CN113094892A - Oil concentration prediction method based on data elimination and local partial least squares - Google Patents
Oil concentration prediction method based on data elimination and local partial least squares Download PDFInfo
- Publication number
- CN113094892A CN113094892A CN202110362070.1A CN202110362070A CN113094892A CN 113094892 A CN113094892 A CN 113094892A CN 202110362070 A CN202110362070 A CN 202110362070A CN 113094892 A CN113094892 A CN 113094892A
- Authority
- CN
- China
- Prior art keywords
- data
- predicted
- modeling
- elimination
- squares
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000008030 elimination Effects 0.000 title claims abstract description 16
- 238000003379 elimination reaction Methods 0.000 title claims abstract description 16
- 239000003208 petroleum Substances 0.000 claims abstract description 12
- 238000001228 spectrum Methods 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004451 qualitative analysis Methods 0.000 claims description 5
- 238000001237 Raman spectrum Methods 0.000 abstract description 9
- 238000001514 detection method Methods 0.000 description 10
- 230000003595 spectral effect Effects 0.000 description 9
- 238000012795 verification Methods 0.000 description 6
- 238000001069 Raman spectroscopy Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 239000000126 substance Substances 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000005553 drilling Methods 0.000 description 3
- 239000012530 fluid Substances 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- 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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a petroleum concentration prediction method based on data elimination and local partial least squares, which comprises the following steps of: s1, after preprocessing and characteristic extraction are carried out on the spectrum data, calculating the Euclidean distance between the data to be predicted and the modeling data, wherein when the characteristic extraction is carried out on the spectrum data, a wavelet transform method is adopted to continuously decompose the data, a decomposed low-frequency part is selected as the characteristic of the spectrum, and then the distance between the data to be predicted and the modeling data is calculated to find a modeling data set which is close to the data to be predicted and is used as a neighbor set of the data to be predicted; and S2, calculating and comparing the correlation coefficients of the adjacent set and the data to be predicted, establishing a rejection rule, and rejecting part of modeling data and the data to be predicted. The invention adopts partial data to carry out modeling and removes the data to a certain extent, thereby completing the concentration prediction of the Raman spectrum, ensuring the precision of the prediction result within a certain range and ensuring the prediction result of the model to be more accurate.
Description
Technical Field
The invention relates to the technical field of qualitative analysis and quantitative calculation of substance concentration of petroleum, in particular to a petroleum concentration prediction method based on data elimination and local partial least squares.
Background
In recent years, with the development of technology, the oil detection technology is more advanced and more difficult, and therefore, the requirements on the oil detection technology are more and more stringent. How to change petroleum detection from traditional ground detection to underground detection, how to analyze and predict Raman spectra quickly and accurately, how to reduce time consumption and capital investment in the detection process are problems which are difficult to solve at present but need to be solved. Raman spectroscopy is recognized in more and more fields and has wider and wider application range by virtue of the advantages of simplicity and rapidness in sample analysis, no need of preparing samples in advance, obvious characteristics of obtained spectral signals and the like. The Raman spectrometer can realize rapid detection on the sample, and has small volume and convenient movement. The Raman technology is applied to underground detection of oil gas, so that the oil gas can be converted from ground detection to underground detection, the information of Raman spectrum can be fed back in time, and the time consumption and the investment of capital are reduced. However, when analyzing raman spectra, how to perform rapid analysis on spectral data, how to perform accurate qualitative analysis on a substance, and how to perform reliable quantitative calculation on a substance are the most problems to be solved at present.
Aiming at the problem that the accuracy of the prediction result of the concentration of the drilling fluid mixture is not high by the existing partial least square method, an improved algorithm for removing data and establishing a local prediction model is provided, and the Euclidean distance and the Pearson correlation coefficient are combined with the partial least square method (PLS). The method has the advantages that in consideration of the problem that when modeling is carried out on the whole data, the data inclusion of the model is large, so that the model precision is not enough, the problem is solved by calculating the Euclidean distance between the data and finding the modeling data close to the space position of the data to be predicted to establish the partial least square prediction model. And selecting partial data for modeling, so that the model is more targeted to the data. Meanwhile, the Pearson correlation coefficient aims at the problem that the accuracy of the model is interfered by data with large deviation or 'non-clustered' in the detected data, and abnormal data are removed, so that the prediction result of the model is more accurate and reliable.
