CN106950192A - A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology - Google Patents
A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology Download PDFInfo
- Publication number
- CN106950192A CN106950192A CN201710186272.9A CN201710186272A CN106950192A CN 106950192 A CN106950192 A CN 106950192A CN 201710186272 A CN201710186272 A CN 201710186272A CN 106950192 A CN106950192 A CN 106950192A
- Authority
- CN
- China
- Prior art keywords
- sample
- vegetable protein
- fat
- protein beverage
- soluble solid
- 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 86
- 235000021568 protein beverage Nutrition 0.000 title claims abstract description 67
- 108010082495 Dietary Plant Proteins Proteins 0.000 title claims abstract description 66
- 238000005516 engineering process Methods 0.000 title claims abstract description 12
- 238000010183 spectrum analysis Methods 0.000 title claims abstract description 8
- 238000001514 detection method Methods 0.000 title claims description 32
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 53
- 102000004169 proteins and genes Human genes 0.000 claims abstract description 53
- 239000007787 solid Substances 0.000 claims abstract description 49
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims abstract description 4
- 238000007796 conventional method Methods 0.000 claims abstract description 3
- 238000002329 infrared spectrum Methods 0.000 claims description 18
- 230000002068 genetic effect Effects 0.000 claims description 16
- 238000005259 measurement Methods 0.000 claims description 15
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 230000006835 compression Effects 0.000 claims description 13
- 238000007906 compression Methods 0.000 claims description 13
- 239000000126 substance Substances 0.000 claims description 11
- 235000013305 food Nutrition 0.000 claims description 10
- 239000004575 stone Substances 0.000 claims description 8
- 238000007689 inspection Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 235000013361 beverage Nutrition 0.000 claims description 6
- 239000012141 concentrate Substances 0.000 claims description 6
- 150000001875 compounds Chemical class 0.000 claims description 5
- 239000003094 microcapsule Substances 0.000 claims description 5
- 238000005903 acid hydrolysis reaction Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 230000008676 import Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 238000012821 model calculation Methods 0.000 claims description 2
- 230000003595 spectral effect Effects 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 230000008033 biological extinction Effects 0.000 claims 1
- 230000003993 interaction Effects 0.000 claims 1
- 238000003860 storage Methods 0.000 claims 1
- 238000010200 validation analysis Methods 0.000 claims 1
- 238000012795 verification Methods 0.000 abstract description 7
- 230000008901 benefit Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 3
- 238000002835 absorbance Methods 0.000 description 2
- 239000003153 chemical reaction reagent Substances 0.000 description 2
- 238000002512 chemotherapy Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 235000016709 nutrition Nutrition 0.000 description 2
- 230000035764 nutrition Effects 0.000 description 2
- 210000004885 white matter Anatomy 0.000 description 2
- BJEPYKJPYRNKOW-REOHCLBHSA-N (S)-malic acid Chemical compound OC(=O)[C@@H](O)CC(O)=O BJEPYKJPYRNKOW-REOHCLBHSA-N 0.000 description 1
- FEWJPZIEWOKRBE-XIXRPRMCSA-N Mesotartaric acid Chemical compound OC(=O)[C@@H](O)[C@@H](O)C(O)=O FEWJPZIEWOKRBE-XIXRPRMCSA-N 0.000 description 1
- 238000004497 NIR spectroscopy Methods 0.000 description 1
- FEWJPZIEWOKRBE-UHFFFAOYSA-N Tartaric acid Natural products [H+].[H+].[O-]C(=O)C(O)C(O)C([O-])=O FEWJPZIEWOKRBE-UHFFFAOYSA-N 0.000 description 1
- 239000002253 acid Substances 0.000 description 1
- 210000000593 adipose tissue white Anatomy 0.000 description 1
- BJEPYKJPYRNKOW-UHFFFAOYSA-N alpha-hydroxysuccinic acid Natural products OC(=O)C(O)CC(O)=O BJEPYKJPYRNKOW-UHFFFAOYSA-N 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 235000011090 malic acid Nutrition 0.000 description 1
- 239000001630 malic acid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 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
- G01N21/359—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 invention discloses a kind of vegetable protein beverage protein based on near-infrared spectral analysis technology, fat and soluble solid content method for quick, this method is by under the modeling conditions of setting, using near-infrared spectral analysis technology and suitable chemometrics method is selected, the near-infrared Quantitative Prediction Model of protein in representative vegetable protein beverage, fat and soluble solid content is set up respectively；Checking optimization is carried out using verification sample the set pair analysis model, the measure to unknown vegetable protein beverage sample protein matter to be measured, fat and soluble solid content is finally realized with the near-infrared forecast model.The conventional method determined compared to existing protein, fat and soluble solid, the inventive method is simple to operate, quick, accurate, the forecast model practicality that the inventive method is obtained is stronger, can solve the active demand that enterprise controls in time to vegetable protein beverage quality and stability.
