CN115236024A - Training method, determination method and device for model for determining content of total acids and total esters in wine - Google Patents
Training method, determination method and device for model for determining content of total acids and total esters in wine Download PDFInfo
- Publication number
- CN115236024A CN115236024A CN202210769887.5A CN202210769887A CN115236024A CN 115236024 A CN115236024 A CN 115236024A CN 202210769887 A CN202210769887 A CN 202210769887A CN 115236024 A CN115236024 A CN 115236024A
- Authority
- CN
- China
- Prior art keywords
- wine
- total
- wine sample
- content
- training
- 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 618
- 150000002148 esters Chemical class 0.000 title claims abstract description 203
- 239000002253 acid Substances 0.000 title claims abstract description 201
- 238000012549 training Methods 0.000 title claims abstract description 176
- 238000000034 method Methods 0.000 title claims abstract description 69
- 150000007513 acids Chemical class 0.000 title claims description 38
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 104
- 239000000126 substance Substances 0.000 claims abstract description 57
- 238000012795 verification Methods 0.000 claims description 55
- 238000001228 spectrum Methods 0.000 claims description 45
- 238000005259 measurement Methods 0.000 claims description 33
- 230000002159 abnormal effect Effects 0.000 claims description 32
- 238000012216 screening Methods 0.000 claims description 23
- 238000012937 correction Methods 0.000 claims description 16
- 238000010238 partial least squares regression Methods 0.000 claims description 15
- 230000003595 spectral effect Effects 0.000 claims description 13
- 238000000513 principal component analysis Methods 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 10
- 238000010200 validation analysis Methods 0.000 description 8
- HEMHJVSKTPXQMS-UHFFFAOYSA-M Sodium hydroxide Chemical compound [OH-].[Na+] HEMHJVSKTPXQMS-UHFFFAOYSA-M 0.000 description 6
- 238000004140 cleaning Methods 0.000 description 6
- 239000000796 flavoring agent Substances 0.000 description 6
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000002835 absorbance Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000004448 titration Methods 0.000 description 3
- 239000003513 alkali Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000003153 chemical reaction reagent Substances 0.000 description 2
- 235000019634 flavors Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- KJFMBFZCATUALV-UHFFFAOYSA-N phenolphthalein Chemical compound C1=CC(O)=CC=C1C1(C=2C=CC(O)=CC=2)C2=CC=CC=C2C(=O)O1 KJFMBFZCATUALV-UHFFFAOYSA-N 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 239000002585 base Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003330 mid-infrared imaging Methods 0.000 description 1
- 150000007524 organic acids Chemical class 0.000 description 1
- 238000010992 reflux Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 235000020097 white wine Nutrition 0.000 description 1
Images
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
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 application provides a training method, a measuring method and a device for a model for measuring the contents of total acid and total ester in wine. The training method comprises the following steps: acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the first wine sample set, chemical determination content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in the second wine sample set and chemical determination content of total ester of each wine sample in the second wine sample set; the method comprises the steps of taking mid-infrared spectrum information of all wine samples in a first wine sample set and mid-infrared spectrum information of all wine samples in a second wine sample set as input of a partial least square regression model, taking chemical determination content of all wine samples in the first wine sample set and chemical determination content of all wine samples in the second wine sample set as output of the partial least square regression model, training the partial least square regression model, and obtaining a determination wine total acid and total ester content model which is accurately and quickly used for detecting the total acid and total ester content in wine.
Description
Technical Field
The application relates to the technical field of detection of contents of total acids and total esters of wine, in particular to a training method, a determination method and a device for a model for determining contents of total acids and total esters of wine.
Background
The total acid and the total ester are used as key physicochemical indexes of the wine and play an important role in the aspects of the aroma generation and the flavor coordination of the wine body. Generally, the calibration of total acid and total ester in wine is carried out by chemical titration. The calibration method has the advantages of complex and time-consuming detection process, large consumption of chemical reagents and high technical dependence on operators, and is difficult to meet the urgent requirement for rapidly detecting and calibrating total acid and total ester in wine.
Therefore, a method for rapidly and efficiently measuring the alcohol is needed.
Disclosure of Invention
The application provides a training method, a measuring method and a device for a model for measuring the contents of total acid and total ester in wine.
The first aspect of the application provides a training method for a model for measuring the contents of total acids and total esters in wine, and the training method comprises the following steps:
acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the first wine sample set, chemical determination content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in the second wine sample set and chemical determination content of total ester of each wine sample in the second wine sample set;
taking mid-infrared spectrum information of each wine sample in the first wine sample set and mid-infrared spectrum information of each wine sample in the second wine sample set as input of a first partial least square regression model, taking the chemical determination content of each wine sample total acid in the first wine sample set and the chemical determination content of each wine sample total ester in the second wine sample set as output of the first partial least square regression model, training the first partial least square regression model to obtain a model for determining the content of the wine total acid and the total ester, wherein the model for determining the content of the wine total acid and the total ester comprises a mid-infrared determination wine total acid content model and a mid-infrared determination wine total ester content model,
the first wine sample set and the second wine sample set are both from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acid through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total ester through the original wine sample set.
The training method for determining the content model of the total acids and the total esters of the wine provided by the first aspect of the application can obtain the model for determining the content of the total acids and the total esters of the wine, which is accurately and quickly used for detecting the content of the total acids and the total esters in the wine, so that the contents of the total acids and the total esters in the wine, especially the new wine, can be quickly and efficiently determined, the detection efficiency is improved, and the detection accuracy is ensured.
In some optional embodiments of the first aspect of the present application, the step of screening the raw wine sample set for wine samples with abnormal total acid in the training method comprises:
acquiring first pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in the original wine sample set and chemical determination content of total acid of all wine samples in the original wine sample set;
taking mid-infrared spectrum information of all the wine samples in the original wine sample set as input of a second partial least square method regression model, taking the chemical determination content of all the wine samples in the original wine sample set as output of the second partial least square method regression model, training the second partial least square method regression model, and obtaining a pre-training model for determining the total acid content of the wine;
based on the outliers in the pre-training model for determining the total acid content of the wine, the wine samples with abnormal total acid are screened from the original wine sample set to obtain a first wine sample set.
