CN109520962A - A kind of grape wine near infrared spectrum detection method - Google Patents
A kind of grape wine near infrared spectrum detection method Download PDFInfo
- Publication number
- CN109520962A CN109520962A CN201710845125.8A CN201710845125A CN109520962A CN 109520962 A CN109520962 A CN 109520962A CN 201710845125 A CN201710845125 A CN 201710845125A CN 109520962 A CN109520962 A CN 109520962A
- Authority
- CN
- China
- Prior art keywords
- sample
- near infrared
- wine
- model
- variable
- 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
- 235000014101 wine Nutrition 0.000 title claims abstract description 77
- 235000009754 Vitis X bourquina Nutrition 0.000 title claims abstract description 41
- 235000012333 Vitis X labruscana Nutrition 0.000 title claims abstract description 41
- 235000014787 Vitis vinifera Nutrition 0.000 title claims abstract description 41
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 title claims abstract description 21
- 240000006365 Vitis vinifera Species 0.000 title 1
- 238000000034 method Methods 0.000 claims abstract description 45
- 241000219095 Vitis Species 0.000 claims abstract description 40
- 230000001476 alcoholic effect Effects 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012937 correction Methods 0.000 claims abstract description 8
- 238000005259 measurement Methods 0.000 claims abstract description 8
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims abstract description 7
- 238000001228 spectrum Methods 0.000 claims description 14
- 230000003595 spectral effect Effects 0.000 claims description 9
- 238000002474 experimental method Methods 0.000 claims description 7
- 239000000126 substance Substances 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000004566 IR spectroscopy Methods 0.000 claims description 6
- 238000012821 model calculation Methods 0.000 claims description 4
- ISWSIDIOOBJBQZ-UHFFFAOYSA-N Phenol Chemical compound OC1=CC=CC=C1 ISWSIDIOOBJBQZ-UHFFFAOYSA-N 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000002835 absorbance Methods 0.000 claims description 2
- 230000000717 retained effect Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 4
- 238000004587 chromatography analysis Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 150000001720 carbohydrates Chemical class 0.000 description 2
- 239000010453 quartz Substances 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 241000143252 Idaea infirmaria Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000855 fermentation Methods 0.000 description 1
- 230000004151 fermentation Effects 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 239000003205 fragrance Substances 0.000 description 1
- 235000019990 fruit wine Nutrition 0.000 description 1
- 125000000524 functional group Chemical group 0.000 description 1
- 238000004817 gas chromatography Methods 0.000 description 1
- 238000004128 high performance liquid chromatography Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 239000002366 mineral element Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000050 nutritive effect Effects 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 208000011117 substance-related disease Diseases 0.000 description 1
- 238000002235 transmission spectroscopy Methods 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
- 150000003722 vitamin derivatives Chemical class 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical group O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3577—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing liquids, e.g. polluted water
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention relates to a kind of grape wine near infrared spectrum detection methods, belong to grape wine quality detection technique field.Key step of the present invention: (1) using the alcoholic strength of the wine samples of alcohol meter method measurement different cultivars, the near infrared spectrum of near infrared spectrometer acquisition wine samples is used；(2) sample is randomly divided into training set and forecast set, variable is screened using MC-UVE and GA algorithm, PPLS analysis is carried out, alcoholic strength regression model is established, according to the reliability of correction root-mean-square error (RMSEC) value judgment models of model；(3) alcoholic strength of forecast set sample is predicted, calculates the correlation (R of forecast set sample predictions value and measured value2) and forecast set predicted root mean square error (RMSEP), the estimated performance of judgment models is carried out with this.The present invention combines MC-UVE, GA algorithm and PPLS analysis method, provides a kind of method of wine composition rapid quantitative detection, pretreatment is simple, easy to operate, quick nondestructive, so that on-line checking is more efficient.
Description
Technical field
The present invention relates to a kind of grape wine near infrared spectrum detection methods, belong to grape wine quality detection technique field.