Description
Technical field
The invention belongs to Food Quality and Safety rapid detection technical field, and in particular to one kind utilizes near infrared spectrum point
Method for quick of the analysis technology to protein, fat and soluble solid content in vegetable protein beverage.
Background technology
Vegetable protein beverage, in Vehicles Collected from Market, due to its nutrition, health care, natural, mouthfeel is excellent the advantages of and enjoy and disappear
Expense person likes.Protein, fat and soluble solid are directly affected as three big main components in vegetable protein beverage
Local flavor, nutrition and the stability of beverage.Therefore, to the protein in vegetable protein beverage, fat and soluble solid content
Control in time is carried out very necessary, to ensureing that beverage quality is played an important role.
The measure of protein, fat and soluble solid is respectively Kjeldahl's method, acid in current vegetable protein beverage
Hydrolyze method and compound microcapsule, still, these method generally existing detection process take, operate the defect such as complex, and to inspection
The technical merit of survey personnel requires higher, it is impossible to realize the high-volume fast quantification inspection of Contents of Main Components in vegetable protein beverage
Survey, it is impossible to product quality and stability are controlled in time, being not suitable for enterprise is used for the quick detection of sample.
Near infrared spectrum（Near infrared spectroscopy, NIR）Technology is developed rapidly in recent years
Fast nondestructive evaluation technology, may be used with nearly all relevant with hydric group sample physico-chemical properties analyses, with it is simple to operate,
Sample without pre-treatment, be easily achieved the advantage quickly analyzed, it is extensive in fields such as food, medicine, petrochemical industry, agriculture and animal husbandries
Using market prospects are good.But at home up to now there is not yet using near-infrared spectral analysis technology to vegetable protein beverage
The relevant report that middle composition is used for quickly detecting.
The content of the invention
Present invention aims at provide protein, fat in a kind of vegetable protein beverage based on near-infrared spectral analysis technology
The method for quick of fat and soluble solid content, this method compared to the complex operation present in existing detection method, take
The defect such as power is time-consuming, it is quick, convenient to have the advantages that.
Contents of Main Components method for quick, comprises the following steps in vegetable protein beverage provided by the present invention：
1）Collect variety classes, the vegetable protein beverage sample of different brands, randomly select it is some as sample sets, using near red
External spectrum instrument gathers the atlas of near infrared spectra of sample sets；
2）Detection obtains the measured value of protein in sample sets vegetable protein beverage sample, fat and soluble solid content；
3）Modeling sample is chosen, according to 2:1 ratio, takes Kennard-Stone（K-S）Method carries out sample set division, is divided into
Calibration set and checking collect, and by step 1）In in obtained calibration set vegetable protein beverage sample near infrared spectrum data difference
With step 2）In obtain protein, fat and soluble solid content measured value it is corresponding, it is soft by Chemical Measurement
Part sets up the quantitative calibration models of vinifera sample mesotartaric acid and malic acid content, and quantitative to building with checking collection sample
Calibration model carries out external certificate, finally gives Contents of Main Components near-infrared prediction correcting model in vegetable protein beverage；
4）Testing sample is taken, by step 1）In spectral measurement condition gather testing sample near infrared spectrum data, import school
In positive forecast model, through model calculation, you can obtain the protein in unknown vegetable protein beverage sample, fat and solubility and consolidate
Shape thing content.
In above-mentioned detection method, step 1）In, collected vegetable protein beverage sample number is at least 800, is preferably
800~1200.The method of the atlas of near infrared spectra of vegetable protein beverage sample is as follows in the collection sample sets：By sample before measurement
Product, which to be placed unify at room temperature, stands 30min, and the spectra collection of sample is carried out using transflector mode, is per portion sample volume
25ml, sample measurement temperature is 40 degrees Celsius, and sample pours into specimen cup, is put into transflector lid, and drives bubble away with transflector lid,
Near infrared light covers generation diffusing reflection after passing through sample in transflector, and diffusing and passing through sample enters in detector, scanning
Number of times is 32 times, and wave-length coverage is 4000-10000cm-1, thickness of sample is fixed as 0.3mm under transflector lid, and each sample is repeated
Fill sample and gather 2 spectrum, spectrum is stored in absorbance log (1/R) form.
The vegetable protein beverage sample covers different sources, different dates of manufacture, the sample of different model specification.
The near infrared spectrometer is NIRMasterM54P Fourier Transform Near Infrared instruments（Switzerland walks the limited public affairs of fine jade
Department）.
The chemo metric software is provided by Bu Qi Co., Ltds of Switzerland.
