CN110057777A - The quantitative detecting method of vomitoxin in a kind of flour - Google Patents
The quantitative detecting method of vomitoxin in a kind of flour Download PDFInfo
- Publication number
- CN110057777A CN110057777A CN201910442284.2A CN201910442284A CN110057777A CN 110057777 A CN110057777 A CN 110057777A CN 201910442284 A CN201910442284 A CN 201910442284A CN 110057777 A CN110057777 A CN 110057777A
- Authority
- CN
- China
- Prior art keywords
- flour
- sample
- don
- vomitoxin
- concentration
- 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
- LINOMUASTDIRTM-QGRHZQQGSA-N deoxynivalenol Chemical compound C([C@@]12[C@@]3(C[C@@H](O)[C@H]1O[C@@H]1C=C(C([C@@H](O)[C@@]13CO)=O)C)C)O2 LINOMUASTDIRTM-QGRHZQQGSA-N 0.000 title claims abstract description 39
- LINOMUASTDIRTM-UHFFFAOYSA-N vomitoxin hydrate Natural products OCC12C(O)C(=O)C(C)=CC1OC1C(O)CC2(C)C11CO1 LINOMUASTDIRTM-UHFFFAOYSA-N 0.000 title claims abstract description 39
- 235000013312 flour Nutrition 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000001228 spectrum Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 11
- 230000003595 spectral effect Effects 0.000 claims description 10
- 239000000843 powder Substances 0.000 claims description 7
- IOLCXVTUBQKXJR-UHFFFAOYSA-M potassium bromide Chemical compound [K+].[Br-] IOLCXVTUBQKXJR-UHFFFAOYSA-M 0.000 claims description 6
- 238000001035 drying Methods 0.000 claims description 5
- 239000007787 solid Substances 0.000 claims description 5
- 238000004611 spectroscopical analysis Methods 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 239000002202 Polyethylene glycol Substances 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 239000008367 deionised water Substances 0.000 claims description 3
- 229910021641 deionized water Inorganic materials 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 229920001223 polyethylene glycol Polymers 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 238000002156 mixing Methods 0.000 claims description 2
- 239000006228 supernatant Substances 0.000 claims 1
- 238000000926 separation method Methods 0.000 abstract description 12
- 238000002329 infrared spectrum Methods 0.000 abstract description 10
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000004445 quantitative analysis Methods 0.000 abstract description 5
- 238000005259 measurement Methods 0.000 abstract description 3
- 238000011160 research Methods 0.000 abstract description 3
- 238000003062 neural network model Methods 0.000 abstract description 2
- 238000012216 screening Methods 0.000 abstract description 2
- 229930002954 deoxynivalenol Natural products 0.000 description 27
- 210000002569 neuron Anatomy 0.000 description 7
- 241000209140 Triticum Species 0.000 description 6
- 235000021307 Triticum Nutrition 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 3
- 235000013305 food Nutrition 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000004615 ingredient Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 206010047700 Vomiting Diseases 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 102000004169 proteins and genes Human genes 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000008673 vomiting Effects 0.000 description 2
- 241000894006 Bacteria Species 0.000 description 1
- 206010012735 Diarrhoea Diseases 0.000 description 1
- 239000004593 Epoxy Substances 0.000 description 1
- 206010037660 Pyrexia Diseases 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 229920002472 Starch Polymers 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 231100000570 acute poisoning Toxicity 0.000 description 1
- 239000012773 agricultural material Substances 0.000 description 1
- 150000001336 alkenes Chemical class 0.000 description 1
- 208000022531 anorexia Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 238000011097 chromatography purification Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 206010061428 decreased appetite Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 235000014113 dietary fatty acids Nutrition 0.000 description 1
- 238000005485 electric heating Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 229930195729 fatty acid Natural products 0.000 description 1
- 239000000194 fatty acid Substances 0.000 description 1
- 150000004665 fatty acids Chemical class 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 125000000468 ketone group Chemical group 0.000 description 1
- 150000002576 ketones Chemical class 0.000 description 1
- 238000004811 liquid chromatography Methods 0.000 description 1
- 239000007791 liquid phase Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004570 mortar (masonry) Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 235000019698 starch Nutrition 0.000 description 1