In some alternative embodiments of the first aspect of the present application, the step of screening the raw wine sample set for wine samples with total ester abnormalities in the training method comprises:
acquiring second pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all the wine samples in the original wine sample set and chemical determination content of total esters of all the wine samples in the original wine sample set;
taking mid-infrared spectrum information of all the wine samples in the original wine sample set as input of a third partial least square regression model, taking the chemical determination content of all the wine samples in the original wine sample set as output of the third partial least square regression model, and training the third partial least square regression model to obtain a pre-training model for determining the total ester content of the wine;
and (4) screening out the wine samples with abnormal total ester from the original wine sample set based on the outlier in the pre-training model for determining the content of the total ester of the wine to obtain a second wine sample set.
In some optional embodiments of the first aspect of the present application, the step of training the first partial least squares regression model further comprises:
dividing the first wine sample set into a first correction set, a first verification set and a first external verification set by adopting principal component analysis, and performing primary training on a first partial least square regression model based on the first correction set and the first verification set to obtain a plurality of training submodels for determining total acidity of wine;
taking the mid-infrared spectrum information of each wine sample in the first external verification set as the input of each wine total acid determination training submodel, and obtaining the total acid content predicted by the training submodel of each wine sample in the first external verification set output by each wine total acid determination training submodel;
dividing the second wine sample set into a second correction set, a second verification set and a second external verification set by adopting principal component analysis, and performing primary training on the first partial least square regression model based on the second correction set and the second verification set to obtain a plurality of wine total ester determination training submodels;
taking mid-infrared spectrum information of the second external verification concentrated wine sample as input of each measured wine total ester training submodel to obtain the predicted total ester content of the training submodel of each wine sample in the second external verification concentrated output by each measured wine total ester training submodel;
predicting the total acid content according to the training submodels of all wine samples in a first external verification set output by all wine total acid determination training submodels, predicting the total ester content according to the training submodels of all wine samples in a second external verification set output by all wine total ester determination training submodels, determining the chemical determination content of all wine samples in the first external verification set and the chemical determination content of all wine samples in a second external verification set, and determining the number of main components by taking a preset error range as a determination basis to obtain a wine total acid determination model and a wine total ester determination model comprising a middle infrared wine total acid determination model and a middle infrared wine total ester determination model.
In some optional embodiments of the first aspect of the present application, the mid-infrared spectrum information of each wine sample in the set of raw wine samples has a continuous spectrum with a wavelength range of 400cm -1 ～4000cm -1 。
In some alternative embodiments of the first aspect of the present application, the first spectrum in the mid-infrared spectral information of each of the wine samples in the first wine sample set is a break spectrum and the second spectrum in the mid-infrared spectral information of each of the wine samples in the second wine sample set is also a break spectrum.
In some alternative embodiments of the first aspect of the present application, the first spectrum includes two bands spaced apart by 4000cm, respectively -1 ～3791cm -1 And 2943cm -1 ～1670cm -1 。
In some alternative embodiments of the first aspect of the present application, the second spectrum comprises three bands spaced apart by 4000cm each -1 ～3791cm -1 ，2727cm -1 ～2518cm -1 And 2090cm -1 ～1670cm -1 。
In some optional embodiments of the first aspect of the present application, the bands in the first spectrum and the second spectrum are obtained by performing a combined interval partial least squares operation according to mid-infrared spectrum information of each wine sample in the original wine sample set, a chemically-determined content of each wine sample in the original wine sample set, and comparing cross-validation mean square deviations, where the bands of continuous spectra in the mid-infrared spectrum information of each wine sample in the original wine sample set are equally divided into N sub-bands, and N is greater than 1;
preferably, N takes the value 17, 55 or 85.
In a second aspect, the present application provides a method for determining the content of total acids and total esters in wine, the method comprising:
acquiring mid-infrared spectrum information of a wine sample to be detected;
inputting the mid-infrared spectrum information of the wine sample to be measured into a model for measuring the content of the total acid and the total ester of the wine to obtain the content of the total acid and the total ester of the wine sample output by the model for measuring the content of the total acid and the total ester of the wine,
wherein the model for determining the content of total acids and total esters is obtained by pre-training the training method in the first aspect of the present application.
In a third aspect, the present application provides an apparatus for training a model for determining total acid and total ester content of wine, the apparatus comprising:
the first acquisition unit is used for acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the first wine sample set, chemical measurement content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in the second wine sample set and chemical measurement content of total ester of each wine sample in the second wine sample set;
and the model training unit is used for taking the mid-infrared spectrum information of the wine samples in the first wine sample set and the mid-infrared spectrum information of the wine samples in the second wine sample set as the input of a first partial least square regression model, taking the chemical measurement content of the total acids of the wine samples in the first wine sample set and the chemical measurement content of the total esters of the wine samples in the second wine sample set as the output of the first partial least square regression model, training the first partial least square regression model to obtain a total acid and total ester content measurement model, wherein the total acid and total ester content measurement model comprises a total acid content measurement model and a total ester content measurement model, the first wine sample set and the second wine sample set are from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acids through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total esters through the original wine sample set.
In a fourth aspect, the present application provides an apparatus for determining the total acid and total ester content of wine, the apparatus comprising:
the second acquisition unit is used for acquiring mid-infrared spectrum information of the wine sample to be detected;
and the wine sample total acid and total ester determination unit is used for inputting the mid-infrared spectrum information of the wine sample to be determined into a wine total acid and total ester content determination model, and acquiring the contents of the wine sample total acid and total ester output by the wine total acid and total ester content determination model, wherein the wine total acid and total ester content determination model is obtained by pre-training the device in the third aspect of the application.