In above-mentioned detection method, step 2）In, the method for the detection is respectively GB/T 5009.5-2010《Egg in food
The measure of white matter》Kjeldahl's method, GB/T 5009.6-2003《Fatty measure in food》Acid-hydrolysis method and GB/T
12143-2008《Beverage universaling analysis method》Compound microcapsule.
In above-mentioned detection method, step 3）In, the near-infrared quantitative calibration models to being set up carry out external certificate
The step of：By step 1）The near infrared spectrum of obtained checking collection vegetable protein beverage sample imports set up quantitative correction mould
In type, the predicted value for concentrating sample protein matter, fat and soluble solid content is verified, by itself and step 2）Described in
Checking concentrates the measured value of the protein, fat and soluble solid content of vegetable protein beverage sample to compare, and carries out model
The inspection of the degree of accuracy, if predicted value（%）With measured value（%）The absolute value and measured value of difference（%）The ratio between in the range of setting,
Then the quantitative calibration models can use；Conversely, then needing repeat step 3）, Optimization Modeling condition is until the quantitative calibration models can
With.
The quantitative calibration models are set up and obtained using following steps successively：
A, preprocessing procedures：The preprocessing procedures are selected from following at least one：Multiplicative scatter correction（MSC）、
Savitzky-Golay convolution is smooth, Savitzky-Golay first derivatives, wavelet transformation（WT）And NCL；
B, sample sets division methods：The sample sets division methods be selected from it is following any one：Randomized, Kennard-Stone
（K-S）Method, SPXY methods；
C, variable compression method：The variable compression method be selected from it is following any one：CARS methods, without information variable null method,
The interval PLS of combination, genetic algorithm；
D, Chemical Measurement modeling method：The Chemical Measurement modeling method be selected from it is following any one：PLS
（PLS）, main composition returns（PCR）, least square method supporting vector machine（LS-SVM）Or artificial neural network method（ANN）.
The quantitative calibration models are specifically set up and obtained using following steps successively：
A, preprocessing procedures：NCL；
B, sample sets division methods：Kennard-Stone（K-S）Method；
C, variable compression method：The interval PLS of combination, genetic algorithm；
D, Chemical Measurement modeling method：PLS（PLS）.
In above-mentioned detection method, step 3）In, the variable compression method realizes step：By vegetable protein beverage
The full spectrum of 4000-10000cm-1 are divided into k subinterval（K=10 ~ 40, interval 5）, it is just different respectively under different subinterval numbers
Number of combinations（1~4）It is combined interval offset minimum binary（SiPLS）Calculate, then with the minimum marks of validation-cross standard deviation RMSECV
The modeling wave band that brigadier SiPLS is filtered out is calculated using genetic algorithm, filter out respectively vegetable protein beverage protein,
The best modeled variable of fat and soluble solid.
In above-mentioned detection method, step 3）In, the operational factor of genetic algorithm is set in the variable compression method：
Initial population size 80, mutation probability Pm=0.01, crossover probability Pc=0.5, maximum factor number 10, genetic iteration number of times 120 times,
Optimal modeling variable is determined with RMSECV values.
In above-mentioned detection method, step 3）In, the best modeled variable is：
1）Each index of vegetable protein beverage filtered out through SiPLS models wave band and is respectively：Protein（4404~4800、5604~
6000、6804~7200、9204~9600cm-1）；Fat（4244~4480、4724~4960、5684~5920、8804~9040cm-1）；Soluble solid（4000~4480、4964~5200、7604~7840cm-1）；
2）Each index best modeled variable of vegetable protein beverage filtered out through genetic algorithm is respectively：Protein（4404、
4460、4500、4528、4552、4580、4596、4612、4624、4628、4672、4680、4700、4716、5608、5616、
5620、5624、5640、5644、5676、5680、5688、5700、5708、5720、5732、5752、5768、5772、5780、
5788、5792、5796、5800、5804、5808、5816、5820、5824、5828、5840、5844、5852、5864、5936、
5964、6816、6964、7112、7148 cm-1）；Fat（4272、4292、4312、4316、4320、4324、4328、4332、
4344、4348、4352、4356、4380、4384、4404、4408、4432、4436、4464、4800、4832、4928、4936、
5692、5716、5724、5728、5732、5736、5740、5744、5748、5756、5788、5800、5804、5808、5812、
5816、5828、5832、5836、5844、5852、5856、5864、5868、5872、5876、5892、5920、8868、8936
cm-1）；Soluble solid（4096、4228、4232、4244、4248、4288、4292、4300、4340、4368、4372、
4376、4380、4384、4388、4392、4396、4400、4404、4408、4412、4416、4420、4428、4432、4436、
4440、4444、4452、4456、4460、4464、4468、4472、4476、4480、4980、4988、5036、5080、7620、
7672 cm-1）.