Background technique
Grape wine is a kind of higher spirituosity drink of nutritive value, is drunk in right amount beneficial to body.In recent years, with me
The import volume of the raising of state's living standards of the people, grape wine also increases.The weather of different sources, picking condition, is made soil
Wine technology causes the diversification of the distinctive flavor of grape wine and varietal wine, grape wine quality be mainly reflected in composition at
Point and content on, if phenolic substances mainly determines the fragrance and color of wine, carbohydrate mainly determines alcohol content, and pectase determines first
Alcohol content etc..Therefore good grape contains higher sugar, mineral element, vitamin and pectase etc., the grape wine of fermentation
Quality is also superior.Therefore in grape wine ingredient carry out rapid quantitative detection, with meet the online Quality Detection of grape wine and
The demand of quality identification.
According to reported document it is found that wine composition rapid quantitative detection mainly utilizes infrared spectroscopy, chromatography skill
Art etc. combines chemometrics method.Quantitative detection, sample pretreatment complex steps, inspection are carried out to grape wine using chromatographic technique
Survey process is time-consuming, cannot achieve on-line quick detection.Compared with high performance liquid chromatography and gas chromatography, infrared spectrum technology
It is more fast and convenient, more it is able to satisfy the requirement of grape wine on-line quick detection.
Detection is carried out to grape wine using infrared spectrum technology and is mainly based upon in molecule chemical bond or functional group to red
The position of the absorption of outer light and intensity difference, to obtain molecular composition, content and structure of information.However, infrared spectrum technology
It is larger by external environmental interference, so including there are many noise unrelated with sample information, using red in infrared spectroscopy information
When external spectrum composes modeling entirely, the efficiency of modeling and the estimated performance of model will affect.Therefore, it is necessary to in infrared spectroscopy
Variable is screened, and selection contributes biggish variable to model, thus simplified model, raising modeling efficiency, to be grape wine
Quickly detection provides a new way.
Summary of the invention
To solve the above-mentioned problems, the present invention provides one kind to be based on Monte Carlo without information variable null method (MC-UVE)
With the near infrared spectrum Variable Selection method of genetic algorithm (GA), and further progress probability offset minimum binary (PPLS) analyze,
The method for establishing regression model promotes the rule of China's Wine Market to realize the requirement of the online rapid quantitative detection of grape wine
Plasticity safeguards the legitimate rights and interests of consumer.
The novelty of the present invention is being selected based on MC-UVE and GA algorithm near-infrared wavelength, it is different from traditional
It is modeled using all-wave length, can effectively remove external environment, instrument state etc. using the characteristic wavelength modeling selected
Interference information caused by factor, to improve the predictability of modeling efficiency and model.PPLS points are carried out to characteristic wavelength simultaneously
Regression model is established in analysis, and PPLS analyzes the probability distribution that can improve error and pivot that PLS analysis method is ignored.
The method that the present invention detects grape wine is the wine samples for first selecting several different sources, is used
National Standard Method detects its main component, while acquiring the near infrared spectrum of wine sample.Then sample is randomly divided into
Training set and forecast set, to the near infrared spectrum of training set sample using Monte Carlo without information variable null method (MC-UVE) and
Genetic algorithm (GA) carries out variables choice, filters out characteristic wavelength, carries out probability partial least squares analysis (PPLS), establishes and returns
Model.It is predicted using forecast set sample, the correlation of comparison prediction value and measured value.
To achieve the goals above, the present invention provides the following technical scheme that
The first purpose of the invention is to provide a kind of method of near infrared detection wine composition, the method is to utilize
Near infrared spectrometer acquires the near infrared spectrum of multiple wine samples, using Monte Carlo without information variable null method and heredity
Algorithm screens variable, and the characteristic wavelength selected carries out probability partial least squares analysis, establishes MC-UVE-GA-PPLS mould
Type；When detecting unknown wine samples, the infrared spectroscopy of sample to be tested is first obtained, then the light absorption value under optimal variable is inputted
MC-UVE-GA-PPLS model can obtain wine samples component content to be detected.
In one embodiment of the invention, the characteristic variable is screened according to the index of stability of variable.
In one embodiment of the invention, the method is used to detect the alcoholic strength of grape wine, total phenol content or total
Sugared content.
In one embodiment of the invention, when the method is used to detect dregs of grape wine precision, selected characteristic variable
Wave number be 4018.93cm-1、4022.78cm-1、4026.64cm-1、4234.91cm-1、4238.77cm-1、4242.63cm-1。
In one embodiment of the invention, the near infrared spectrum is collected at 1000-2500nm.