Drawings
FIG. 1 is a schematic flow chart of a training method for determining the content of total acids and total esters in the wine in the first embodiment of the first aspect of the present application;
FIG. 2 is a mid-infrared spectrum of a wine sample. (ii) a
FIG. 3 is a model of the mid-infrared determination of total acid content of wine according to an embodiment of the first aspect of the present application;
FIG. 4 is a mid-infrared determined total acid content of wine sub-model obtained via the first validation set in the mid-infrared determined total acid content of wine model of FIG. 3;
FIG. 5 is a mid-infrared determined total acid content of wine sub-model obtained via the first calibration set in the mid-infrared determined total acid content of wine model of FIG. 3;
FIG. 6 is a model of the mid-infrared determination of total ester content of wine according to one embodiment of the first aspect of the present application;
FIG. 7 is a mid-infrared determined total ester of wine content sub-model obtained via a second validation set in the mid-infrared determined total ester of wine content model of FIG. 6;
FIG. 8 is a mid-infrared determined total ester content sub-model obtained via a second calibration set from the mid-infrared determined total ester content model of FIG. 6;
FIG. 9 is a schematic flow chart of a method for determining total acid and total ester content of wine according to an embodiment of the second aspect of the present application.
Detailed Description
The present application will be described in further detail below with reference to the drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The inventor researches in the field of wine detection and finds that the total acid and the total ester are key physicochemical indexes of the wine, particularly the Maotai-flavor new white wine, and play an important role in the aspects of wine body aroma development and flavor coordination. At present, the method specified by the traditional national standard for the two indexes is a chemical titration method, has the defects of complex and time-consuming detection process, consumption of chemical reagents and the like, has certain requirements on the technical level of operators, needs to be equipped with a large number of professional technicians every day to consume time for repeated analysis, and cannot meet the urgent requirement on quick detection of the sauce-flavor new wine.
In view of the above, the present application aims to provide a novel training method, a novel determination method and a novel determination device for a model for determining the content of total acids and total esters in wine. The technical contents provided by the present application are described below with reference to examples.
As shown in fig. 1, the present application provides, in a first aspect, a training method for a model for determining total acid and total ester content of wine, the training method comprising:
s10: acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the first wine sample set, chemical determination content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in the second wine sample set and chemical determination content of total ester of each wine sample in the second wine sample set;
s20: taking mid-infrared spectrum information of each wine sample in the first wine sample set and mid-infrared spectrum information of each wine sample in the second wine sample set as input of a first partial least square regression model, taking the chemical determination content of each wine sample total acid in the first wine sample set and the chemical determination content of each wine sample total ester in the second wine sample set as output of the first partial least square regression model, training the first partial least square regression model to obtain a model for determining the content of the wine total acid and the total ester, wherein the model for determining the content of the wine total acid and the total ester comprises a mid-infrared determination wine total acid content model and a mid-infrared determination wine total ester content model,
wherein the first wine sample set and the second wine sample set are both from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acid through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total ester through the original wine sample set.
The training method for determining the total acid and total ester content model of the wine provided by the first aspect of the application can obtain the total acid and total ester content model of the wine which is accurately and quickly used for detecting the total acid and total ester content of the wine, so that the total acid and total ester content of the wine, especially fresh wine, can be quickly and efficiently determined, the detection efficiency is improved, the detection accuracy is ensured, the warehousing acceptance efficiency of the sauce-flavor fresh wine is also improved, the labor cost is reduced, and the production economic benefit is ensured.
The first Partial Least Squares Regression model mentioned in the first aspect of the present application belongs to the PLS (Partial Least Squares Regression) Regression model.
In some optional embodiments of the first aspect of the present application, in the step of obtaining the chemical content measurement of the total acids and total esters of each wine sample in the wine sample set from which the abnormal sample has been screened, the total acids of each wine sample in the wine sample set from which the abnormal sample has been screened are measured by using the method for measuring the total acids in the food of national standard for food safety of GB 12456-2021. And (3) determining the total esters of all wine samples in the wine sample set with abnormal samples screened out by adopting a GB/T10345-2007 white spirit analysis method.
The specific measurement principle is as follows: organic acid in the white spirit is neutralized and titrated by adopting sodium hydroxide solution by taking phenolphthalein as an indicator, and the content of total acid is calculated by the amount of the sodium hydroxide consumed standard titration solution. The determination of total ester uses alkali to neutralize free acid in sample, then adds a certain amount of alkali accurately, heats and refluxes to saponify ester. The total ester content was calculated from the amount of base consumed.
In some optional embodiments of the first aspect of the present application, the mid-infrared spectrum information of the wine sample is obtained by scanning the sample with an infrared spectrometer to obtain a mid-infrared spectrum, and then converting the scanned mid-infrared spectrum into a plurality of spectral data points of the reaction mid-infrared spectrum by using Unscrambler X10.3 spectral analysis software. Each spectral data point consists of both (wave number, absorbance). The specific intermediate infrared spectrum collection process comprises the steps of starting up and preheating the infrared spectrometer, cleaning and zeroing equipment, pouring a wine sample into a test cup, automatically extracting the sample by a pump, cleaning a pipeline, and collecting the spectrum of the sampleThe wave number of the scanning is in the range of 400cm -1 ～4000cm -1 And automatically extracting and feeding sample for three times for each wine sample, and scanning to obtain the spectrum of the wine sample.
In some optional embodiments of the first aspect of the present application, the step of screening the raw wine sample set for wine samples with total acid anomalies in the training method comprises:
acquiring first pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in the original wine sample set and chemical determination content of total acid of all wine samples in the original wine sample set;
taking mid-infrared spectrum information of all the wine samples in the original wine sample set as input of a second partial least square method regression model, taking the chemical determination content of all the wine samples in the original wine sample set as output of the second partial least square method regression model, training the second partial least square method regression model, and obtaining a pre-training model for determining the total acid content of the wine;
based on the outliers in the pre-training model for determining the total acid content of the wine, the wine samples with abnormal total acid are screened from the original wine sample set to obtain a first wine sample set.
In some examples of these embodiments, the second Partial Least Squares Regression model belongs to a PLS (Partial Least Squares Regression) Regression model.
In some examples of these embodiments, after obtaining the pre-training model for determining total acid content, the model is found to predict a wine sample with poor total acid content effect, i.e. a wine sample with abnormal total acid content, and the corresponding sample point of the wine sample generally has a large or positive or negative residual error, and then the sample point is defined as an outlier in the pre-training model for determining total acid content, and the wine sample with poor total acid content prediction effect in the model (i.e. the wine sample with abnormal total acid content) is screened out from the original wine sample set, so as to obtain a first wine sample set.