The present invention for vegetable protein beverage protein, fat and soluble solid content measure provide it is a kind of completely newly
Quickly, accurately, simple detection method, quality and stability control and whole industry meaning weight to vegetable protein beverage
Greatly.
Compared with prior art, technical advantage of the invention is embodied in following aspect：（1）Operating procedure of the present invention is simple,
Operating process is succinct, for the manpower and materials cost needed for enterprise saves traditional detection；（2）It is main suitable for vegetable protein beverage
The rapidly and efficiently analysis of component content, with detection speed is fast, analysis efficiency is high, the advantage of stability and favorable reproducibility, is used
The present invention can be reduced in the measure that protein, fat and soluble solid content in vegetable protein beverage are completed in 40 seconds
Testing cost greatly improves detection efficiency simultaneously；（3）Without using any chemical reagent in detection process, also environment is not produced
There is provided point of protein, fat and soluble solid content in a kind of reliable vegetable protein beverage of green for any pollution
Analysis method.
Brief description of the drawings
Fig. 1 is the vegetable protein beverage protein of modeling, the spectrogram of fatty and soluble solid content.
Fig. 2, Fig. 3, Fig. 4 are vegetable protein beverage protein, fat and soluble solid content near-infrared forecast model
Design sketch.
Embodiment
The method of the present invention is illustrated below by specific embodiment, but the invention is not limited in this, it is all at this
Any modification, equivalent and improvement made within the spirit and principle of invention etc., should be included in the protection model of the present invention
Within enclosing.
Experimental method described in following embodiments, unless otherwise specified, is conventional method；The reagent and material, such as
Without specified otherwise, commercially obtain.
Embodiment 1, the vegetable protein beverage Contents of Main Components method for quick based on near-infrared spectral analysis technology
Concrete operation step is as follows：
（1）Collect sample and collection sample near infrared spectrum：Variety classes, the vegetable protein beverage sample of different brands are collected,
According to 2:1 ratio, takes Kennard-Stone（K-S）Method carries out sample set division, is divided into Calibration and verification sample
Collection, 786 parts of sample composition Calibrations of final choice, 388 parts of samples are verification sample collection.Sample sets cover different sources,
Different dates of manufacture, the sample of different model specification；
The spectra collection of sample is carried out using transflector mode, sample pours into specimen cup, be put into transflector lid, and use transflector lid
Drive bubble away, near infrared light covers generation diffusing reflection after passing through sample in transflector, diffusing and passing through sample enters detection
In device, scanning times are 32 times, and wave-length coverage is 4000-10000cm-1.Thickness of sample is fixed as 0.3mm under transflector lid, often
Individual sample repeats dress sample and gathers 2 spectrum.Collect 2 spectrum of gained and add vegetable protein drink in absorbance log (1/R) form
Expect the near infrared spectrum picture library of sample, finally the spectrum of obtained Calibration is as shown in Figure 1.
（2）Protein, fat and soluble solid content in chemical determination vegetable protein beverage sample：Respectively according to
According to GB/T 5009.5-2010《The measure of Protein in Food》Kjeldahl's method, GB/T 5009.6-2003《Fat in food
The measure of fat》Acid-hydrolysis method and GB/T 12143-2008《Beverage universaling analysis method》Compound microcapsule measure in sample
The measured value of protein, fat and soluble solid content, builds the near-infrared model of Calibration and verification sample collection
Reference value data set.
（3）The foundation of near-infrared quantitative calibration models and verification sample set pair near-infrared quantitative calibration models are tested and commented
Valency：By step（1）The atlas of near infrared spectra of gained Calibration respectively with step（2）The correcting sample of middle gained concentrates sample
Protein, fat and soluble solid content measured value correspond, set up plant respectively by chemo metric software
The quantitative calibration models of protein, fat and soluble solid content in thing protein beverage, the quantitative calibration models are successively
Set up and obtained using following steps：
A, preprocessing procedures：NCL；
B, sample sets division methods：Kennard-Stone（K-S）Method；
C, variable compression method：The interval PLS of combination, genetic algorithm；
D, Chemical Measurement modeling method：PLS.
Wherein variable compression method is achieved by the steps of：By vegetable protein beverage 4000-10000cm-1 full spectrum point
For k subinterval（K=10 ~ 40, interval 5）, under different subinterval numbers, respectively with regard to various combination number（1~4）It is combined interval
Offset minimum binary（SiPLS）Calculate, then the modeling ripple for being filtered out SiPLS with the minimum standards of validation-cross standard deviation RMSECV
Section is calculated using genetic algorithm, and the protein, fat and soluble solid of vegetable protein beverage are filtered out respectively most
Good modeling variable.
Genetic algorithm operational factor is set to：Initial population size 50, mutation probability Pm=0.01, crossover probability Pc=0.5,
Maximum factor number 10, genetic iteration number of times 120 times determines optimal modeling variable with RMSECV values.