In one embodiment of the invention, the near infrared spectrum passes through a kind of following preprocess method or any two
Kind preprocess method, which combines, to be pre-processed: Savitzky-Golay 9 points of smooth, multiplicative scatter corrections, baseline correction, single order
Derivative or second dervative preprocess method.
In one embodiment of the invention, described method includes following steps:
(1) select the grape wine of different sources different cultivars as experiment sample；
(2) content for measuring wine samples test substance uses the close red of near infrared spectrometer acquisition wine samples
External spectrum；
(3) sample is randomly divided into training set and forecast set, the near-infrared data of training set sample is imported in matlab,
Variable is screened using MC-UVE and GA algorithm, the characteristic wavelength selected carries out PPLS analysis, establishes test substance content
Regression model, according to the reliability of correction root-mean-square error (RMSEC) value judgment models of model；
(4) it is predicted using test substance content of the model built to forecast set sample, calculates forecast set sample predictions
Correlation (the R of value and measured value2) and forecast set predicted root mean square error (RMSEP), the prediction of judgment models is carried out with this
Performance.
(5) sample to be tested is taken, according to the near infrared spectrum number of the spectral measurement condition acquisition sample to be tested in step (2)
According to, and imported in prediction model after being pre-processed, through model calculation, the component content of unknown wine samples can be obtained.
In one embodiment of the invention, described method includes following steps:
(1) select the grape wine of different sources different cultivars as experiment sample；
(2) using the alcoholic strength of alcohol meter method measurement wine samples, wine samples are acquired using near infrared spectrometer
Near infrared spectrum；
(3) sample is randomly divided into training set and forecast set, the near-infrared data of training set sample is imported in matlab,
Variable is screened using MC-UVE and GA algorithm, the characteristic wavelength selected carries out PPLS analysis, establishes alcoholic strength and returns mould
Type, according to the reliability of correction root-mean-square error (RMSEC) value judgment models of model；
(4) it is predicted using alcoholic strength of the model built to forecast set sample, calculates forecast set sample predictions value and reality
Correlation (the R of measured value2) and forecast set predicted root mean square error (RMSEP), the estimated performance of judgment models is carried out with this.
(5) sample to be tested is taken, according to the near infrared spectrum number of the spectral measurement condition acquisition sample to be tested in step (2)
According to, and imported in prediction model after being pre-processed, through model calculation, the alcoholic strength of unknown wine samples can be obtained.
In one embodiment of the invention, in above-mentioned steps (2), selective transmission mode, when experiment, sets sample
In the quartz colorimetric utensil of 1mm, wine samples are contained to cuvette 2/3, are averaged for each sample parallel acquisition 10 times.
In one embodiment of the invention, in above-mentioned steps (3), by 117 samples be randomly divided into training set and
Forecast set (2:1), is modeled using the near infrared spectrum data of training set sample.
In one embodiment of the invention, in above-mentioned steps (5), variables choice is carried out by MC-UVE and GA
Afterwards, the information unrelated with sample is rejected.The number of run of MC-UVE is N=1000, number of principal components A=15, is stablized with variable
Index RI (reliability index) is screening foundation, and 60% with RI maximum value is threshold value, when the absolute value of the RI of variable
When greater than threshold value, which is retained.
Beneficial effect of the present invention
The screening of alcoholic strength near infrared spectrum variable and the foundation of calibration model in grape wine of the invention, to different product
Kind, growing environment, different picking condition and the made grape wine of brewing technology alcoholic strength detected.Acquire grape wine sample
The near infrared light spectrum information of product combines MC-UVE, GA algorithm and PPLS analysis method, uses MC-UVE and GA algorithm pair first
Spectral information is screened, and establishes PPLS regression model according to selected characteristic wavelength, and then to the alcoholic strength of forecast set sample
It is predicted.
The present invention quickly detects for grape wine and provides a new way.Compared with chromatographic technique, infrared spectrum technology is more
Add easy to be quick；Compared with traditional modeling using full spectrum, after Variable Selection, garbage can be rejected, simplify mould
Type improves modeling efficiency and model stability.
Detailed description of the invention
Fig. 1: 20 wine sample NIR transmittance spectroscopy figures are randomly selected；
Fig. 2: it is distributed using stability of the MC-UVE algorithm to the variable that alcoholic strength is predicted；
Fig. 3: Quan Guangpu PLS model result forecast set alcoholic strength measured value and predicted value；
Fig. 4: MC-UVE-GA-PPLS model result forecast set alcoholic strength measured value and predicted value.