In some optional embodiments of the first aspect of the present application, the step of screening the raw wine sample set for wine samples with total ester abnormalities in the training method comprises:
acquiring second pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all the wine samples in the original wine sample set and chemical determination content of total esters of all the wine samples in the original wine sample set;
taking mid-infrared spectrum information of all wine samples in the original wine sample set as input of a third partial least square regression model, taking chemical determination content of all wine samples in the original wine sample set as output of the third partial least square regression model, and training the third partial least square regression model to obtain a pre-training model for determining the total ester content of the wine;
and screening out the wine samples with abnormal total esters from the original wine sample set based on the outlier in the pre-training model for determining the content of the total esters of the wine to obtain a second wine sample set.
In some examples of these embodiments, the third Partial Least Squares Regression model belongs to a PLS (Partial Least Squares Regression) Regression model.
In some examples of these embodiments, after obtaining the pre-training model for determining total ester content of wine, it is found that the model predicts a wine sample with poor total ester content effect, i.e. a wine sample with abnormal total ester, and the corresponding sample point of the wine sample usually has a large or positive or negative residual error, and then the sample point is defined as an outlier in the pre-training model for determining total ester content of wine, and the wine sample with poor total ester content prediction effect in the model (i.e. the wine sample with abnormal total ester) is screened from the original wine sample set, so as to obtain a second wine sample set.
In some optional embodiments of the first aspect of the present application, the step S20 of training the first partial least squares regression model further includes:
s21: dividing the first wine sample set into a first correction set, a first verification set and a first external verification set by adopting principal component analysis, and performing primary training on a first partial least square regression model based on the first correction set and the first verification set to obtain a plurality of training submodels for determining total acidity of wine;
s22: taking the mid-infrared spectrum information of each wine sample in the first external verification set as the input of each wine total acid determination training submodel, and obtaining the total acid content predicted by the training submodel of each wine sample in the first external verification set output by each wine total acid determination training submodel;
s23: dividing the second wine sample set into a second correction set, a second verification set and a second external verification set by adopting principal component analysis, and performing primary training on the first partial least square regression model based on the second correction set and the second verification set to obtain a plurality of sub models for determining the total esters of wine;
s24: taking mid-infrared spectrum information of the second external verification concentrated wine sample as input of each measured wine total ester training submodel to obtain the predicted total ester content of the training submodel of each wine sample in the second external verification concentrated output by each measured wine total ester training submodel;
s25: predicting the total acid content according to the training submodels of all wine samples in a first external verification set output by all wine total acid determination training submodels, predicting the total ester content according to the training submodels of all wine samples in a second external verification set output by all wine total ester determination training submodels, determining the chemical determination content of all wine samples in the first external verification set and the chemical determination content of all wine samples in a second external verification set, and determining the number of main components by taking a preset error range as a determination basis to obtain a wine total acid determination model and a wine total ester determination model comprising a middle infrared wine total acid determination model and a middle infrared wine total ester determination model.
In some optional embodiments of the first aspect of the present application, the mid-infrared spectrum information of each of the wine samples in the set of raw wine samples is a continuous spectrum with a wavelength range of 400cm -1 ～4000cm -1 。
In some examples of these embodiments, the mid-infrared spectral information of each of the wine samples in the set of raw wine samples is obtained by scanning the sample with an infrared spectrometer to obtain a mid-infrared spectrum and converting the scanned mid-infrared spectrum into a plurality of spectral data points of a reaction mid-infrared spectrum using Unscrambler X10.3 spectral analysis software. Each spectral data point consists of both (wave number, absorbance). The specific intermediate infrared spectrum acquisition process comprises the steps of preheating the infrared spectrometer after starting up, cleaning and zeroing equipment, pouring a wine sample into a test cup, automatically extracting the sample by a pump, cleaning a pipeline, and then performing sample spectrum acquisition, wherein the scanning wave number range is 400cm -1 ～4000cm -1 And automatically extracting and feeding sample for three times for each wine sample, and scanning to obtain the spectrum of the wine sample.
In these examples, it is not clear which band ranges in the mid-infrared spectrum information of the wine sample correspond to the sensitivity of the total acid content of the wine sample, nor which band ranges in the mid-infrared spectrum of the wine sample correspond to the sensitivity of the total ester content of the wine sample, before obtaining the model for determining the total acid and total ester content of the wine. Therefore, the mid-infrared spectrum information of each wine sample in the original wine sample set is that the spectrum is continuous and the wave band range is 400cm -1 ～4000cm -1 (i.e., the full mid-infrared spectral band).
In some alternative embodiments of the first aspect of the present application, the first spectrum in the mid-infrared spectral information of each of the wine samples in the first wine sample set is a break spectrum and the second spectrum in the mid-infrared spectral information of each of the wine samples in the second wine sample set is also a break spectrum.
In some optional embodiments of the first aspect of the present application, the first spectrum comprises two bands spaced apart by 4000cm each -1 ～3791cm -1 And 2943cm -1 ～1670cm -1 . In these examples, 4000cm -1 ～3791cm -1 And 2943cm -1 ～1670cm -1 The band range of the mid-infrared spectrum sensitive to the total acid content of the wine sample. The selection of the waveband range sensitive to the total acid content of the wine sample greatly improves the efficiency and accuracy of model training, and a better model for measuring the total acid content of the wine by the intermediate infrared is obtained.
In some alternative embodiments of the first aspect of the present application, the second spectrum comprises three bands spaced apart by 4000cm each -1 ～3791cm -1 ，2727cm -1 ～2518cm -1 And 2090cm -1 ～1670cm -1 . In these examples, 4000cm -1 ～3791cm -1 ，2727cm -1 ～2518cm -1 And 2090cm -1 ～1670cm -1 The band range of the mid-infrared spectrum sensitive to the total acid content of the wine sample. The selection of the band range sensitive to the total ester content of the wine sample greatly improves the efficiency of model trainingAnd the accuracy, a better model for measuring the total ester content of the wine by the mid-infrared is obtained.