Each index of vegetable protein beverage gone out through SiPLS preliminary screenings models wave band and is respectively：Protein（4404~4800、
5604~6000、6804~7200、9204~9600cm-1）；Fat（4244~4480、4724~4960、5684~5920、8804~
9040cm-1）；Soluble solid（4000~4480、4964~5200、7604~7840cm-1）.
Each index best modeled variable of vegetable protein beverage filtered out through genetic algorithm is respectively：Protein（4404、
4460、4500、4528、4552、4580、4596、4612、4624、4628、4672、4680、4700、4716、5608、5616、
5620、5624、5640、5644、5676、5680、5688、5700、5708、5720、5732、5752、5768、5772、5780、
5788、5792、5796、5800、5804、5808、5816、5820、5824、5828、5840、5844、5852、5864、5936、
5964、6816、6964、7112、7148 cm-1）；Fat（4272、4292、4312、4316、4320、4324、4328、4332、
4344、4348、4352、4356、4380、4384、4404、4408、4432、4436、4464、4800、4832、4928、4936、
5692、5716、5724、5728、5732、5736、5740、5744、5748、5756、5788、5800、5804、5808、5812、
5816、5828、5832、5836、5844、5852、5856、5864、5868、5872、5876、5892、5920、8868、8936
cm-1）；Soluble solid（4096、4228、4232、4244、4248、4288、4292、4300、4340、4368、4372、
4376、4380、4384、4388、4392、4396、4400、4404、4408、4412、4416、4420、4428、4432、4436、
4440、4444、4452、4456、4460、4464、4468、4472、4476、4480、4980、4988、5036、5080、7620、
7672 cm-1）.
Corresponding vegetable protein beverage protein, fat and soluble solid content near-infrared forecast model design sketch point
It is not shown as shown in Figure 2, Figure 3 and Figure 4, understand that the near-infrared of vegetable protein beverage protein, fat and soluble solid is pre- from figure
Measured value is evenly distributed on diagonal both sides with measured value, illustrates that model can be to sample protein matter, fat and soluble solid
Content is predicted.By verification sample concentrate sample according to step（1）Middle modeling sample concentrates sample identical processing mode
Handled, by step（1）The near infrared spectrum of obtained checking collection sample is imported in set up quantitative calibration models, is obtained
Verification sample concentrates the predicted value of protein, fat and soluble solid content, by itself and step（2）Described in verify massive planting
The measured value of protein, fat and soluble solid content is compared in thing protein beverage, carries out the inspection of model accuracy, if
Predicted value（%）With measured value（%）The absolute value and measured value of difference（%）The ratio between within the scope of Standard, then it is described quantitative
Calibration model can use；Conversely, then needing repeat step（3）, Optimization Modeling condition is until the quantitative calibration models can use.Quantitative school
Positive model the result is as shown in table 1.It can be seen that, model the result is good, and the degree of accuracy meets demand.
The checking collection near-infrared of table 1 protein, fat and soluble solid interpretation of result
（4）Testing sample is tested：50 vegetable protein beverage samples for having neither part nor lot in modeling are constituted into Prediction, respectively foundation
GB/T 5009.5-2010《The measure of Protein in Food》Kjeldahl's method, GB/T 5009.6-2003《It is fatty in food
Measure》Acid-hydrolysis method and GB/T 12143-2008《Beverage universaling analysis method》Compound microcapsule measure egg in sample
White matter, fat and soluble solid content, are used as measured value.Same step（1）Enter under the conditions of identical near infrared spectrum scanning
Row near infrared spectrum scanning, by gained spectroscopic data input step（3）It is protein in gained vegetable protein beverage, fatty and solvable
Property solid content quantitative calibration models in, obtain protein in vegetable protein beverage to be measured, fat and soluble solid
The predicted value of content, according to national standard, protein, fat and soluble solid model inspection result in vegetable protein beverage
It is as shown in table 2 below, as known from Table 2：Protein, fat and soluble solid index have 98%, 100%, 100% test respectively
As a result national standard is met, illustrates that the present invention is measured protein, fat and containing for soluble solid in vegetable protein beverage
There is good application effect in fixed.
