CN111141704B - Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine - Google Patents
Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine Download PDFInfo
- Publication number
- CN111141704B CN111141704B CN202010106463.1A CN202010106463A CN111141704B CN 111141704 B CN111141704 B CN 111141704B CN 202010106463 A CN202010106463 A CN 202010106463A CN 111141704 B CN111141704 B CN 111141704B
- Authority
- CN
- China
- Prior art keywords
- near infrared
- infrared spectrum
- data
- temperature
- regression
- 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.)
- Active
Links
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 72
- 238000000855 fermentation Methods 0.000 title claims abstract description 71
- 230000004151 fermentation Effects 0.000 title claims abstract description 71
- 238000012544 monitoring process Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 49
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims abstract description 45
- 238000000605 extraction Methods 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 238000012360 testing method Methods 0.000 claims description 37
- 235000019674 grape juice Nutrition 0.000 claims description 33
- 230000003595 spectral effect Effects 0.000 claims description 24
- 239000013598 vector Substances 0.000 claims description 21
- 238000005259 measurement Methods 0.000 claims description 17
- 239000000523 sample Substances 0.000 claims description 17
- 238000002835 absorbance Methods 0.000 claims description 13
- 238000004497 NIR spectroscopy Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000011049 filling Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 239000013307 optical fiber Substances 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 239000007791 liquid phase Substances 0.000 claims description 3
- 229910052700 potassium Inorganic materials 0.000 claims description 2
- 238000005429 filling process Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 12
- 238000007405 data analysis Methods 0.000 abstract description 3
- 238000012994 industrial processing Methods 0.000 abstract description 2
- 238000009776 industrial production Methods 0.000 abstract description 2
- 238000010801 machine learning Methods 0.000 abstract description 2
- 239000002994 raw material Substances 0.000 description 9
- 238000005070 sampling Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- RWPGFSMJFRPDDP-UHFFFAOYSA-L potassium metabisulfite Chemical compound [K+].[K+].[O-]S(=O)S([O-])(=O)=O RWPGFSMJFRPDDP-UHFFFAOYSA-L 0.000 description 6
- 229940043349 potassium metabisulfite Drugs 0.000 description 6
- 235000010263 potassium metabisulphite Nutrition 0.000 description 6
- 238000007789 sealing Methods 0.000 description 6
- 239000000243 solution Substances 0.000 description 6
- 240000004808 Saccharomyces cerevisiae Species 0.000 description 5
- 235000014680 Saccharomyces cerevisiae Nutrition 0.000 description 5
- 241000219095 Vitis Species 0.000 description 5
- 235000009754 Vitis X bourquina Nutrition 0.000 description 5
- 235000012333 Vitis X labruscana Nutrition 0.000 description 5
- 235000014787 Vitis vinifera Nutrition 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000012417 linear regression Methods 0.000 description 5
- 230000001954 sterilising effect Effects 0.000 description 5
- 238000003756 stirring Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000000047 product Substances 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- QTBSBXVTEAMEQO-UHFFFAOYSA-N Acetic acid Chemical compound CC(O)=O QTBSBXVTEAMEQO-UHFFFAOYSA-N 0.000 description 3
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 3
- 241000219094 Vitaceae Species 0.000 description 3
- 238000013124 brewing process Methods 0.000 description 3
- 238000011478 gradient descent method Methods 0.000 description 3
- 235000021021 grapes Nutrition 0.000 description 3
- 230000005484 gravity Effects 0.000 description 3
- 238000004659 sterilization and disinfection Methods 0.000 description 3
- 241000894006 Bacteria Species 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000003306 harvesting Methods 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- CHHHXKFHOYLYRE-UHFFFAOYSA-M 2,4-Hexadienoic acid, potassium salt (1:1), (2E,4E)- Chemical compound [K+].CC=CC=CC([O-])=O CHHHXKFHOYLYRE-UHFFFAOYSA-M 0.000 description 1
- 229920001661 Chitosan Polymers 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 230000009036 growth inhibition Effects 0.000 description 1
- 230000003993 interaction Effects 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
- 230000003204 osmotic effect Effects 0.000 description 1
- 235000010241 potassium sorbate Nutrition 0.000 description 1
- 239000004302 potassium sorbate Substances 0.000 description 1
- 229940069338 potassium sorbate Drugs 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 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
- G01N21/359—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
- 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)
- Distillation Of Fermentation Liquor, Processing Of Alcohols, Vinegar And Beer (AREA)
Abstract
The invention discloses a near infrared spectrum-based real-time monitoring method for an ice wine temperature-control slow fermentation process, and belongs to the field of process industrial production and processing. The method comprises the following steps: the method comprises the steps of obtaining a near infrared spectrum data set, preprocessing near infrared spectrum data, establishing a feature extraction model, establishing a regression monitoring model and monitoring the temperature control slow fermentation process of the ice wine in real time based on the near infrared spectrum. The invention organically combines the microscopic monitoring technology and the data analysis algorithm by utilizing the machine learning characteristic extraction and regression monitoring model and integrates the microscopic monitoring technology and the data analysis algorithm into the process of brewing the ice wine, so that the microscopic trend of the monitoring on the sugar degree and the alcohol potential in the temperature-controlled slow fermentation link can be more directly reflected.