In some optional embodiments of the first aspect of the present application, the bands in the first spectrum and the second spectrum are obtained by performing a combined interval partial least squares operation according to mid-infrared spectrum information of each wine sample in the original wine sample set, a chemically measured content of each wine sample total acid in the original wine sample set, and a chemically measured content of each wine sample total ester in the original wine sample set, and comparing cross-validation mean square deviations, wherein the bands of continuous spectra in the mid-infrared spectrum information of each wine sample in the original wine sample set are equally divided into N sub-bands, and N is greater than 1.
In some examples of these embodiments, N takes the value 17, 55, or 85.
[ EXAMPLES OF THE PREFERRED EMBODIMENT OF THE FIRST AREA OF THE APPLICATION ]
Taking 718 Maotai-flavor liquor original liquor samples to form an original liquor sample set.
Acquiring first pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in an original wine sample set and chemical determination content of total acid of all wine samples in the original wine sample set, the mid-infrared spectrum information of all wine samples in the original wine sample set is scanned on the samples by an infrared spectrometer to obtain a mid-infrared spectrum, and the scanned mid-infrared spectrum is converted into a plurality of spectrum data points of a reaction mid-infrared spectrum by Unscrambler X10.3 spectrum analysis software. Each spectral data point consists of both (wave number, absorbance). The specific intermediate infrared spectrum collection process comprises the steps of starting up and preheating the infrared spectrometer, cleaning and zeroing equipment, pouring a wine sample into a test cup, automatically extracting the sample by a pump, cleaning a pipeline, and then carrying out spectrum collection on the sample, wherein the scanning waveband is 400cm -1 ～4000cm -1 And automatically extracting and sampling each wine sample for three times and scanning to obtain the spectrum of the concentrated wine sample of the original wine sample. FIG. 2 shows a mid-infrared spectrum of a wine sample.
Taking mid-infrared spectrum information of all the wine samples in the original wine sample set as input of a second partial least square method regression model, taking the chemical determination content of all the wine samples in the original wine sample set as output of the second partial least square method regression model, training the second partial least square method regression model, and obtaining a pre-training model for determining the total acid content of the wine;
based on the determination of the outliers in the pre-training model of the total acid content of the wine, 26 wine samples with abnormal total acid are screened from the original wine sample set to obtain a first wine sample set, and the first wine sample set comprises 692 wine samples.
In some optional embodiments of the first aspect of the present application, the step of screening the raw wine sample set for wine samples with total ester abnormalities in the training method comprises:
acquiring second pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in the original wine sample set and chemical determination content of total esters of all wine samples in the original wine sample set;
taking mid-infrared spectrum information of all the wine samples in the original wine sample set as input of a third partial least square regression model, taking the chemical determination content of all the wine samples in the original wine sample set as output of the third partial least square regression model, and training the third partial least square regression model to obtain a pre-training model for determining the total ester content of the wine;
based on the determination of the outliers in the pre-training model of the total ester content of the wine, 30 wine samples with abnormal total esters are screened from the original wine sample set to obtain a second wine sample set, wherein the second wine sample set comprises 688 wine samples.
And selecting a first spectrum in the mid-infrared spectrum information of each wine sample in the first wine sample set, and selecting a second spectrum in the mid-infrared spectrum information of each wine sample in the second wine sample set.
And the wave bands in the first spectrum and the second spectrum are obtained by performing combined interval partial least square operation according to the mid-infrared spectrum information of each wine sample in the original wine sample set, the chemical measurement content of each wine sample in the original wine sample set and the chemical measurement content of each wine sample total ester in the original wine sample set, and comparing and interactively verifying the mean square error, wherein the wave bands of continuous spectra in the mid-infrared spectrum information of each wine sample in the original wine sample set are equally divided into N sub-wave bands, and the value of N is 17, 55 or 85.
The first spectrum comprises two spaced bands of 4000cm each -1 ～3791cm -1 And 2943cm -1 ～1670cm -1 。
The second spectrum comprises three spaced bands of 4000cm each -1 ～3791cm -1 ，2727cm -1 ～2518cm -1 And 2090cm -1 ～1670cm -1 。
Dividing a first wine sample set into a first calibration set, a first validation set and a first external validation set by using principal component analysis, and carrying out primary training on a first partial least square regression model based on the first calibration set and the first validation set to obtain a plurality of training submodels for determining total alcohol content, wherein the ratio of the number of samples of the first calibration set to the number of samples of the first calibration set of the first validation set is 2.
And (3) taking the mid-infrared spectrum information of each wine sample in the first external verification set as the input of each wine total acid determination training submodel, and obtaining the total acid content predicted by the training submodel of each wine sample in the first external verification set output by each wine total acid determination training submodel.
And dividing a second wine sample set into a second correction set, a second verification set and a second external verification set by adopting principal component analysis, and carrying out primary training on a first partial least square regression model based on the second correction set and the second verification set to obtain a plurality of wine total ester training submodels, wherein the ratio of the number of samples of the second correction set to the number of samples of a first correction set of the second verification set is 2.
And (4) taking the mid-infrared spectrum information of the second external verification concentrated wine sample as the input of each measured wine total ester training submodel to obtain the predicted total ester content of each wine sample training submodel output by each measured wine total ester training submodel.
Predicting the total acid content according to the training submodel of each wine sample in the first external verification set output by each wine total acid determination training submodel, predicting the total ester content according to the training submodel of each wine sample in the second external verification set output by each wine total ester determination training submodel, determining the chemical determination content of each wine sample in the first external verification set and the chemical determination content of each wine sample in the second external verification set, and determining the number of main components by taking a preset error range as a determination basis to obtain a wine total acid determination model and a total ester determination model comprising a mid-infrared wine total acid determination model and a mid-infrared wine total ester determination model.
FIG. 3 shows a model for measuring total acid content of wine by using mid-infrared light in this embodiment. FIG. 4 shows the mid-infrared determined total acid content wine sub-model obtained via the first validation set in the mid-infrared determined total acid content wine model. FIG. 5 illustrates a mid-infrared determined total acid content sub-model via the first calibration set in the mid-infrared determined total acid content wine model.