The vegetable protein beverage protein of table 2, fat and soluble solid model inspection result
Claims (10)
1. a kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology, bag
Include following step：
1）Collect variety classes, the vegetable protein beverage sample of different brands, randomly select it is some as sample sets, using near red
External spectrum instrument gathers the atlas of near infrared spectra of sample sets；
2）Protein in sample sets vegetable protein beverage sample, fat and soluble solid are obtained using conventional method detection to contain
The measured value of amount；
3）Modeling sample is chosen, according to 2:1 ratio, takes Kennard-Stone（K-S）Method carries out sample set division, is divided into
Calibration set and checking collect, and by step 1）In in obtained calibration set vegetable protein beverage sample near infrared spectrum data difference
With step 2）In obtain protein, fat and soluble solid content measured value it is corresponding, pass through variable compression method
Select best modeled variable, and by Chemical Measurement modeling method set up protein in vegetable protein beverage sample, fat and
The quantitative calibration models of soluble solid content, and external certificate is carried out to built quantitative calibration models with checking collection sample,
Finally give Contents of Main Components near-infrared prediction correcting model in vegetable protein beverage；
4）Testing sample is taken, by step 1）In spectral measurement condition gather testing sample near infrared spectrum data, import school
In positive forecast model, through model calculation, you can obtain the protein in unknown vegetable protein beverage sample, fat and solubility and consolidate
Shape thing content.
2. detection method according to claim 1, it is characterised in that：Step 1）In, plant egg in the collection sample sets
The method of the atlas of near infrared spectra of white drink sample is as follows：Sample to be placed unify at room temperature before measurement and stands 30min, is used
Transflector mode carries out the spectra collection of sample, is 25ml per portion sample volume, sample measurement temperature is 40 degrees Celsius, sample
Specimen cup is poured into, transflector lid is put into, and drives with transflector lid bubble away, near infrared light covers hair after passing through sample in transflector
Raw diffusing reflection, diffusing and passing through sample enters in detector, and scanning times are 32 times, and wave-length coverage is 4000-
10000cm-1；Thickness of sample is fixed as 0.3mm under transflector lid, and each sample repeats dress sample and gathers 2 spectrum, and spectrum is with extinction
Log (1/R) form of spending storage；
The vegetable protein beverage sample covers different sources, different dates of manufacture, the sample of different model specification.
3. the detection method according to claim 1-2, it is characterised in that：Step 2）In, the method for the detection is respectively
GB/T 5009.5-2010《The measure of Protein in Food》Kjeldahl's method, GB/T 5009.6-2003《It is fatty in food
Measure》Acid-hydrolysis method and GB/T 12143-2008《Beverage universaling analysis method》Compound microcapsule.
4. the detection method according to any one of claim 1-3, it is characterised in that：Step 3）In, the quantitative correction
Model is set up and obtained using following steps successively：
A, preprocessing procedures：The preprocessing procedures are selected from following at least one：Multiplicative scatter correction,
Savitzky-Golay convolution is smooth, Savitzky-Golay first derivatives, wavelet transformation（WT）And NCL；
B, sample sets division methods：The sample sets division methods be selected from it is following any one：Randomized, Kennard-Stone
（K-S）Method, SPXY methods；
C, variable compression method：The variable compression method is selected from following any one or two kinds：CARS methods, without information variable eliminate
Method, the interval PLS of combination, genetic algorithm；
D, Chemical Measurement modeling method：The Chemical Measurement modeling method be selected from it is following any one：PLS,
Main composition recurrence, least square method supporting vector machine or artificial neural network method.
5. detection method according to claim 4, it is characterised in that：Step 3）In, the quantitative calibration models are adopted successively
Set up and obtained with following steps：
A, preprocessing procedures：NCL；
B, sample sets division methods：Kennard-Stone（K-S）Method；
C, variable compression method：The interval PLS of combination, genetic algorithm；
D, Chemical Measurement modeling method：PLS.
6. detection method according to claim 5, it is characterised in that：Step 3）In, the variable compression method passes through such as
Lower step is realized：By vegetable protein beverage 4000-10000cm-1Full spectrum is divided into k subinterval（K=10 ~ 40, interval 5）, not
With under the number of subinterval, respectively with regard to various combination number（1~4）It is combined interval offset minimum binary（SiPLS）Calculate, then with interaction
The validation criteria minimum standards of difference RMSECV are calculated the SiPLS modeling wave bands filtered out using genetic algorithm, are sieved respectively
Select the best modeled variable of the protein, fat and soluble solid of vegetable protein beverage.
7. detection method according to claim 6, it is characterised in that：Step 3）In, the variable compression method, heredity is calculated
Method operational factor is set to：Initial population size 80, mutation probability Pm=0.01, crossover probability Pc=0.5, maximum factor number 10,
Genetic iteration number of times 120 times, optimal modeling variable is determined with RMSECV values.