Therefore, we propose a petroleum concentration prediction method based on data culling and local partial least squares to solve the above problems.
Disclosure of Invention
The invention aims to solve the problem that the accuracy of the prediction result of the concentration of a drilling fluid mixture by using the existing partial least square method is not high in the prior art, and provides an oil concentration prediction method based on data elimination and local partial least square.
A petroleum concentration prediction method based on data elimination and local partial least squares comprises the following steps:
s1, after preprocessing and characteristic extraction are carried out on the spectral data, calculating the Euclidean distance between the data to be predicted and the modeling data, wherein when the characteristic extraction is carried out on the spectral data, the dimensions are compressed by continuously decomposing the data by adopting a wavelet transform method, the main information of the data is extracted, and after the dimensions are reduced, the running time of a program is greatly shortened; selecting the decomposed low-frequency part as the characteristic of the spectrum, and then finding a modeling data set which is close to the data to be predicted by calculating the distance between the data to be predicted and the modeling data to be used as a neighbor set of the data to be predicted;
s2, calculating and comparing correlation coefficients of the adjacent set and the data to be predicted, establishing a rejection rule, and rejecting part of modeling data and the data to be predicted;
and S3, establishing a partial least square prediction model according to the residual modeling data and predicting the concentration of the data to be predicted.
Preferably, a rule for removing data is established according to the pearson correlation coefficient, and whether the data of the neighboring set meets the removal rule is judged by calculating the pearson correlation coefficient of the data of the neighboring set, so as to remove the data in the neighboring set.
Preferably, a pearson correlation coefficient between the data to be predicted and the data of the neighboring set which is not removed is calculated, and whether the data to be predicted meets a removal rule is judged, so that the data to be predicted is removed.
Preferably, a concentration prediction model is established by a partial minimum two-way method for the data of the neighboring set which are not removed, and the data to be predicted which are not removed are input into the prediction model for concentration prediction, so that qualitative analysis and quantitative calculation are realized.
Compared with the prior art, the invention has the beneficial effects that:
1. when the features of the Raman spectrum data are extracted, a wavelet transformation method is selected to decompose the spectrum data. The low frequency part of the wavelet is selected after each decomposition. And the low-frequency part of the data after L-layer decomposition is used as the characteristic of the spectral data. This not only shortens the dimensionality of the data, but also preserves the main features of the spectrum. When modeling is carried out on the characteristic variables, the overfitting phenomenon caused by overlarge data dimension is avoided, and the running time of a program is reduced.
2. When the data is selected to establish the prediction model, the data which is closer to the data to be predicted is selected to establish the prediction model according to the distance. Different from the situation that all data are selected to establish a prediction model, the selected data can represent the characteristics of the data to be predicted, and when the prediction model is established, the model is more targeted and has higher accuracy.
3. And further removing the data on the basis of selecting partial data for modeling. Through calculating the Pearson correlation coefficient, the correlation magnitude among the data is compared, the data is removed, abnormal data or data which are close in distance and have more difference in result are removed, and the like. This step makes the data more "regular" and the accuracy of the model is again improved.
4. And after the data are processed, establishing a partial least square concentration prediction model by using the residual data. When the data to be predicted is predicted, the correlation coefficients of the data to be predicted and the residual modeling data are calculated, compared and eliminated, and then the prediction and output are performed through a PLS model, so that the prediction result is closer to the real concentration value, and the error is ensured to be within a certain range.
Drawings
FIG. 1 is a flow chart of an algorithm of a method for predicting petroleum concentration based on data culling and local partial least squares according to the present invention;
FIG. 2 is a detailed operational flow diagram of the present invention;
FIG. 3 is a comparison graph of verification results of global modeling and local modeling according to an embodiment of the present invention;
FIG. 4 is a comparison chart of the verification results before and after data culling in the embodiment of the invention.
Detailed Description
The methods involved in the present invention will be more clearly and more fully described below with reference to the accompanying drawings in which embodiments of the invention are shown. It is obvious that the implementation examples in the following description are only a part of implementation examples of the present invention, and not all implementation examples.