Specific embodiment
Selected grape wine meets GB15037-2006 " grape wine " requirement.
Embodiment 1: the foundation and detection of near infrared spectrum detection dregs of grape wine precision model
1) select 117 grape wine of different sources different cultivars as experimental study object, sample sealed storage is in 4 DEG C
In refrigerator, sample is placed at 25 DEG C after 2h before experiment and is tested.
2) sample alcoholic strength content:
Using the alcoholic strength of alcohol meter method measurement wine samples, alcohol meter method is according to GB15038-2006 " grape wine, fruit
Wine universaling analysis method " standard is measured.Each sample is measured in parallel twice, is averaged, specific data are as shown in table 1.
1 wine samples alcoholic strength reference value of table
3) grape wine near infrared spectrum:
Using the near infrared spectrum of near infrared spectrometer acquisition wine samples, booting preheating 1h before testing；Selective transmission
Mode carries out near infrared spectra collection to sample, and sample is placed in the quartz colorimetric utensil of 1mm by when experiment, and wine samples contain
To cuvette 2/3, it is averaged for each sample parallel acquisition 10 times；20 wine samples randomly selected it is close red
Outer transmitted light spectrogram as shown in Figure 1, it can be seen from the figure that the grape wine of different sources have similar near ir absorption peaks,
Wherein, in 4996cm-1The absorption peak at place, absorbance illustrate that the near infrared light for penetrating sample at this time is very faint, easily close to 3
It is influenced by external environments such as noises, therefore it should be rejected in analysis.6862cm-1、4996cm-1Stronger suction at two
Receive peak respectively to the frequency multiplication of-OH of carbohydrate and water group in grape wine and related, the 4300-4500cm of sum of fundamental frequencies absorption-1Wave
Small peak in section is mainly related with the sum of fundamental frequencies absorption of the c h bond of each component in sample, and 5250-6000cm-1Peak be then and C-H
Two frequencys multiplication of key absorb related.
4) the alcoholic strength regression model for combining PLS analysis method to establish using full spectrum
It uses full spectral limit to carry out PLS analysis to after step 3) Pretreated spectra, establishes regression model.It can from table 2
Out, the performance of model is influenced less after spectrum is pre-processed, the quantitative model R established using original spectrum2With RMSEC points
Not Wei 0.997 and 0.053, therefore be further analyzed using original spectral data.
The alcoholic strength PLS model that table 2 is established using different preprocess methods
5) variables choice based on MC-UVE method
117 samples are randomly divided into training set and forecast set (2:1), use the near infrared spectrum data of training set sample
It is modeled, the full spectroscopic data of collected near-infrared is imported in matlab, pass through each of MC-UVE alcoholic strength obtained
The stability value of variable is as shown in Figure 2.Dotted line indicates threshold value, and absolute value is selected to be greater than the variable of threshold value for modeling.For MC-
UVE algorithm selects the quantity of variable very crucial, if variable negligible amounts, useful information can be ignored；If by uncorrelated
Variable include, it will influence the predictability of model.Calibration model RMSEC calculation method is 10 variables of every increase, then
It is recalculated.Meanwhile as the hidden variable of GA-PLS mode input (LVs), number is set as 1 to 10, according to R2 preReally
Fixed optimal LV value.When LV value is 4, highest R is obtained2 pre.The MC-UVE-PLS model established using 29 variables
Prediction result is as shown in table 3, compared with entire spectrum PLS model, R2 preSlightly it is reduced to 0.942, RMSEP 0.220.