- 239000008107 starch Substances 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000004885 tandem mass spectrometry Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 231100000167 toxic agent Toxicity 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000017260 vegetative to reproductive phase transition of meristem Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N1/00—Sampling; Preparing specimens for investigation
- G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
- G01N1/40—Concentrating samples
-
- 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
-
- 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
- G01N2021/3595—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using FTIR
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
- G01N2201/1296—Using chemometrical methods using neural networks
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 quantitative detecting methods of vomitoxin in flour.A kind of beneficiation technologies are first passed through, high concentration DON flour sample is made, then the DON in high concentration DON flour sample is measured using near-infrared spectrum technique, is finally converted into the DON content of original flour sample again.This research promotes DON concentration using separation and concentration technology, then establishes the DON Quantitative Analysis Model based on BP neural network, related coefficient reaches 0.92 or more, and modeling works well.The R of neural network model has larger promotion, and RMSEP is also down to 0.273ppm from 0.426, further increases the Stability and adaptability of model.It is relatively small using the flour forecast set error mean square root after separation and concentration, illustrate that it differs smaller with true value, when large-scale flour mill's detection DON content, the method that directly scanning flour spectrum can be used carries out factory measurement, provides a kind of preliminary screening detection method for it.
Description
Technical field
The present invention relates to a kind of quantitative detecting methods of vomitoxin in flour.
Background technique
Wheat is one of global staple food crop, flour as wheat main converted products people diet
Occupy very big specific gravity in structure, however wheat is influenced that vomitoxin can be generated by head blight during own growth, also known as
Deoxynivalenol (DON) this toxic compounds.Vomitoxin is due to can cause the vomiting reaction of the mankind and animal
Name.The acute poisonings symptoms such as vomiting, fever, anorexia, diarrhea can occur after people has excessively eaten by the food of DON pollution.
Therefore, DON levels should be lower than 1ppm in China's regulation flour product.
Currently, mainly having for the measuring method of DON: liquid chromatogram-tandem mass spectrometry;Immunoaffinity chromatography purification is high
Effect liquid phase chromatogram method;Tlc Determination method etc..The pretreatment of these detection method samples is at high cost, and detection cycle is long, operation
Difficulty is big, and professional technician is needed to operate, and equipment is expensive, has been unable to meet the market demand.In recent years, near infrared spectrum skill
The advantages that art is quick, lossless, pollution-free because of its, be widely used to crops, agricultural and sideline product quantitative and qualitative analysis in.But
It is, since the near infrared light spectrum signal of most its special component of agricultural material is all very faint, especially when certain ingredient contains
When measuring very low, the signal that near infrared spectrum can reflect is just weaker, and various factors interferes in addition, therefore, limits near infrared spectrum
Detectability of the technology to low content ingredient, it is however generally that, when the content of certain ingredient is lower than 5ppm, using near infrared spectrum
Technology carries out quantitative analysis to it with regard to highly difficult, and detection error is larger.
Summary of the invention
The technical problem to be solved by the present invention is to overcome the defects of the prior art, provide one kind.
In order to solve to determine using near-infrared spectrum technique it when the content of vomitoxin in flour is lower than 5ppm
Amount analysis is just highly difficult, and the larger problem of detection error, the present invention first passes through a kind of beneficiation technologies, and high concentration DON flour is made
Then sample measures the DON in high concentration DON flour sample using near-infrared spectrum technique, is finally converted into again
The DON content of original flour sample.