Examples
Referring to fig. 1-2, a method for predicting petroleum concentration based on data elimination and local partial least squares is based on the fact that when a partial least square model is established by adopting all data for prediction at present, the data can affect each other, and the model precision is affected. Moreover, when abnormal data exists in the data, the accuracy of the model is seriously affected, and the error of the prediction result of the abnormal data is large, so that the judgment of the operator on the concentration of the substance is influenced, and serious loss is caused. Therefore, an improved algorithm capable of selecting and eliminating data is established:
firstly, spectrum preprocessing and feature extraction are carried out on data, and the preprocessing mainly comprises three parts of denoising, correcting and standardizing. In actual measurement, the Raman spectrum is influenced by emission noise, instrument noise and readout noise, and in order to avoid the problems, a polynomial moving average filtering method is adopted to carry out denoising processing on the Raman spectrum; raman spectroscopy is affected not only by noise but also by the fluorescent background of the sample, container, etc., causing the true signal to be swamped. A polynomial fitting method is chosen to solve this problem. Firstly, fitting background information through a polynomial according to a spectrum signal, subtracting the background from original data, and finally obtaining an analysis spectrum; the influence of dimension and magnitude caused by laser intensity in a Raman spectrum is removed by adopting z-score standardization; and (3) performing feature extraction on the spectral data by adopting a wavelet transform method, and extracting 62-dimensional variables from 1961-dimensional variables for subsequent analysis.
In order to find a data set close to the data to be predicted in the modeling data, a local prediction model is built. By calculating the distance between the data to be predicted and the modeling data, and giving a distance threshold tau, a set of modeling data with the distance from the modeling data smaller than the threshold tau, called a neighbor set, is found from the modeling data. The present invention uses euclidean distances to find a neighbor set. Euclidean distance is the most common method for measuring the distance between two groups of high-dimensional data in space, and the formula is as follows:
wherein X ═ X1,x2,…,xn]，Y＝[y1,y2,…,yn]Representing two sets of data in a high-dimensional space, respectively.
And according to the found adjacent set, establishing a rule for removing the data by comparing the correlation or the similarity between the data, and removing the data. The invention adopts the Pearson correlation coefficient to establish the elimination rule. The calculation formula is as follows:
wherein:
The culling rules are given as follows:
the method comprises the following steps: calculating a correlation coefficient between Raman spectrum data according to the formula (2);
step two: given a minimum value p of the correlation coefficientminComparing the correlation coefficients obtained by the previous step, and when the correlation coefficients are smaller than the minimum value rhominWhen the correlation degree between the two variables is low, the correlation degree is marked as 0;
step three: setting a proportional value delta, judging the proportional value of the number of 0 in the data in the number of the samples in the proximity set, and when the proportional value is greater than delta, determining that the data needs to be removed.
And finally, establishing a partial least square prediction model by using the data reserved in the steps to realize qualitative analysis and quantitative calculation of the data to be measured.
The data of the invention is spectral data obtained by performing Raman laser irradiation on a petroleum drilling fluid mixture prepared in a laboratory, the concentration of the spectral data respectively comprises 0%, 5%, 10% and 15%, each concentration comprises 40 groups of data, and the length of each data is 1961 dimension. These 40 sets of data were divided into two parts, one part 30 as modeling data and the remaining 10 as verification data. When the spectral data is subjected to feature extraction, 5 layers are selected as the number of decomposition layers. The maximum value of the euclidean distance is chosen to be 20. In the elimination rule, the selection of the correlation coefficient is lower than 0.8, and the data with the proportion of more than 0.5 is eliminated.
Referring to fig. 3, for comparison, fig. 3 shows a comparison of the validation results of the pre-processing and wavelet transformation of data to create a PLS prediction model and a local PLS prediction model based on a neighbor set. FIG. 4 is a graph showing a comparison of the results of verification of a local PLS model before data removal and a local PLS model after data removal. It can be seen from the comparison of fig. 3 that when a prediction model is built by using data of all concentrations, the error of the result is large, and several points are seriously deviated from the true value, while in the result of building a local prediction model, the predicted values are relatively close to the true concentration value. Intuitively, the local PLS prediction model has better fitting degree and higher precision. In order to prove that the model has universality, the data is disturbed, and new modeling data and new verification data are reselected.