8. the detection method according to claim 6-7, it is characterised in that：Step 3）In, the best modeled variable is：
1）Each index of vegetable protein beverage filtered out through SiPLS models wave band and is respectively：Protein（4404~4800、5604~
6000、6804~7200、9204~9600cm-1）；Fat（4244~4480、4724~4960、5684~5920、8804~9040cm-1）；Soluble solid（4000~4480、4964~5200、7604~7840cm-1）；
2）Each index best modeled variable of vegetable protein beverage filtered out through genetic algorithm is respectively：Protein（4404、
4460、4500、4528、4552、4580、4596、4612、4624、4628、4672、4680、4700、4716、5608、5616、
5620、5624、5640、5644、5676、5680、5688、5700、5708、5720、5732、5752、5768、5772、5780、
5788、5792、5796、5800、5804、5808、5816、5820、5824、5828、5840、5844、5852、5864、5936、
5964、6816、6964、7112、7148 cm-1）；Fat（4272、4292、4312、4316、4320、4324、4328、4332、
4344、4348、4352、4356、4380、4384、4404、4408、4432、4436、4464、4800、4832、4928、4936、
5692、5716、5724、5728、5732、5736、5740、5744、5748、5756、5788、5800、5804、5808、5812、
5816、5828、5832、5836、5844、5852、5856、5864、5868、5872、5876、5892、5920、8868、8936
cm-1）；Soluble solid（4096、4228、4232、4244、4248、4288、4292、4300、4340、4368、4372、
4376、4380、4384、4388、4392、4396、4400、4404、4408、4412、4416、4420、4428、4432、4436、
4440、4444、4452、4456、4460、4464、4468、4472、4476、4480、4980、4988、5036、5080、7620、
7672 cm-1）.
9. the detection method according to any one of claim 1-8, it is characterised in that：Step 3）In, it is described to being set up
Near-infrared quantitative calibration models carry out external certificate the step of：By step 1）Obtained checking collection vegetable protein beverage sample
Near infrared spectrum is imported in set up quantitative calibration models, is verified concentration sample protein matter, fat and soluble solid
The predicted value of thing content, by itself and step 2）Described in checking concentrate the protein, fatty and solvable of vegetable protein beverage sample
Property solid content measured value compare, carry out model accuracy inspection, if predicted value（%）With measured value（%）Difference it is exhausted
To value and measured value（%）The ratio between in the range of setting, then the quantitative calibration models can use；Conversely, then needing repeat step 3）,
Optimization Modeling condition is until the quantitative calibration models can use.
10. the detection method according to any one of claim 1-9, it is characterised in that：The near infrared spectrometer is
NIRMasterM54P Fourier Transform Near Infrared instruments（Bu Qi Co., Ltds of Switzerland）.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710186272.9A CN106950192A (en) | 2017-03-27 | 2017-03-27 | A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710186272.9A CN106950192A (en) | 2017-03-27 | 2017-03-27 | A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106950192A true CN106950192A (en) | 2017-07-14 |
Family
ID=59472468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710186272.9A Pending CN106950192A (en) | 2017-03-27 | 2017-03-27 | A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106950192A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108226090A (en) * | 2016-12-15 | 2018-06-29 | 中国农业机械化科学研究院 | A kind of method of component content detection model structure |
| CN111257279A (en) * | 2019-12-20 | 2020-06-09 | 杭州娃哈哈精密机械有限公司 | Near-infrared detection system for on-line determination of content of functional components in milk beverage |
| CN112525855A (en) * | 2020-11-20 | 2021-03-19 | 广东省农业科学院蔬菜研究所 | Detection method and device for quality parameters of pumpkin fruits and computer equipment |
| CN112798555A (en) * | 2020-12-25 | 2021-05-14 | 江苏大学 | Modeling method for improving adaptability of coarse protein correction model of wheat flour |
| CN113267458A (en) * | 2021-05-21 | 2021-08-17 | 河南科技大学 | Method for establishing quantitative prediction model of soluble protein content of sweet potatoes |
| CN115015153A (en) * | 2022-06-29 | 2022-09-06 | 南京农业大学 | Formula apple puree quality evaluation method based on spectral reconstruction technology |
| CN117036963A (en) * | 2023-10-08 | 2023-11-10 | 中国农业大学 | Detection and estimation method for typical grassland surface plant quantity |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279168A (en) * | 2011-07-20 | 2011-12-14 | 浙江大学 | Near-infrared spectroscopic technology-based method for fast and undamaged analysis of nutritional quality of whole cottonseed |
| CN102393376A (en) * | 2011-10-14 | 2012-03-28 | 上海海洋大学 | Support vector regression-based near infrared spectroscopy for detecting content of multiple components of fish ball |
| CN105588819A (en) * | 2016-03-11 | 2016-05-18 | 江西出入境检验检疫局检验检疫综合技术中心 | Method for conducting near-infrared rapid detection on component content in infant formula rice flour |
-
2017
- 2017-03-27 CN CN201710186272.9A patent/CN106950192A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102279168A (en) * | 2011-07-20 | 2011-12-14 | 浙江大学 | Near-infrared spectroscopic technology-based method for fast and undamaged analysis of nutritional quality of whole cottonseed |