6) variables choice based on GA algorithm
After MC-UVE selection, variable number is reduced to 29, in order to be further simplified to model, by MC-
Input of 29 variables that UVE is selected as GA algorithm.Spectral variables selection result after GA operation is as shown in table 3, by institute
The spectral variables number of selection establishes PLS calibration model (MC-UVE-GA-PLS), and is predicted with unknown sample, R2 preAnd
RMSEP is shown in Table 3.Meanwhile PPLS analysis is carried out to based on full spectrum data set and by the variable of MC-UVE-GA algorithms selection,
MC-UVE-GA-PPLS model is established, selecting number of wavelengths is 6, and the parameter of model is R2 pre=0.946, RMSEP=0.215 are high
In the parameter of MC-UVE-GA-PLS model.Show that probability partial least squares analysis method can be improved the performance of model, 6 waves
Length is considered as the optimal variable as PPLS mode input, is 4018.93cm respectively-1、4022.78cm-1、4026.64cm-1、
4234.91cm-1、4238.77cm-1、4242.63cm-1。
7) characteristic wavelength selected in step 6) is subjected to PPLS analysis, alcoholic strength regression model is established, according to model
Correct the reliability of root-mean-square error (RMSEC) value judgment models；Compared with the PLS model based on full spectrum, MC-UVE-GA-
The R of PPLS model2 preDeclined, RMSEP is risen, but variable used in MC-UVE-GA-PPLS model is far below complete
PLS model is composed, this will be helpful to Simplified prediction model, realize the requirement that grape wine quickly detects.Fig. 3 shows two models
The scatter plot of prediction result, it depicts the distribution of predicted value and measured value: good linear dependence is presented in they.
The alcoholic strength forecast set of the different models of table 3
Although the present invention has been described by way of example and in terms of the preferred embodiments, it is not intended to limit the invention, any to be familiar with this skill
The people of art can do various change and modification, therefore protection model of the invention without departing from the spirit and scope of the present invention
Enclosing subject to the definition of the claims.
Claims (10)
1. a kind of method of near infrared detection wine composition, which is characterized in that the method is adopted using near infrared spectrometer
The near infrared spectrum for collecting multiple wine samples, after being pre-processed to spectrum, using Monte Carlo without information variable null method
Variable is screened with genetic algorithm, the characteristic wavelength selected carries out probability partial least squares analysis, establishes MC-UVE-GA-
PPLS model；When detecting unknown wine samples, infrared spectroscopy detection is first carried out, then the absorbance under characteristic variable is imported
MC-UVE-GA-PPLS model can obtain wine samples component content to be detected.
2. method according to claim 1, which is characterized in that the characteristic variable is sieved according to the index of stability of variable
Choosing.
3. method according to claim 1, which is characterized in that the method is used to detect the alcoholic strength of grape wine, total phenol contains
Amount or total sugar content.
4. method according to claim 3, which is characterized in that when the method is used to detect dregs of grape wine precision, selected spy
The wave number for levying variable is 4018.93cm-1、4022.78cm-1、4026.64cm-1、4234.91cm-1、4238.77cm-1、
4242.63cm-1。
5. method according to claim 1, which is characterized in that the near infrared spectrum is acquired at 1000-2500nm
It arrives.
6. method according to claim 1, which is characterized in that the near infrared spectrum by a kind of following preprocess method or
Any two kinds of preprocess methods are combined and are pre-processed: Savitzky-Golay 9 points of smooth, multiplicative scatter corrections, baseline school
Just, first derivative or second dervative preprocess method.
7. method according to claim 1, which is characterized in that described method includes following steps:
(1) select the grape wine of different sources different cultivars as experiment sample；
(2) content for measuring the test substance of wine samples uses the near-infrared of near infrared spectrometer acquisition wine samples
Spectrum, and spectrum is pre-processed；
(3) sample is randomly divided into training set and forecast set, the near-infrared data of training set sample is imported in matlab, are used
MC-UVE and GA algorithm screens variable, and the characteristic wavelength selected carries out PPLS analysis, establishes returning for test substance content
Return model, according to the reliability of the correction root-mean-square error value judgment models of model；
(4) predicted using test substance content of the model built to forecast set sample, calculate forecast set sample predictions value and
The correlation of measured value and the predicted root mean square error of forecast set, the estimated performance of judgment models is carried out with this；
(5) sample to be tested is taken, according to the near infrared spectrum data of the spectral measurement condition acquisition sample to be tested in step (2), and
It is imported in prediction model after being pre-processed, through model calculation, the component content of unknown wine samples can be obtained.