Referring to fig. 4, the result of fig. 4 is the verification result after regrouping, and it can be seen from fig. 4 that the true concentration is 5% at the 11 th point, but the predicted result is 10%. By finding the adjacent set of the data to be predicted, the data concentration in the adjacent set of the data to be predicted is respectively 5%, 5% and 15%, so that the judgment of the similarity of the data by means of the distance alone is not accurate enough. Therefore, a rule to cull data is added. Through comparison, the elimination rule can eliminate data which are close to each other but have more different results, so that the accuracy of the model is further improved.
For more obvious comparison, the invention selects the Mean Absolute Error (MAE), the mean square error (RMSE) and the goodness of fit R2The three performance indicators were evaluated for the predicted results, which are shown in tables 1-2 below:
TABLE 1 comparison of Global modeling and local modeling
TABLE 2 comparison results before and after data culling
As can be seen from the results in Table 1, the overall performance of the modeling by selecting part of the data is better than that of the modeling result of all the data, R2The local modeling precision is better as the improvement is about 0.01, the RMSE and the MAE are reduced by about 0.003. Comparing the results in table 2, it can be seen that when abnormal data occurs in the data, the calculation result of the performance index is seriously affected, and a prediction model is established after the data is removed, so that the model precision is further improved, and the result is more reliable.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (4)
1. A petroleum concentration prediction method based on data elimination and local partial least squares is characterized by comprising the following steps:
s1, after preprocessing and characteristic extraction are carried out on the spectrum data, calculating the Euclidean distance between the data to be predicted and the modeling data, wherein when the characteristic extraction is carried out on the spectrum data, a wavelet transform method is adopted to continuously decompose the data, a decomposed low-frequency part is selected as the characteristic of the spectrum, and then the distance between the data to be predicted and the modeling data is calculated to find a modeling data set which is close to the data to be predicted and is used as a neighbor set of the data to be predicted;
s2, calculating and comparing correlation coefficients of the adjacent set and the data to be predicted, establishing a rejection rule, and rejecting part of modeling data and the data to be predicted;
and S3, establishing a partial least square prediction model according to the residual modeling data and predicting the concentration of the data to be predicted.
2. The method for predicting the oil concentration based on the data elimination and the local partial least squares as claimed in claim 1, wherein a rule for eliminating data is established according to a Pearson correlation coefficient, and the data in the neighbor set is eliminated by calculating the Pearson correlation coefficient of the data in the neighbor set to judge whether the data in the neighbor set meets the elimination rule.
3. The method for predicting the petroleum concentration based on data elimination and local partial least squares as claimed in claim 2, characterized in that a pearson correlation coefficient between the data to be predicted and the data of the neighboring set which is not eliminated is calculated, whether the data to be predicted meets elimination rules is judged, and therefore the data to be predicted is eliminated.