FIG. 6 shows a model of the mid-infrared determination of total ester content of wine for this particular example. FIG. 7 shows the mid-infrared determined total ester content sub-model via the second validation set in the mid-infrared determined total ester content model. FIG. 8 shows a mid-infrared determined total ester content submodel obtained via the second calibration set in the mid-infrared determined total ester content model.
The model for measuring the total acid content of the wine by the intermediate infrared shown in figure 3 and the model for measuring the total ester content of the wine by the intermediate infrared shown in figure 6 jointly form a model for measuring the total acid content and the total ester content of the wine.
In this specific embodiment, the predetermined measurement error range of the total acid content is between-3.52% and 5.32%, the determined number of principal components is 15, and table 1 shows the verification result of the first external verification set on the obtained mid-infrared measurement wine total acid content model.
TABLE 1
In this specific embodiment, the predetermined measurement error range of the total ester content is between-9.50% and 4.05%, the determined number of principal components is 16, and table 2 shows the verification result of the obtained model for measuring the total acid content of the wine by using the mid-infrared method by using the second external verification set.
TABLE 2
Table 3 shows the specific parameters of the model for determining the total acid and total ester content of the wine in the specific example, wherein RMSE is root mean Square error, and R-Square is correlation coefficient R 2 ，R 2 The closer to 1, the closer to 0 the RMSE, the better the model.
TABLE 3
As can be seen from table 3, the model for determining the content of total acids and total esters trained by the training method for determining the content of total acids and total esters provided in the first aspect of the present application can perform rapid and accurate detection and calibration on the content of total acids and total esters in wine.
As shown in fig. 9, the second aspect of the present application provides a method for determining the content of total acids and total esters in wine, the method comprising:
acquiring mid-infrared spectrum information of a wine sample to be detected;
inputting the mid-infrared spectrum information of the wine sample to be measured into a model for measuring the content of the total acid and the total ester of the wine to obtain the content of the total acid and the total ester of the wine sample output by the model for measuring the content of the total acid and the total ester of the wine,
wherein, the model for measuring the total acid and total ester content of the wine is obtained by training in advance by the training method provided by the first aspect of the application.
In a third aspect, the present application provides an apparatus for training a model for determining the contents of total acids and total esters in wine, the apparatus comprising:
the first acquisition unit is used for acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the first wine sample set, chemical measurement content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in the second wine sample set and chemical measurement content of total ester of each wine sample in the second wine sample set;
and the model training unit is used for taking the mid-infrared spectrum information of all the wine samples in the first wine sample set and the mid-infrared spectrum information of all the wine samples in the second wine sample set as the input of a first partial least square regression model, taking the chemical measurement content of the total acids of all the wine samples in the first wine sample set and the chemical measurement content of the total esters of all the wine samples in the second wine sample set as the output of the first partial least square regression model, training the first partial least square regression model to obtain a measurement wine total acids and total ester content model, wherein the measurement wine total acids and total ester content model comprises a mid-infrared measurement wine total acids content model and a mid-infrared measurement wine total ester content model, the first wine sample set and the second wine sample set are both from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acids through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total esters through the original wine sample set.
In a fourth aspect, the present application provides an apparatus for determining the total acid and total ester content of a wine, the apparatus comprising:
the second acquisition unit is used for acquiring mid-infrared spectrum information of the wine sample to be detected;
and the wine sample total acid and total ester determination unit is used for inputting the mid-infrared spectrum information of the wine sample to be determined into a wine total acid and total ester content determination model and acquiring the content of the wine sample total acid and total ester output by the wine total acid and total ester content determination model, wherein the wine total acid and total ester content determination model is obtained by pre-training the device for training the wine total acid and total ester content determination model provided by the third aspect of the application.
In a second aspect, the present application provides a method for determining the content of total acids and total esters in wine, the method comprising:
acquiring mid-infrared spectrum information of a wine sample to be detected;
inputting the mid-infrared spectrum information of the wine sample to be detected into the wine total acid and total ester content measuring model to obtain the contents of the wine sample total acid and total ester output by the wine total acid and total ester content measuring model,
wherein, the model for measuring the content of the total alcohol acid and the total ester is obtained by pre-training the training method for measuring the content of the total alcohol acid and the total ester as provided by the first aspect of the application.
In a third aspect of the present application, there is provided an apparatus for training a model for determining the total acid and total ester content of wine, the apparatus comprising:
the first acquisition unit is used for acquiring training data, wherein the training data comprises mid-infrared spectrum information of each wine sample in the wine sample set with abnormal samples screened out and chemical measurement content of total acid and total ester of each wine sample in the wine sample set with abnormal samples screened out;
and the model training unit is used for taking the mid-infrared spectrum information of each wine sample in the wine sample set with the abnormal samples screened out as the input of a first partial least square regression model, taking the chemical measurement content of the total acid and the total ester of each wine sample in the wine sample set with the abnormal samples screened out as the output of the first partial least square regression model, and training the first partial least square regression model to obtain a model for measuring the content of the total acid and the total ester of the wine.
In a fourth aspect of the present application, there is provided an apparatus for determining the total acid and total ester content of wine, the apparatus comprising:
the second acquisition unit is used for acquiring mid-infrared spectrum information of the wine sample to be detected;
a wine sample total acid and total ester determination unit, configured to input mid-infrared spectrum information of a wine sample to be determined into a wine total acid and total ester content determination model, and obtain contents of total acid and total ester of the wine sample output by the wine total acid and total ester content determination model, where the wine total acid and total ester content determination model is obtained by pre-training the apparatus according to claim 9.
It should be understood that at least a portion of the steps in the method flow diagrams shown in fig. 1 and 9 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be alternated or performed with other steps or at least a portion of the sub-steps or stages of other steps.