1, in a kind of flour vomitoxin quantitative detecting method, which comprises the following steps:
S1: sample enrichment processing
Polyethylene glycol and deionized water is added after mixing into sample, is centrifuged with centrifuge, then takes appropriate upper layer
The stillness of night is placed in glassware, is put in constant temperature drying box, and it is spare that solid taking-up will be precipitated;
S2 spectral scan
The solid that will be precipitated in S1 is ground into fine powder and is uniformly mixed with potassium bromide powder, is packed into mold, is pressed into
The piece pressed is placed on scanning optical spectrum on instrument by piece;
S3 acquires spectrum
The S2 spectroscopic data obtained is converted into csv file, the drawing of Origin8.0 software is imported, obtains the spectrum of sample
Figure;
S4 Pretreated spectra
Rejecting abnormalities sample is simultaneously smooth by spectral subtraction the atmospheric background, correction baseline and Savitzky-Golay multiple spot;
S5 modeling parameters are chosen
Take mapminmax () by data normalization to (- 1,1), training objective error is set as 10^-8;Learning rate is set
Being equipped with is 0.1, and the number of plies of hidden layer is set as 3, and hidden layer node takes 8.Then the data for randomly selecting 2/3rds samples are made
It is used for model training for training data, remaining one third sample to carry out model verifying as test data;
The model based on BP neural network is established when modeling in S5
The BP of BP neural network refers to Back Propagation (backpropagation), the as backpropagation of error, letter
It number propagating forward, a branch of the BP neural network as ANN is similar with most of neural network, main difference is that
The difference of neuron, wherein typical neuron models are as schemed:
By this model, each neuron receives the input signal from other neurons, and each signal passes through
One connection with weight is transmitted, and neuron adds up these signals to obtain a total input value, then by total input value
It is compared with the threshold value of neuron, final output finally is calculated through activation primitive, output again can be as neuron later
Input hand in layer.
Assuming that the node of input layer is n, the number of nodes of hidden layer is l, and the number of nodes of output layer is m, and input layer is to implying
The weight of layer is wij, the weight of hidden layer to output layer is wjk, input layer to hidden layer is biased to aj, hidden layer to output layer
Be biased to bk, learning rate τ, excitation function is g (x), takes sigmoid function, form is
The output of hidden layer are as follows:
The output of output layer are as follows:
Take error formula are as follows:(wherein i=1 ... n, j=1 ... l, k=1 ... m.)
To make error function reach minimum value, i.e. minE, during error back propagation, we used under gradient
Drop method, so that the weight for calculating input layer to hidden layer updates:
It obtains:
Weight more new formula of the hidden layer to output layer are as follows: (wjk=wjk+τHjek)
When the difference between adjacent error twice is less than specified numerical value, algorithm reaches the setting such as specified the number of iterations
When condition, judge that algorithm has been restrained.
The beneficial effects obtained by the present invention are as follows being: the present invention first passes through a kind of beneficiation technologies, and high concentration DON flour sample is made
This, then measures the DON in high concentration DON flour sample using near-infrared spectrum technique, is finally converted into original again
The DON content of beginning flour sample.This research promotes DON concentration using separation and concentration technology, then establishes and is based on BP neural network
DON Quantitative Analysis Model, related coefficient reaches 0.92 or more, and modeling works well.Opposite Shen Fei et al., neural network model
R have larger promotion, RMSEP is also down to 0.273ppm from 0.426, further increases the Stability and adaptability of model.
It is relatively small using the flour forecast set error mean square root after separation and concentration, illustrate that it differs smaller with true value, large-scale flour
When DON content detects in factory, the method that directly scanning flour spectrum can be used carries out factory measurement, provides a kind of preliminary screening for it
Detection method.
Detailed description of the invention
Attached drawing is used to provide further understanding of the present invention, and constitutes part of specification, with reality of the invention
It applies example to be used to explain the present invention together, not be construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the spectrogram of all samples in embodiment;
Fig. 2 is the mahalanobis distance distribution map of sample;
Fig. 3 is pretreated sample light spectrogram;
Fig. 4 is the matched curve of model after separation and concentration
Specific embodiment
Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings, it should be understood that preferred reality described herein
Apply example only for the purpose of illustrating and explaining the present invention and is not intended to limit the present invention.
Embodiment
1.1 experimental materials and device
In jan flowering wheat, in different fields, the reagent for spraying various concentration deoxynivalenol bacterium makes wheat difference journey
Then degree morbidity is gathered in after wheat is mature, by the process flow of national standard powder and standard, is made containing various concentration DON's
Standard flour amounts to 103 experiment samples.