Description
Technical Field
The invention relates to a near infrared spectrum-based real-time monitoring method for an ice wine temperature-control slow fermentation process, and belongs to the field of process industrial production and processing.
Background
In 1794, ice wine, also called ice wine, originally appeared in frankeny, germany, the definition of ice wine in the technical specification of chinese grape wine was: harvesting grape, freezing grape at temperature below-7 deg.C, harvesting, squeezing to obtain grape juice, and brewing with the grape juice.
In general, the ice wine brewing process comprises: picking, sorting, squeezing frozen grapes to obtain concentrated grape juice, adding into a container, sterilizing the concentrated grape juice, slowly fermenting at a controlled temperature, performing secondary filtration sterilization, clarifying a product, and filling into wine; factors influencing the quality of the ice wine are various, and typical factors comprise picking temperature, grape variety, fermentation temperature and the like. In addition, in the temperature-controlled slow fermentation process, the sugar degree and alcohol potential of the ice grape juice which are continuously changed are important quality indexes for finally evaluating the ice wine; in the temperature-controlled slow fermentation process, if the sugar concentration is too low, growth inhibition or yeast cracking can be caused, even fermentation is stopped or retarded, even sub-lethal sugar concentration can stimulate the high osmotic pressure stress reaction of yeast, so that the production amount of various byproducts such as glycerol and acetic acid is increased, and the product yield is reduced; whether the alcohol potential can reach and maintain the desired concentration as desired is critical to the evaluation of the ice wine itself and the susceptibility to contamination by spoilage bacteria. Therefore, the real-time monitoring of the sugar content and the alcohol potential of the ice grape juice in the process of brewing the ice wine is very important in the process of temperature-controlled slow fermentation.
In the prior art, a hydrometer is generally used for measuring the specific gravity, the sugar degree and the alcohol potential of an ice grape juice raw material in the temperature-controlled slow fermentation process, and the hydrometer is used for converting an empirical table obtained by priori knowledge based on the measured specific gravity to obtain the percentage content of special components in liquid; however, for the temperature control slow fermentation links of different ice grape juice raw materials, the hydrometer based on the experience table is insufficient in the aspects of measurement precision and reliability; the sugar degree and alcohol potential of the existing ice grape juice temperature-controlled slow fermentation link are only measured macroscopically, and since the macroscopical change is caused by the continuous accumulation of microscopic change, compared with other macroscopical monitoring technologies such as a hydrometer and the like, the existing ice wine brewing fermentation process mostly uses, so that the microscopic change in the ice grape juice in the temperature-controlled slow fermentation link is difficult to be reflected quickly and accurately; the temperature control slow fermentation link cannot be shown on the microscopic molecular level, so that errors occur when the fermentation degree or quality of the ice grape juice is judged, and the problems of more serious economic and energy loss caused by estimated loss, long time for taking material supplementing measures, failure in stopping loss in time and the like are caused.