The above-described embodiments are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (12)
1. A training method for a model for measuring the content of total acids and total esters in wine is characterized by comprising the following steps:
acquiring training data, wherein the training data comprises mid-infrared spectrum information of all wine samples in a first wine sample set, chemical determination content of total acids of all wine samples in the first wine sample set, mid-infrared spectrum information of all wine samples in a second wine sample set and chemical determination content of total esters of all wine samples in the second wine sample set;
taking the mid-infrared spectrum information of each wine sample in the first wine sample set and the mid-infrared spectrum information of each wine sample in the second wine sample set as the input of a first partial least square regression model, taking the chemical determination content of each wine sample in the first wine sample set and the chemical determination content of each wine sample total ester in the second wine sample set as the output of the first partial least square regression model, training the first partial least square regression model to obtain a model for determining the content of the total acids and the total esters of the wine, wherein the model for determining the content of the total acids and the total esters of the wine comprises a mid-infrared determination wine total acid content model and a mid-infrared determination wine total ester content model,
wherein the first wine sample set and the second wine sample set are both from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acid through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total ester through the original wine sample set.
2. The training method of claim 1, wherein the step of screening out wine samples with total acid anomalies from the set of raw wine samples in the training method comprises:
acquiring first pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in an original wine sample set and chemical determination content of total acid of all wine samples in the original wine sample set;
taking the mid-infrared spectrum information of each wine sample in the original wine sample set as the input of a second partial least square regression model, taking the chemical determination content of the total acid of each wine sample in the original wine sample set as the output of the second partial least square regression model, and training the second partial least square regression model to obtain a pre-training model for determining the total acid content of the wine;
and screening out wine samples with abnormal total acid from the original wine sample set based on the outliers in the pre-training model for determining the total acid content of the wine to obtain the first wine sample set.
3. Training method according to claim 1, wherein the step of screening out the wine sample with total ester abnormality from the set of raw wine samples in the training method comprises:
acquiring second pre-training data, wherein the pre-training data comprises mid-infrared spectrum information of all wine samples in the original wine sample set and chemical determination content of total esters of all wine samples in the original wine sample set;
taking mid-infrared spectrum information of all wine samples in the original wine sample set as input of a third partial least square regression model, taking chemical determination content of all wine samples in the original wine sample set as output of the third partial least square regression model, and training the third partial least square regression model to obtain a pre-training model for determining the content of the total esters of the wine;
and screening out the wine samples with abnormal total esters from the original wine sample set based on the outliers in the pre-training model for determining the content of the total esters of the wine to obtain the second wine sample set.
4. A training method as claimed in claim 1, wherein the step of training the first partial least squares regression model further comprises:
dividing the first wine sample set into a first correction set, a first verification set and a first external verification set by adopting principal component analysis, and carrying out primary training on the first partial least square regression model based on the first correction set and the first verification set to obtain a plurality of sub models for determining the total acidity of wine;
taking the mid-infrared spectrum information of each wine sample in the first external verification set as the input of each wine total acid determination training submodel to obtain the total acid content predicted by the training submodel of each wine sample in the first external verification set output by each wine total acid determination training submodel;
dividing the second wine sample set into a second correction set, a second verification set and a second external verification set by adopting principal component analysis, and performing primary training on the first partial least square regression model based on the second correction set and the second verification set to obtain a plurality of measured wine total ester training submodels;
taking the mid-infrared spectrum information of the second externally verified concentrated wine sample as the input of each measured wine total ester training submodel to obtain the predicted total ester content of each wine sample in the second externally verified concentrated wine sample output by each measured wine total ester training submodel;
and determining the number of main components according to the training submodel predicted total acid content of each wine sample in the first external verification set output by each wine total acid measurement training submodel, the training submodel predicted total ester content of each wine sample in the second external verification set output by each wine total ester measurement training submodel, the chemical measured content of each wine sample in the first external verification set and the chemical measured content of each wine sample in the second external verification set, and taking a preset error range as a determination basis to obtain the wine total acid measurement model and the total ester measurement model comprising the mid-infrared wine total acid measurement model and the mid-infrared wine total ester measurement model.
5. The training method of any one of claims 1 to 3, wherein the mid-infrared spectrum information of each of the wine samples in the set of raw wine samples has a continuous spectrum with a wavelength range of 400cm -1 ～4000cm -1 。
6. Training method according to claim 5, wherein a first spectrum in the mid-infrared spectral information of each wine sample in the first set of wine samples is a break spectrum and a second spectrum in the mid-infrared spectral information of each wine sample in the second set of wine samples is also a break spectrum.
7. Training method according to claim 6, wherein said first spectrum comprises two bands spaced apart by 4000cm each -1 ～3791cm -1 And 2943cm -1 ～1670cm -1 。
8. A training method as claimed in claim 6, wherein said second spectrum comprises three bands spaced apart by 4000cm -1 ～3791cm -1 ，2727cm -1 ～2518cm -1 And 2090cm -1 ～1670cm -1 。
9. The training method as claimed in claim 6, wherein the bands in the first spectrum and the second spectrum are obtained by performing a combined interval partial least squares operation according to the mid-infrared spectrum information of each wine sample in the original wine sample set, the chemically measured content of the total acid of each wine sample in the original wine sample set and the chemically measured content of the total ester of each wine sample in the original wine sample set, and comparing cross validation mean square deviations, wherein the bands of the continuous spectrum in the mid-infrared spectrum information of each wine sample in the original wine sample set are equally divided into N sub-bands, and N is greater than 1;
preferably, N takes the value 17, 55 or 85.
10. A method for measuring the content of total acids and total esters in wine is characterized by comprising the following steps:
acquiring mid-infrared spectrum information of a wine sample to be detected;
inputting the mid-infrared spectrum information of the wine sample to be detected into a wine total acid and total ester content measuring model to obtain the contents of the wine sample total acid and total ester output by the wine total acid and total ester content measuring model,
wherein the model for measuring the content of total acids and total esters is obtained by pre-training according to the training method of any one of claims 1 to 9.
11. An apparatus for training a model for determining the content of total acids and total esters in wine, the apparatus comprising:
the system comprises a first acquisition unit, a second acquisition unit and a control unit, wherein the first acquisition unit is used for acquiring training data, and the training data comprises mid-infrared spectrum information of each wine sample in a first wine sample set, chemical determination content of total acid of each wine sample in the first wine sample set, mid-infrared spectrum information of each wine sample in a second wine sample set and chemical determination content of total ester of each wine sample in the second wine sample set;
and a model training unit, configured to use the mid-infrared spectrum information of each wine sample in the first wine sample set and the mid-infrared spectrum information of each wine sample in the second wine sample set as inputs of a first partial least squares regression model, use the chemically-measured content of the total acids in each wine sample in the first wine sample set and the chemically-measured content of the total esters in each wine sample in the second wine sample set as outputs of the first partial least squares regression model, train the first partial least squares regression model, and obtain a total acid and total ester content model, where the total acid and total ester content model includes a total acid content model for mid-infrared measurement and a total ester content model for infrared measurement, where the first wine sample set and the second wine sample set are both from the same original wine sample set, the first wine sample set is obtained by screening out wine samples with abnormal total acids through the original wine sample set, and the second wine sample set is obtained by screening out wine samples with abnormal total esters through the original wine sample set.