8. method according to claim 7, which is characterized in that described method includes following steps:
(1) select the grape wine of different sources different cultivars as experiment sample；
(2) using the alcoholic strength of alcohol meter method measurement wine samples, the close of near infrared spectrometer acquisition wine samples is used
Infrared spectroscopy, and spectrum is pre-processed；
(3) sample is randomly divided into training set and forecast set, the near-infrared data of training set sample is imported in matlab, are used
MC-UVE and GA algorithm screens variable, and the characteristic wavelength selected carries out PPLS analysis, establishes alcoholic strength regression model, root
According to the correction root-mean-square error value of model, the reliability of judgment models；
(4) it is predicted using alcoholic strength of the model built to forecast set sample, calculates forecast set sample predictions value and measured value
Correlation and forecast set predicted root mean square error, the estimated performance of judgment models is carried out with this；
(5) sample to be tested is taken, according to the near infrared spectrum data of the spectral measurement condition acquisition sample to be tested in step (2), and
It is imported in prediction model after being pre-processed, through model calculation, the alcoholic strength of unknown wine samples can be obtained.
9. method according to claim 8, which is characterized in that be to be randomly divided into 117 samples according to 2:1 in the method
Training set and forecast set are modeled using the near infrared spectrum data of training set sample.
10. method according to claim 8, which is characterized in that the number of run of the MC-UVE is N=1000, principal component
Number A=15 is screening foundation with variable index of stability RI, and 60% with RI maximum value is threshold value, when the absolute value of the RI of variable
When greater than threshold value, which is retained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710845125.8A CN109520962A (en) | 2017-09-19 | 2017-09-19 | A kind of grape wine near infrared spectrum detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710845125.8A CN109520962A (en) | 2017-09-19 | 2017-09-19 | A kind of grape wine near infrared spectrum detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109520962A true CN109520962A (en) | 2019-03-26 |
Family
ID=65769000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710845125.8A Pending CN109520962A (en) | 2017-09-19 | 2017-09-19 | A kind of grape wine near infrared spectrum detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109520962A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109916849A (en) * | 2019-04-04 | 2019-06-21 | 新疆大学 | Method based near infrared spectrum correlation analysis test sample physicochemical property |
| CN111027025A (en) * | 2019-12-11 | 2020-04-17 | 华北电力大学（保定） | Method for screening infrared spectrum characteristic wavelength to predict wine quality parameters |
| CN111504944A (en) * | 2020-06-02 | 2020-08-07 | 江南大学 | Statistical monitoring method of citric acid fermentation liquefied clear liquid based on near infrared spectrum |
| CN112697745A (en) * | 2021-01-20 | 2021-04-23 | 西安电子科技大学 | Method for measuring alcohol content of white spirit |
| CN113361610A (en) * | 2021-06-10 | 2021-09-07 | 北方民族大学 | Intelligent identification method and system for wine production place |
| CN113376106A (en) * | 2021-06-04 | 2021-09-10 | 深圳技术大学 | Method for measuring alcoholic strength |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102525481A (en) * | 2011-12-14 | 2012-07-04 | 山东大学 | Detection method and system for alcohol content in human body on the basis of near infrared spectrum |
| CN105911017A (en) * | 2016-04-12 | 2016-08-31 | 中华人民共和国张家港出入境检验检疫局 | Detection method for simultaneously and rapidly determining five indexes in wine |
-
2017
- 2017-09-19 CN CN201710845125.8A patent/CN109520962A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102525481A (en) * | 2011-12-14 | 2012-07-04 | 山东大学 | Detection method and system for alcohol content in human body on the basis of near infrared spectrum |
| CN105911017A (en) * | 2016-04-12 | 2016-08-31 | 中华人民共和国张家港出入境检验检疫局 | Detection method for simultaneously and rapidly determining five indexes in wine |
Non-Patent Citations (3)
| Title |
|---|
| BAO YANFEI ET AL.: "NIR Detection of Alcohol Content Based on GA-PLS", 《APPLIED MECHANICS AND MATERIALS》 * |
| DENGFEI JIE ET AL.: "Variable selection for partial least squares analysis of soluble solids content in watermelon using near-infrared diffuse transmission technique", 《JOURNAL OF FOOD ENGINEERING》 * |