The near infrared spectrum is a rapid, high-efficiency, nondestructive and pollution-free measurement technology and is widely applied to various occasions such as pharmacy, biology, industry and the like. The near infrared spectrum technology integrates the near infrared spectrum measurement technology, chemometrics and computer science into a whole to realize quantitative and qualitative analysis of samples. Compared with other spectrum technologies, the near infrared spectrum has the characteristic of weak absorption, so that the sample can be directly analyzed without pretreatment such as dilution and the like. The method is based on the absorption spectrum generated by the interaction of the molecules (mainly C-H, O-H and N-H covalent bonds) of a measured sample and transmitted light or reflected light, and has the advantages of simplicity, quickness, no need of sample pretreatment and the like. The near infrared spectroscopy measurement step generally includes selecting and collecting a standard sample set for establishing a spectral correction model, with the remaining standard samples as a validation set for evaluating the extrapolation capability of the model; preprocessing the spectral data and establishing a model; optimizing and checking the performance of the correction model; and (6) carrying out sample measurement.
However, the near infrared spectrum data consisting of the wave number-absorbance element vector acquired by using the infrared detector has the characteristics of high dimension, multiple linear correlation, information redundancy and the like, so that the problems that the complexity of a monitoring model is increased, the monitoring model is difficult to accurately establish and the like are caused during modeling.
Disclosure of Invention
The method aims to solve the problems that in the prior art, the processing difficulty of near infrared spectrum data is high, the complexity of a monitoring model is increased during modeling, the monitoring model is difficult to establish correctly and the like; the invention provides a near infrared spectrum-based real-time monitoring method for an ice wine temperature-control slow fermentation process, which comprises the following steps:
step 1: installing a transmission type liquid phase optical fiber probe of a near infrared spectrometer on a hole position of a cover of the ice grape juice temperature-controlled slow fermentation vessel, and collecting spectral data in the ice grape juice temperature-controlled slow fermentation process; the spectral data is near infrared spectral data consisting of vectors of wave number-absorbance elements in the temperature-controlled slow fermentation process of the ice wine, and the near infrared spectral data acquired at the time t is set as Xt，Xt＝{x1,x2,x3,…,xM}; putting a specific gravity meter into a temperature-controlled slow fermentation vessel according to a certain proportionReading time interval, converting and recording sugar degree T at T moment based on empirical tabletAnd alcohol potential AtN times in total; will be read by a hydrometer and record the resulting sugar degree TtAnd alcohol potency AtAs regression label vector YtThe regression label vector is prior index information required in the training process of the regression model; will YtNear infrared spectral data X corresponding to timetMatching is performed based on the time t correspondence relationship, thereby forming a time sequence consisting of n bars and { X }t,Yt}＝{x1,x2…,xM,Tt,AtD, i.e.:
step 2: rejecting abnormal or damaged data in the near infrared spectrum data set, filling missing data and correcting error data; performing vector normalization and first derivative smoothing filtering on a near infrared spectrum data set of the ice grape juice in the temperature-controlled slow fermentation link;
and step 3: carrying out feature extraction and establishing a feature extraction model;
and 4, step 4: establishing a regression monitoring model, wherein the regression monitoring model comprises data set division and regression monitoring model class selection, inputting a test set into the regression monitoring model, and taking a mean square error and goodness of fit as performance evaluation indexes;
and 5: and carrying out real-time monitoring on the temperature-controlled slow fermentation process of the ice wine according to the trained model.
In one embodiment of the present invention, the process of rejecting abnormal or damaged data in the near infrared spectrum data set comprises: calculating tiTemporal near infrared spectral data
Absolute value of residual error of the r-th element of (1)
If it is
The measurement is bad, where σ is the standard deviation σ of the normal distribution (σ, μ) of the variables to which the measurement belongs.