Laboratory apparatus: Nicolet iS10 Fourier transformation infrared spectrometer: U.S. Thermo Nicolet
Corporation company;High performance liquid chromatograph (is furnished with UV detector): Japanese Shimadzu Corporation;101-1A type electric heating constant temperature
Drying box: Shanghai Jin Yu scientific instrument Co., Ltd;Infrared tablet press machine: Tianjin daylight optical instrument manufacture;Automatic classification vibration
Sieve: Zibo Long Zi environmental protection equipment Co., Ltd;JM-A20002 type electronic balance: the upper positive Medical Instruments Co., Ltd of sea light.
1.2 test method
103 samples are acquired in experiment altogether, each sample is divided into two parts, and national food safety standard is pressed by first part
Immunoaffinity chromatography purifies its DON content of high effective liquid chromatography for measuring in GB5009.111-2016.Second part is through separating
After enrichment processing, spectral scan experiment is carried out.
1.2.1 sample enrichment is handled
From each sample of second part, 5g flour is taken, 1g polyethylene glycol is added and 200ml deionized water is uniformly mixed
Afterwards, 8min is centrifuged under conditions of 7000r/min with centrifuge, then taking 10ml solution, the stillness of night is placed in glassware at the middle and upper levels
In, it is put in constant temperature drying box, for 24 hours with 35 DEG C of temperature drying, it is spare that solid taking-up will be precipitated.
1.2.2 spectral scan
The flour sample by enrichment processing for taking 2mg, is ground into fine powder and potassium bromide (spectrum in the agate mortar
It is pure) powder (200mg, 200 mesh of granularity) be uniformly mixed, be packed into mold in, press 20s that piece is made on tablet press machine with 20MPa pressure.
The piece pressed is placed on scanning optical spectrum on instrument, the spectral wavelength ranges of the instrument are 400~4000cm-1, resolution ratio 4cm-1, mean scan number be 32, instrument preheats at least 60min before formal scanning, and experimental temperature is controlled at 20 DEG C or so, as far as possible
It reduces environment temperature to impact experimental result, it is ensured that the consistency of the spectroscopic data collected.
2.1 acquisition spectrum
The preparation of sample and the scanning of spectrum are the bases followed by quantitative analysis work, and obtained data are to determine
The source of subsequent data analysis and processing method selection, will have a direct impact on model learning effect and predictive ability.
The spectroscopic data of two parts sample collection before and after separation and concentration is converted into csv file, imports Origin8.0 software
Drawing, the spectrogram for obtaining all samples are as shown in Figure 2.DON chemical structure is 3 α, 7 α, 15- trihydroxy -12,13- epoxy list
Hold mould -8 ketone of -9 alkene of spore, molecular formula, the presence of unsaturated ketone group makes it have absorption peak under short wavelength UV, but with it is many other
The ultraviolet absorption peak of substance here overlaps, and belongs to non-characteristic.
The curve of spectrum in Fig. 1 is observed it is recognized that while DON content is different from each sample, but its curve of spectrum exists
Very much like variation tendency is all shown in entire band, especially at wave crest and trough, shows each sample in chemistry
There is similitude in terms of constituent and characteristic wavelength.
In surveyed wave band, 1677~1537cm-1 is the characteristic absorption peak of protein;1525~1479cm-1 is fat
Characteristic absorption peak;1785cm-1 is fatty acid characteristic absorption peak;And 1824cm-1 absorption value very little, it may be miscellaneous for noise etc.
Peak.And the generation of DON is mainly related to starch, protein and fat, so can completely be read by the variation of spectrum wave crest
To the information of substance contained therein.
After separation and concentration, the DON concentration of sample can promote 5 times or so.By sample or it is divided into training set and forecast set,
Guarantee that the sample after dividing can be uniformly distributed and have good representativeness to bulk sample sheet, to enhance the accuracy and representative of model
Property.Table 1 is DON routine data statistical analysis after sample separation and concentration.