| CN102393376A (en) * | 2011-10-14 | 2012-03-28 | 上海海洋大学 | Support vector regression-based near infrared spectroscopy for detecting content of multiple components of fish ball |
| CN105588819A (en) * | 2016-03-11 | 2016-05-18 | 江西出入境检验检疫局检验检疫综合技术中心 | Method for conducting near-infrared rapid detection on component content in infant formula rice flour |
Non-Patent Citations (1)
| Title |
|---|
| 邱燕燕 等: "近红外法测定豆浆蛋白质、脂肪和可溶性固形物含量", 《中国粮油学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108226090A (en) * | 2016-12-15 | 2018-06-29 | 中国农业机械化科学研究院 | A kind of method of component content detection model structure |
| CN108226090B (en) * | 2016-12-15 | 2020-02-07 | 中国农业机械化科学研究院 | Method for constructing component content detection model |
| CN111257279A (en) * | 2019-12-20 | 2020-06-09 | 杭州娃哈哈精密机械有限公司 | Near-infrared detection system for on-line determination of content of functional components in milk beverage |
| CN112525855A (en) * | 2020-11-20 | 2021-03-19 | 广东省农业科学院蔬菜研究所 | Detection method and device for quality parameters of pumpkin fruits and computer equipment |
| CN112798555A (en) * | 2020-12-25 | 2021-05-14 | 江苏大学 | Modeling method for improving adaptability of coarse protein correction model of wheat flour |
| CN113267458A (en) * | 2021-05-21 | 2021-08-17 | 河南科技大学 | Method for establishing quantitative prediction model of soluble protein content of sweet potatoes |
| CN115015153A (en) * | 2022-06-29 | 2022-09-06 | 南京农业大学 | Formula apple puree quality evaluation method based on spectral reconstruction technology |
| CN117036963A (en) * | 2023-10-08 | 2023-11-10 | 中国农业大学 | Detection and estimation method for typical grassland surface plant quantity |
| CN117036963B (en) * | 2023-10-08 | 2024-01-26 | 中国农业大学 | Detection and estimation method for typical grassland surface plant quantity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106950192A (en) | A kind of method of Contents of Main Components quick detection in vegetable protein beverage based on near-infrared spectral analysis technology | |
| Zhang et al. | Rapid determination of leaf water content using VIS/NIR spectroscopy analysis with wavelength selection | |
| CN102879353B (en) | The method of content of protein components near infrared detection peanut | |
| Ren et al. | Multi-variable selection strategy based on near-infrared spectra for the rapid description of dianhong black tea quality | |
| CN107064047A (en) | A kind of Fuji apple quality damage-free detection method based near infrared spectrum | |
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN102507480B (en) | Method for nondestructively and quickly measuring moisture content of tea leaf based on 12 characteristic wavelengths | |
| CN111965140B (en) | Wavelength point recombination method based on characteristic peak | |
| CN106841083A (en) | Sesame oil quality detecting method based on near-infrared spectrum technique | |
| CN104020129A (en) | Method for discriminating fermentation quality of congou black tea based on near-infrared-spectroscopy-combined amino acid analysis technology | |
| CN109540836A (en) | Near infrared spectrum pol detection method and system based on BP artificial neural network | |
| CN109669023A (en) | A kind of soil attribute prediction technique based on Multi-sensor Fusion | |
| CN109211829A (en) | A method of moisture content in the near infrared spectroscopy measurement rice based on SiPLS | |
| CN111272696A (en) | Method for rapidly detecting essence doped in Pu' er tea | |
| CN105784628A (en) | Method for detecting chemical composition of soil organic matter with mid-infrared spectra | |
| CN102393376A (en) | Support vector regression-based near infrared spectroscopy for detecting content of multiple components of fish ball | |
| Wang et al. | Onsite nutritional diagnosis of tea plants using micro near-infrared spectrometer coupled with chemometrics | |
| CN105758819A (en) | Method for detecting organic components of soil by utilizing near infrared spectrum | |
| CN102937575B (en) | Watermelon sugar degree rapid modeling method based on secondary spectrum recombination | |
| CN104778349B (en) | One kind is used for rice table soil nitrogen application Classified Protection | |
| CN103411895B (en) | Pseudo-near infrared spectrum identification method mixed by pearl powder | |
| CN110487746A (en) | A method of baby cabbage quality is detected based near infrared spectrum | |
| CN105486663A (en) | Method for detecting stable carbon isotopic ratio of soil through near infrared spectrum | |
| CN104596979A (en) | Method for measuring cellulose of reconstituted tobacco by virtue of near infrared reflectance spectroscopy technique | |
| CN109540832A (en) | A kind of detection method based on total nitrogen in the near-to-mid infrared spectrum large-scale milch cow farms liquid manure of 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170714 |