4. The method for predicting the petroleum concentration based on the data elimination and the local partial least squares as claimed in claim 3, wherein a concentration prediction model is built on the data of the adjacent set which is not eliminated through a partial minimum two method, and the data to be predicted which is not eliminated is input into the prediction model for concentration prediction, so that qualitative analysis and quantitative calculation are realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110362070.1A CN113094892A (en) | 2021-04-02 | 2021-04-02 | Oil concentration prediction method based on data elimination and local partial least squares |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110362070.1A CN113094892A (en) | 2021-04-02 | 2021-04-02 | Oil concentration prediction method based on data elimination and local partial least squares |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113094892A true CN113094892A (en) | 2021-07-09 |
Family
ID=76673586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110362070.1A Pending CN113094892A (en) | 2021-04-02 | 2021-04-02 | Oil concentration prediction method based on data elimination and local partial least squares |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113094892A (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5892228A (en) * | 1996-09-30 | 1999-04-06 | Ashland Inc. | Process and apparatus for octane numbers and reid vapor pressure by Raman spectroscopy |
| CN105117525A (en) * | 2015-07-31 | 2015-12-02 | 天津工业大学 | Bagging extreme learning machine integrated modeling method |
| CN106442402A (en) * | 2016-11-07 | 2017-02-22 | 江南大学 | Method for rapidly detecting content of calcium element in oil sand on basis of near infrared spectrum |
| CN106596456A (en) * | 2016-04-01 | 2017-04-26 | 南京理工大学 | Changeable moving window based selection method of near infrared spectral region of solution |
| US20180017540A1 (en) * | 2016-07-14 | 2018-01-18 | Chevron U.S.A. Inc. | Method for predicting total petroleum hydrocarbon concentration in soils |
| CN107727676A (en) * | 2017-09-14 | 2018-02-23 | 三峡大学 | A kind of heavy metal content in soil modeling method based on to space before partial least squares algorithm |
| CN107748146A (en) * | 2017-10-20 | 2018-03-02 | 华东理工大学 | A kind of crude oil attribute method for quick predicting based near infrared spectrum detection |
| CN108982409A (en) * | 2018-08-08 | 2018-12-11 | 浙江工业大学 | A method of quickly detecting three constituent content of kelp lignocellulosic based near infrared spectrum |
| CN109324014A (en) * | 2018-10-08 | 2019-02-12 | 华东理工大学 | A kind of adaptive oil property near-infrared method for quick predicting |
| CN109374565A (en) * | 2018-09-30 | 2019-02-22 | 华东交通大学 | A kind of methanol gasoline ethanol petrol differentiates and content assaying method |
| CN109916849A (en) * | 2019-04-04 | 2019-06-21 | 新疆大学 | Method based near infrared spectrum correlation analysis test sample physicochemical property |
| CN110687072A (en) * | 2019-10-17 | 2020-01-14 | 山东大学 | Calibration set and verification set selection and modeling method based on spectral similarity |
| US20210020276A1 (en) * | 2018-04-05 | 2021-01-21 | Inesc Tec - Instttuto De Engenharia De Sistemas, Tecnologia E Ciencia | Spectrophotometry method and device for predicting a quantification of a constituent from a sample |
| CN112285056A (en) * | 2020-10-14 | 2021-01-29 | 山东大学 | Method for selecting and modeling personalized correction set of spectrum sample |
| CN112304922A (en) * | 2020-10-29 | 2021-02-02 | 辽宁石油化工大学 | Method for quantitatively analyzing crude oil by Raman spectrum based on partial least square method |
-
2021
- 2021-04-02 CN CN202110362070.1A patent/CN113094892A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5892228A (en) * | 1996-09-30 | 1999-04-06 | Ashland Inc. | Process and apparatus for octane numbers and reid vapor pressure by Raman spectroscopy |
| CN105117525A (en) * | 2015-07-31 | 2015-12-02 | 天津工业大学 | Bagging extreme learning machine integrated modeling method |
| CN106596456A (en) * | 2016-04-01 | 2017-04-26 | 南京理工大学 | Changeable moving window based selection method of near infrared spectral region of solution |
| US20180017540A1 (en) * | 2016-07-14 | 2018-01-18 | Chevron U.S.A. Inc. | Method for predicting total petroleum hydrocarbon concentration in soils |
| CN106442402A (en) * | 2016-11-07 | 2017-02-22 | 江南大学 | Method for rapidly detecting content of calcium element in oil sand on basis of near infrared spectrum |
| CN107727676A (en) * | 2017-09-14 | 2018-02-23 | 三峡大学 | A kind of heavy metal content in soil modeling method based on to space before partial least squares algorithm |