1 sample DON routine data of table statistical analysis
Tab.1 Partial sample DON content
The rejecting of 2.2 exceptional samples
In spectral-analysis process, the reason of often generating exceptional sample, this is caused to happen, is more, including measurement background
Variation, noise of instrument interference etc. all may make spectroscopic data deviation occur, and it is abnormal abnormal with spectral value often to show as reference value.It is right
It needs to reject exceptional sample before concentration abnormality or spectral singularity sample, modeling, guarantees the accuracy of model, improved mould
The precision of prediction of type.This time exceptional sample is rejected using mahalanobis distance method in research.
Mahalanobis distance method is a kind of method of discrimination based on the distance between one group of point and multivariate space mass center.This method
Sample data is standardized based on sample spectrum matrix and centralization, calculates the space mahalanobis distance of all samples, led to
It crosses given threshold and the sample that sample mahalanobis distance is more than threshold value is determined as exceptional sample.
Mahalanobis distance method assume wherein for from n sample, wherein then sample to center of gravity Mahalanobis square distance define
For.It can wherein be estimated by sample covariance matrix.
Easily card, when n is larger, approximation is obeyed, and critical value can be found by distribution table.At that time, i-th of sample was judged to
It is abnormal.
As shown in Figure 2,11 in mahalanobis distance Near Threshold, 14,89, No. 100 samples easily cause in modeling sample
Model prediction accuracy decline.Therefore sample is identified as exceptional sample and is rejected.Sample after separation and concentration is carried out same
After sample operation, 6 exceptional samples are rejected.
2.3 Pretreated spectras and analysis
It include various make an uproar near infrared spectrum by the interference of outside environmental elements, the objective factors such as unstable of instrument
Sound then needs first to pre-process near infrared spectrum data before modeling, inhibits the scattering of spectrum, drift etc., reduces the mistake of model
Difference, thus the precision of prediction of lift scheme.By reading corresponding bibliography, atmosphere is deducted to original spectrum in this experiment
Background corrects baseline, and Savitzky-Golay multiple spot is smooth, and before and after separation and concentration, two parts sample is by pretreated close red
External spectrum figure such as Fig. 3:
2.4 modeling parameters are chosen
For the convergence for accelerating training network, take mapminmax () by data normalization to (- 1,1), training objective is missed
Difference is set as 10^-8;The setting of learning rate is not answered excessive, although excessive can accelerate to restrain effect when starting, is closed on best
When point, upheaval can be generated, and causes not restraining, is set to 0.1.For the number of plies setting of hidden layer, in order to accelerate net
Its quantity is set as 3 layers in experiment by the training speed of network.Node in hidden layer will affect the performance of network model, take 8.It surveys
Try data: the data of random 83 samples of selection are tested as training data, remaining 20 as test data.
Inspection result such as table 2:
2 DON content prediction result of table
True value and predicted value matched curve after sample separation and concentration are as shown in figure 4, slope is 0.93, intercept 63.86,
The R of matched curve2It is 0.9278, illustrates that fitting effect is preferable.Under the conditions of same concentrations, fitting effect when relatively directly scanning is more
It is accurate.Modeling using separation and concentration to sample predictions collection, calculated result are slightly changed, forecast set root-mean-square error by
516.84 are down to 273.56, and burst error illustrates that model deviates the degree significant decrease of true value between 62.3 to 583.2.
Finally, it should be noted that the foregoing is only a preferred embodiment of the present invention, it is not intended to restrict the invention,
Although the present invention is described in detail referring to the foregoing embodiments, for those skilled in the art, still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features.
All within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in of the invention
Within protection scope.
Claims (4)
1. the quantitative detecting method of vomitoxin in a kind of flour, which comprises the following steps:
S1: sample enrichment processing
Polyethylene glycol and deionized water is added after mixing into sample, is centrifuged with centrifuge, then takes appropriate supernatant
It is placed in glassware, is put in constant temperature drying box, it is spare that solid taking-up will be precipitated;
S2 spectral scan
The solid that will be precipitated in S1 is ground into fine powder and is uniformly mixed with potassium bromide powder, is packed into mold, tabletted, will
The piece pressed is placed on scanning optical spectrum on instrument;
S3 acquires spectrum
The S2 spectroscopic data obtained is converted into csv file, the drawing of Origin8.0 software is imported, obtains the spectrogram of sample;
S4 Pretreated spectra
Rejecting abnormalities sample is simultaneously smooth by spectral subtraction the atmospheric background, correction baseline and Savitzky-Golay multiple spot;
S5 modeling parameters are chosen
Take mapminmax () by data normalization to (- 1,1), training objective error is set as 10^-8;Learning rate is provided with
It is 0.1, the number of plies of hidden layer is set as 3, hidden layer node 8;Then the data of 2/3rds samples are randomly selected as training
Data are used for model training, remaining one third sample carries out model verifying as test data.