In an embodiment of the present invention, the process of filling missing data is: let tjTemporal near infrared spectroscopy
Missing the mth dimension element
Based on a nearest neighbor method, in a sample space formed by a near infrared spectrum data set D, based on the Euclidean distance D:
wherein
Means a certain tiTime of day near infrared spectral vector, ti≠tj，
Is tiTemporal near infrared spectroscopy
The (n) th element of (a),
is tjTime of day and presence of missing data
Near infrared spectrum of
The nth element of (1); according to the above formula, determining
Euclidean distance of
Is recorded as the nearest neighbor point corresponding to the minimum value of
And use
Middle mth dimension element
As
Missing data
And filling all missing data in the near infrared spectrum data set by analogy.
In one embodiment of the present invention, the error data is corrected by: based on the artificial prior knowledge, erroneous data caused by measurement errors, misoperations, and the like are corrected.
In an embodiment of the present invention, the vector normalization process of the near infrared spectrum data set of the ice grape juice in the temperature controlled slow fermentation step is as follows:
wherein M is 1,2, …, M,
is the mean value of the m-dimension near infrared spectrum wave number-absorbance variable, | xmAnd | is the square root of the sum of squares of the m-dimension near infrared spectrum wave number-absorbance variable of the n pieces of near infrared spectrum data.
In one embodiment of the present invention, the first derivative smoothing filter is:
wherein w is a coefficient determining the width of the filter window,
finger tiTemporal near infrared spectroscopy
B refers to the moving sequence number of the filtering window.
In one embodiment of the present invention, the feature extraction algorithm is partial least squares analysis, the feature variable dimension is set to 10, the maximum covariance between the maximum near infrared spectrum variable and the regression label value is the target, and the first main component of X is u1The corresponding weight matrix is p1The first major component of Y is v1The corresponding weight matrix is q1Is provided with
u1＝Xp1,v1＝Yq1
Var(u1)→max，Var(v1)→max
Corr(u1,v1)→max
The optimization goals are as follows:
after the optimization goal is reached, a second main component u is searched from X2And the second major component of Y is v2And optimizing according to the optimizing method, and repeating the steps until the number of the main components reaches the set requirement.
In an embodiment of the present invention, the data dividing process is: the method comprises the steps of taking an original data set D subjected to near infrared spectrum data preprocessing as a sample input, and dividing the original data set D into a training set Dp according to a certain proportionTrain＝{XpTrain,YpTrainAnd test set DpTest＝{XpTest,YpTest}。
In one embodiment of the present invention, the regression monitoring model selects multiple linear regression as the regression monitoring algorithm based on the training set DpTrain＝{XpTrain,YpTrainAnd (4) setting a multivariate linear regression equation with a parameter domain of theta as follows:
hθ(X)＝θTX＝θ0x0+θ1x1+L+θnxn,n＝10
the loss function is established as:
wherein h isθA quantitative model of the mapping relation between the low-dimensional characteristic variable and the sugar degree of the ice grape juice and the alcohol potential in the temperature-controlled slow fermentation link;
and (3) optimizing parameters by adopting a gradient descent method until the loss function converges to a smaller value:
where α is the learning rate.
In an embodiment of the present invention, the inputting the test set into the regression monitoring model, with the mean square error and the goodness of fit as performance evaluation indicators, comprises:
test set DpTest＝{XpTest,YpTestInputting a regression monitoring model, and taking a mean square error and a goodness of fit as performance evaluation indexes:
wherein n istestFor the number of samples in the test set, MSETIs the mean square error corresponding to the sugar degree,
representing near infrared spectral data
Obtaining the sugar degree estimated value under the characteristic extraction and regression model mapping f,
namely the corresponding real sugar degree label value, MSEAIs the mean square error corresponding to the potential of alcohol,
representing near infrared spectral data
Obtaining an alcohol potential estimation value under the conditions of feature extraction and regression model mapping g,
that is, the corresponding real alcohol potential label value is obtained, the mean square error MSE is the average of the distances of all data from the average, and the estimation obtained on the test set by the regression detection model is reflectedThe smaller the value of the degree of dispersion between the evaluation value and the real label value, the better the value;
wherein the content of the first and second substances,
is the average of the brix label values, RT 2Is the goodness of fit corresponding to the brix, goodness of fit RA 2Refers to the fitting degree of the regression monitoring model to the sugar degree and the alcohol potential label value,
the average of the alcohol potential label values.
