20210364441A1 (en) | Method for improving identification accuracy of mixture components by using known mixture raman spectrum | |
| CN109932179B (en) | Rolling bearing fault detection method based on DS self-adaptive spectrum reconstruction | |
| CN105320969A (en) | A heart rate variability feature classification method based on multi-scale Renyi entropy | |
| CN105223482A (en) | The wavelet decomposition two-value denoising method of partial-discharge ultrahigh-frequency signal waveform | |
| CN104020136B (en) | Small echo intermediate value near infrared spectrum denoising method and device | |
| CN112287796B (en) | Radiation source identification method based on VMD-Teager energy operator | |
| CN105426822A (en) | Non-stable signal multi-fractal feature extraction method based on dual-tree complex wavelet transformation | |
| CN111238843A (en) | Fan health evaluation method based on rapid spectrum kurtosis analysis | |
| CN111487213A (en) | Multispectral fusion chemical oxygen demand testing method and device | |
| CN116153329A (en) | CWT-LBP-based sound signal time-frequency texture feature extraction method | |
| CN111523587A (en) | Woody plant species spectrum identification method based on machine learning | |
| CN111089856A (en) | Post-processing method for extracting Raman spectrum weak signal | |
| CN105954223A (en) | Method for improving prediction accuracy of gasoline properties | |
| CN111273126B (en) | Power distribution network topology rapid sensing method | |
| Hassanpour et al. | Adaptive segmentation using wavelet transform | |
| CN109270045A (en) | A kind of rapid fluorescence background suppression method for Raman spectrum | |
| CN113051092A (en) | Fault diagnosis method based on optimized kernel density estimation and JS divergence | |
| CN111639763B (en) | Detection model training method, detection method and device for pollution degree of hydraulic oil | |
| He et al. | Local wavelet packet decomposition of soil hyperspectral for SOM estimation | |
| CN115420726A (en) | Method for rapidly identifying target object by using reconstructed SERS spectrum | |
| CN114444006A (en) | Characterization method of spatial surface frequency characteristics |
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 |