2. the quantitative detecting method of vomitoxin in flour as described in claim 1, which is characterized in that the scanning used in S2
When spectral wavelength ranges be 400~4000cm-1, resolution ratio 4cm-1, mean scan number be 32, instrument is before formal scanning
60min or more is preheated, experimental temperature is controlled at 20 DEG C ± 2 DEG C.
3. the quantitative detecting method of vomitoxin in flour as described in claim 1, which is characterized in that established in S5 and be based on BP
The analysis model of neural network.
4. the quantitative detecting method of vomitoxin in flour as described in claim 1, which is characterized in that in S4 using geneva away from
Exceptional sample is rejected from method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910442284.2A CN110057777A (en) | 2019-05-24 | 2019-05-24 | The quantitative detecting method of vomitoxin in a kind of flour |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910442284.2A CN110057777A (en) | 2019-05-24 | 2019-05-24 | The quantitative detecting method of vomitoxin in a kind of flour |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110057777A true CN110057777A (en) | 2019-07-26 |
Family
ID=67324470
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910442284.2A Pending CN110057777A (en) | 2019-05-24 | 2019-05-24 | The quantitative detecting method of vomitoxin in a kind of flour |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110057777A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103234925A (en) * | 2013-04-12 | 2013-08-07 | 重庆大学 | Method for rapidly determining trace quantity of thallium in drinking water |
CN104593332A (en) * | 2014-10-22 | 2015-05-06 | 中国农业科学院植物保护研究所 | Kit and method for detecting Tilletia controversa Kuhn |
CN105137080A (en) * | 2014-06-06 | 2015-12-09 | 北京华安麦科生物技术有限公司 | Multi-fungal toxin combined rapid detection kit and preparation method and use method thereof |
CN107589205A (en) * | 2017-09-30 | 2018-01-16 | 中南粮油食品科学研究院有限公司 | A kind of method that deoxynivalenol in wheat is detected based on high performance liquid chromatography |
CN107991265A (en) * | 2017-11-17 | 2018-05-04 | 西南大学 | A kind of wheat flour Rubus biflorus Buch quick determination method based on information fusion |
CN108872143A (en) * | 2018-05-22 | 2018-11-23 | 南京农业大学 | A kind of wheat infection head blight level detection method based near infrared spectrum |
CN109557014A (en) * | 2019-01-24 | 2019-04-02 | 广东产品质量监督检验研究院(国家质量技术监督局广州电气安全检验所、广东省试验认证研究院、华安实验室) | A kind of method of lactic acid bacteria number in quick detection acidified milk |
-
2019
- 2019-05-24 CN CN201910442284.2A patent/CN110057777A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103234925A (en) * | 2013-04-12 | 2013-08-07 | 重庆大学 | Method for rapidly determining trace quantity of thallium in drinking water |
CN105137080A (en) * | 2014-06-06 | 2015-12-09 | 北京华安麦科生物技术有限公司 | Multi-fungal toxin combined rapid detection kit and preparation method and use method thereof |
CN104593332A (en) * | 2014-10-22 | 2015-05-06 | 中国农业科学院植物保护研究所 | Kit and method for detecting Tilletia controversa Kuhn |
CN107589205A (en) * | 2017-09-30 | 2018-01-16 | 中南粮油食品科学研究院有限公司 | A kind of method that deoxynivalenol in wheat is detected based on high performance liquid chromatography |
CN107991265A (en) * | 2017-11-17 | 2018-05-04 | 西南大学 | A kind of wheat flour Rubus biflorus Buch quick determination method based on information fusion |
CN108872143A (en) * | 2018-05-22 | 2018-11-23 | 南京农业大学 | A kind of wheat infection head blight level detection method based near infrared spectrum |
CN109557014A (en) * | 2019-01-24 | 2019-04-02 | 广东产品质量监督检验研究院(国家质量技术监督局广州电气安全检验所、广东省试验认证研究院、华安实验室) | A kind of method of lactic acid bacteria number in quick detection acidified milk |