| CN107748146A (en) * | 2017-10-20 | 2018-03-02 | 华东理工大学 | A kind of crude oil attribute method for quick predicting based near infrared spectrum detection |
| US20210020276A1 (en) * | 2018-04-05 | 2021-01-21 | Inesc Tec - Instttuto De Engenharia De Sistemas, Tecnologia E Ciencia | Spectrophotometry method and device for predicting a quantification of a constituent from a sample |
| CN108982409A (en) * | 2018-08-08 | 2018-12-11 | 浙江工业大学 | A method of quickly detecting three constituent content of kelp lignocellulosic based near infrared spectrum |
| CN109374565A (en) * | 2018-09-30 | 2019-02-22 | 华东交通大学 | A kind of methanol gasoline ethanol petrol differentiates and content assaying method |
| CN109324014A (en) * | 2018-10-08 | 2019-02-12 | 华东理工大学 | A kind of adaptive oil property near-infrared method for quick predicting |
| CN109916849A (en) * | 2019-04-04 | 2019-06-21 | 新疆大学 | Method based near infrared spectrum correlation analysis test sample physicochemical property |
| CN110687072A (en) * | 2019-10-17 | 2020-01-14 | 山东大学 | Calibration set and verification set selection and modeling method based on spectral similarity |
| CN112285056A (en) * | 2020-10-14 | 2021-01-29 | 山东大学 | Method for selecting and modeling personalized correction set of spectrum sample |
| CN112304922A (en) * | 2020-10-29 | 2021-02-02 | 辽宁石油化工大学 | Method for quantitatively analyzing crude oil by Raman spectrum based on partial least square method |
Non-Patent Citations (4)
| Title |
|---|
| GUIJUN YANG ET AL.: "Applying wavelet frequency component correlative selection in Raman spectral analysis", 《2009 2ND IEEE INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND INFORMATION TECHNOLOGY》 * |
| ISSAM BARRA ET AL.: "Predicting cetane number in diesel fuels using FTIR spectroscopy and PLS regression", 《VIBRATIONAL SPECTROSCOPY》 * |
| 吴波等: "基于光谱维小波特征的混合像元投影迭代分解", 《电子学报》 * |
| 石鲁珍等: "马氏距离与浓度残差剔除近红外异常样品研究", 《中国农机化学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107179310B (en) | Raman spectrum characteristic peak recognition methods based on robust noise variance evaluation | |
| CN108169213B (en) | Automatic identification method for peak elements of laser-induced breakdown spectroscopy | |
| CN109324013B (en) | Near-infrared rapid analysis method for constructing crude oil property by using Gaussian process regression model | |
| WO2021232757A1 (en) | Method for improving mixture component identification precision by using raman spectra of known mixtures | |
| CN113340874B (en) | Quantitative analysis method based on combination ridge regression and recursive feature elimination | |
| CN106918567B (en) | A kind of method and apparatus measuring trace metal ion concentration | |
| CN103487411A (en) | Method for recognizing steel grade by combining random forest algorithm with laser-induced breakdown spectroscopy | |
| CN102072767A (en) | Wavelength similarity consensus regression-based infrared spectrum quantitative analysis method and device | |
| CN111089856A (en) | Post-processing method for extracting Raman spectrum weak signal | |
| CN115728259A (en) | Multi-component gas analysis method | |
| CN109283153B (en) | Method for establishing quantitative analysis model of soy sauce | |
| CN110987866A (en) | Gasoline property evaluation method and device | |
| CN114878544B (en) | Method for identifying target component from mixture SERS spectrum | |
| CN112666104A (en) | DOAS-based gas concentration inversion method | |
| CN114611582B (en) | Method and system for analyzing substance concentration based on near infrared spectrum technology | |
| US20150112643A1 (en) | Infra-red analysis of diamonds | |
| CN108663334B (en) | Method for searching spectral characteristic wavelength of soil nutrient based on multi-classifier fusion | |
| CN114330411A (en) | Self-adaptive windowed Raman spectrum identification method based on similarity | |
| CN116858822A (en) | Quantitative analysis method for sulfadiazine in water based on machine learning and Raman spectrum | |
| CN113094892A (en) | Oil concentration prediction method based on data elimination and local partial least squares | |
| CN114414524A (en) | Method for rapidly detecting properties of aviation kerosene | |
| CN113791062A (en) | Method for judging fixed substance type based on Raman spectrum | |
| CN111504908A (en) | Rock type identification method and system based on photoacoustic spectroscopy | |
| CN112304918A (en) | Method and device for identifying mixture based on Raman spectrum and Raman spectrum detection equipment | |
| CN115420715B (en) | Laser-induced breakdown spectroscopy abnormal data eliminating method based on spectral feature fusion |
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 |
Application publication date: 20210709 |
|
| RJ01 | Rejection of invention patent application after publication |