12. An apparatus for determining the total acid and total ester content of wine, comprising:
the second acquisition unit is used for acquiring mid-infrared spectrum information of the wine sample to be detected;
a wine sample total acid and total ester determination unit, configured to input mid-infrared spectrum information of the wine sample to be determined into a wine total acid and total ester content determination model, and obtain contents of wine sample total acid and total ester output by the wine total acid and total ester content determination model, where the wine total acid and total ester content determination model is obtained by pre-training the apparatus according to claim 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210769887.5A CN115236024A (en) | 2022-06-30 | 2022-06-30 | Training method, determination method and device for model for determining content of total acids and total esters in wine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210769887.5A CN115236024A (en) | 2022-06-30 | 2022-06-30 | Training method, determination method and device for model for determining content of total acids and total esters in wine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115236024A true CN115236024A (en) | 2022-10-25 |
Family
ID=83671311
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210769887.5A Pending CN115236024A (en) | 2022-06-30 | 2022-06-30 | Training method, determination method and device for model for determining content of total acids and total esters in wine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115236024A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116735811A (en) * | 2023-08-14 | 2023-09-12 | 山东百脉泉酒业股份有限公司 | Method and system for measuring total acid and total ester content of wine |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111426636A (en) * | 2020-04-09 | 2020-07-17 | 米诺杰科技(江苏)有限公司 | Instrument for quickly analyzing chemical component content of wine |
-
2022
- 2022-06-30 CN CN202210769887.5A patent/CN115236024A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111426636A (en) * | 2020-04-09 | 2020-07-17 | 米诺杰科技(江苏)有限公司 | Instrument for quickly analyzing chemical component content of wine |
Non-Patent Citations (4)
| Title |
|---|
| 周军 等: "中红外光谱技术在浓香型白酒原酒关键指标分析中的应用", 《光谱学与光谱分析》, vol. 42, no. 3, 31 March 2022 (2022-03-31), pages 764 - 768 * |
| 王伟: "智能检测技术", 31 December 2021, 机械工业出版社, pages: 45 - 49 * |
| 谷宇: "人工智能基础", 30 November 2021, 机械工业出版社, pages: 33 - 34 * |
| 邹胜琼 等: "中红外模型快速检测白酒基础酒中总酯的含量", 《酿酒》, vol. 48, no. 3, 31 May 2021 (2021-05-31), pages 74 - 77 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116735811A (en) * | 2023-08-14 | 2023-09-12 | 山东百脉泉酒业股份有限公司 | Method and system for measuring total acid and total ester content of wine |
| CN116735811B (en) * | 2023-08-14 | 2023-10-10 | 山东百脉泉酒业股份有限公司 | Method and system for measuring total acid and total ester content of wine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107703097B (en) | Method for constructing model for rapidly predicting crude oil property by using near-infrared spectrometer | |
| CN107817223A (en) | The construction method of quick nondestructive real-time estimate oil property model and its application | |
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN101504363A (en) | Edible fatty acid value detection method based on near-infrared spectrum analysis | |
| CN104749132A (en) | Method for measuring content of azodicarbonamide in flour | |
| CN101216419A (en) | Method for quickly detecting yellow wine quality index | |
| CN106596456A (en) | Changeable moving window based selection method of near infrared spectral region of solution | |
| CN102876816A (en) | Fermentation process statue monitoring and controlling method based on multi-sensor information fusion | |
| CN105548062A (en) | A multi-index rapid nondestructive synchronous detection method for fresh beef | |
| CN104596979A (en) | Method for measuring cellulose of reconstituted tobacco by virtue of near infrared reflectance spectroscopy technique | |
| CN115236024A (en) | Training method, determination method and device for model for determining content of total acids and total esters in wine | |
| CN107991265A (en) | A kind of wheat flour Rubus biflorus Buch quick determination method based on information fusion | |
| CN112179871A (en) | Method for nondestructive detection of caprolactam content in sauce food | |
| CN105738311A (en) | Apple sweetness non-damage quick detection method based on near-infrared spectrum technology | |
| CN102128805A (en) | Method and device for near infrared spectrum wavelength selection and quick quantitative analysis of fruit | |
| CN110987858A (en) | Method for rapidly detecting oil product by using neural network data model | |
| CN104596976A (en) | Method for determining protein of paper-making reconstituted tobacco through ear infrared reflectance spectroscopy technique | |
| CN105954228A (en) | Method for measuring content of sodium metal in oil sand based on near infrared spectrum | |
| US7420170B2 (en) | Fourier transform infrared (FTIR) chemometric method to determine cetane number of diesel fuels containing fatty acid alkyl ester additives | |
| CN111896497A (en) | Spectral data correction method based on predicted value | |
| CN117309776A (en) | Raw chicken ingredient detection method based on hyperspectral and single-tag regression | |
| CN105891148A (en) | Method for measuring VFA concentration in anaerobic wastewater treatment process by means of near infrared spectrum | |
| CN107688007A (en) | A kind of detection method of the heat conduction oil quality based on near-infrared spectral analysis technology | |
| CN104181125A (en) | Method for rapidly determining Kol-bach value of beer malt | |
| CN104374740A (en) | Detection method of bee product |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 564622 Xijiu Town, Xishui County, Zunyi City, Guizhou Province Applicant after: Guizhou Xijiu Co.,Ltd. Address before: 564622 Xijiu Town, Xishui County, Zunyi City, Guizhou Province Applicant before: GUIZHOU MOUTAI WINERY (Group) XIJIU Co.,Ltd. |