| SHUO LI AND JEAN GAO: "Probabilistic Partial Least Square Regression: A Robust Model for Quantitative Analysis of Raman Spectroscopy Data", 《2011 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109916849A (en) * | 2019-04-04 | 2019-06-21 | 新疆大学 | Method based near infrared spectrum correlation analysis test sample physicochemical property |
| CN111027025A (en) * | 2019-12-11 | 2020-04-17 | 华北电力大学（保定） | Method for screening infrared spectrum characteristic wavelength to predict wine quality parameters |
| CN111504944A (en) * | 2020-06-02 | 2020-08-07 | 江南大学 | Statistical monitoring method of citric acid fermentation liquefied clear liquid based on near infrared spectrum |
| CN112697745A (en) * | 2021-01-20 | 2021-04-23 | 西安电子科技大学 | Method for measuring alcohol content of white spirit |
| CN113376106A (en) * | 2021-06-04 | 2021-09-10 | 深圳技术大学 | Method for measuring alcoholic strength |
| CN113361610A (en) * | 2021-06-10 | 2021-09-07 | 北方民族大学 | Intelligent identification method and system for wine production place |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109520962A (en) | A kind of grape wine near infrared spectrum detection method | |
| Hu et al. | Rapid detection of three quality parameters and classification of wine based on Vis-NIR spectroscopy with wavelength selection by ACO and CARS algorithms | |
| Xie et al. | Prediction of titratable acidity, malic acid, and citric acid in bayberry fruit by near-infrared spectroscopy | |
| Barnaba et al. | Portable NIR‐AOTF spectroscopy combined with winery FTIR spectroscopy for an easy, rapid, in‐field monitoring of Sangiovese grape quality | |
| Calvini et al. | Toward the development of combined artificial sensing systems for food quality evaluation: A review on the application of data fusion of electronic noses, electronic tongues and electronic eyes | |
| Xu et al. | Variable selection in visible and near-infrared spectra: Application to on-line determination of sugar content in pears | |
| Cuadrado et al. | Comparison and joint use of near infrared spectroscopy and Fourier transform mid infrared spectroscopy for the determination of wine parameters | |
| Lin et al. | Theory and application of near infrared spectroscopy in assessment of fruit quality: a review | |
| Dambergs et al. | A review of the state of the art, limitations, and perspectives of infrared spectroscopy for the analysis of wine grapes, must, and grapevine tissue | |
| Ríos-Reina et al. | Spectralprint techniques for wine and vinegar characterization, authentication and quality control: Advances and projections | |
| Cozzolino et al. | Measurement of condensed tannins and dry matter in red grape homogenates using near infrared spectroscopy and partial least squares | |
| Dambergs et al. | Rapid measurement of methyl cellulose precipitable tannins using ultraviolet spectroscopy with chemometrics: Application to red wine and inter-laboratory calibration transfer | |
| Li et al. | Nondestructive measurement and fingerprint analysis of soluble solid content of tea soft drink based on Vis/NIR spectroscopy | |
| Li et al. | A simple and nondestructive approach for the analysis of soluble solid content in citrus by using portable visible to near‐infrared spectroscopy | |
| Ouyang et al. | Intelligent sensing sensory quality of Chinese rice wine using near infrared spectroscopy and nonlinear tools | |
| Yu et al. | Nondestructive determination of SSC in Korla fragrant pear using a portable near-infrared spectroscopy system | |
| Chen et al. | Simultaneous measurement of total acid content and soluble salt‐free solids content in Chinese vinegar using near‐infrared spectroscopy | |
| CN102937575B (en) | Watermelon sugar degree rapid modeling method based on secondary spectrum recombination | |
| Assis et al. | Variable selection applied to the development of a robust method for the quantification of coffee blends using mid infrared spectroscopy | |
| Xia et al. | Effect of fruit moving speed on online prediction of soluble solids content of apple using Vis/NIR diffuse transmission | |
| Yu et al. | A portable NIR system for nondestructive assessment of SSC and firmness of Nanguo pears | |
| Zhao et al. | Exploring the use of Near-infrared spectroscopy as a tool to predict quality attributes in prickly pear (Rosa roxburghii Tratt) with chemometrics variable strategy | |
| CN113030011A (en) | Rapid nondestructive testing method and system for sugar content of fruits | |
| Hirri et al. | Prediction of polyphenol fraction in virgin olive oil using mid-infrared attenuated total reflectance attenuated total reflectance accessory–mid-infrared coupled with partial least squares regression | |
| Li et al. | A novel method to determine total sugar of Goji berry using FT-NIR spectroscopy with effective wavelength selection |
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: 20190326 |
|
| RJ01 | Rejection of invention patent application after publication |