Non-Patent Citations (5)
Title |
---|
C. BOLDUAN, J.M. MONTES 等: "Determination of Mycotoxin Concentration by ELISA and Near-Infrared Spectroscopy in Fusarium-Inoculated Maize", 《CEREAL RESEARCH COMMUNICATIONS 》 * |
GREGOR KOS, MARKUS SIEGER等: "A novel chemometric classification for FTIR spectra of mycotoxin regulatory limits", 《FOOD ADDITIVES & CONTAMINANTS: PART A》 * |
KUN LIANG 等: "Comparison of Vis-NIR and SWIR hyperspectral imaging for the non-destructive detection of DON levels in Fusarium head blight wheat kernels and wheat flour", 《INFRARED PHYSICS & TECHNOLOGY》 * |
栢凤女 等: "基于支持向量机的食源性致病菌近红外光谱鉴别", 《食品科学》 * |
王淞等: "小麦粉中呕吐毒素污染调查及膳食暴露分析", 《食品研究与开发》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Fan et al. | Individual wheat kernels vigor assessment based on NIR spectroscopy coupled with machine learning methodologies | |
Huang | Chemometric methods in analytical spectroscopy technology | |
CN102179375A (en) | Nondestructive detecting and screening method based on near-infrared for crop single-grain components | |
CN106596436B (en) | Multi-parameter water quality real-time online monitoring device based on spectrum method | |
CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
CN101216419A (en) | Method for quickly detecting yellow wine quality index | |
CN107703097A (en) | Utilize the method and its application of decay total reflection probe and the model of near infrared spectrometer structure fast prediction oil property | |
CN106198447A (en) | Chemical Mixed Fertilizer main component harmless quantitative detection method based on near-infrared spectrum technique | |
CN108760647A (en) | A kind of wheat content of molds line detecting method based on Vis/NIR technology | |
CN1126515A (en) | Process of measuring chemical and physical parameters for characterisation and classification of aqueous suspensions | |
CN106841066A (en) | A kind of local alignment method in the detection for spectroscopic methodology COD | |
CN110007026A (en) | Trace nitrogen oxides detection method and device in a kind of atmosphere | |
CN101614663A (en) | As seen and near infrared spectrum bee pollen variety discriminating | |
CN103487398B (en) | A kind of analytical method of lysine fermentation liquor | |
CN105527236A (en) | Method for determination of main nutritional components of agricultural product by use of spectroscopy method | |
Sheng et al. | Analysis of protein and fat in milk using multiwavelength gradient-boosted regression tree | |
CN114636687A (en) | Small sample coal quality characteristic analysis system and method based on deep migration learning | |
CN112630180B (en) | Ultraviolet/visible light absorption spectrum model for detecting concentration of organophosphorus pesticide in water body | |
CN103134770A (en) | Method for eliminating influence on infrared spectrum detection of soil total nitrogen content from moisture | |
CN105259136B (en) | Near infrared correction without measuring point temperature correction | |
CN107121408A (en) | The quick nondestructive discrimination method of edible vegetable oil kind | |
CN110057777A (en) | The quantitative detecting method of vomitoxin in a kind of flour | |
CN104198397A (en) | Method for detecting nutrient content of greenhouse crops under nitrogen, phosphorus and potassium interaction | |
EP0187675B1 (en) | Method of detection and quantitative determination of sulfur and sulfur monitor using the method | |
Sun et al. | Visible-NIR spectroscopy and least square support vector machines regression for determination of vitamin C of mandarin fruit |
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: 20190726 |
|
RJ01 | Rejection of invention patent application after publication |