Has the advantages that:
the invention provides a real-time monitoring method for the temperature-controlled slow fermentation process of ice wine based on near infrared spectrum by utilizing near infrared spectrum data acquired in the temperature-controlled slow fermentation link of the ice wine brewing process and sugar degree and alcohol potential data acquired by a hydrometer, wherein a machine learning characteristic extraction and regression monitoring model is utilized, a microscopic monitoring technology and a data analysis algorithm are organically combined and integrated into the ice wine brewing process, and the traditional hydrometer is converted into microscopic on-line monitoring based on near infrared spectrum data modeling from macroscopic fixed measurement based on an artificial experience table; according to the method, near infrared spectrum data consisting of vectors of wave number-absorbance elements in the temperature-controlled slow fermentation process of ice wine are collected, densitometer reading is carried out, the obtained sugar degree and alcohol potential are recorded to be used as prior index information required in the regression model training process, partial least square analysis is adopted for feature extraction, multiple linear regression is selected to be used as a regression monitoring algorithm, a gradient descent method is adopted to optimize parameters, and mean square error and goodness of fit are used as performance evaluation indexes; in the fermentation process on-line monitoring, the sugar degree and alcohol potential of the ice grape juice raw material in the temperature-controlled slow fermentation process can be monitored in real time and recorded on the basis of the previously established feature extraction model and regression monitoring model; the monitoring of the sugar degree and the alcohol potential in the temperature-controlled slow fermentation link can more directly reflect the microscopic trend, so that people can find problems more timely and estimate and remedy loss.
Drawings
FIG. 1 is a flow chart of the implementation procedure of example 1.
FIG. 2 is a graph showing NIR data of the temperature controlled slow fermentation process of example 1.
FIG. 3 is a schematic diagram of the real-time monitoring effect of sugar content during the temperature-controlled slow fermentation process of ice wine based on the near infrared spectrum detection technology in example 1.
FIG. 4 is a schematic diagram of the real-time monitoring effect of alcohol potential in the temperature-controlled slow fermentation process of ice wine based on the near infrared spectrum detection technology in example 1.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Example 1
The embodiment provides a method for monitoring the temperature-controlled slow fermentation process of ice wine in real time based on near infrared spectrum, as shown in fig. 1, the method comprises the following steps:
step 1: installing a transmission type liquid phase optical fiber probe of a near infrared spectrometer on a central hole position of a cover of a fermentation vessel used for the temperature-controlled slow fermentation of the ice grape juice; opening upper computer software OPUS (infrared spectrum software) connected with the instrument, and starting to configure sampling parameters in an OPUS environment; the resolution in the measurement configuration is set to 4 in turn in the OPUS softwarecm-1, the number of scans was set to 16, and the sampling range was set to 4000cm-1-12500cm-1Determining the data type of the result spectrum as absorbance; by measuring a single-channel background spectrum, the background is automatically filtered in the subsequent sampling measurement;
sampling and recording near infrared spectrum data consisting of wave number-absorbance element vectors in the temperature-controlled slow fermentation process of the ice wine according to the sampling parameter setting of the near infrared spectrometer, wherein each near infrared spectrum data is a vector containing 2203 wave number-absorbance elements, and the near infrared spectrum data X consisting of the wave number-absorbance element vectors collected at the moment t is settThen Xt＝{x1,x2,x3,…,x2203}；
Putting a hydrometer into a temperature-controlled slow fermentation vessel, reading at intervals of every 2h, converting and recording the sugar degree T at the T moment based on an empirical tabletAnd alcohol potential AtN times in total;
will be read by a hydrometer and record the resulting sugar degree TtAnd alcohol potency AtAs regression label vector YtThe prior index information required in the training process of the regression model; then, Y is puttNear infrared spectral data X corresponding to timetMatching is performed based on the time t correspondence relationship, and the same { X is formed by N-600M-2203-dimensional elementst,Yt}＝{x1,x2…,x2203,Tt,AtThe data set D of the near infrared spectrum of the temperature-controlled slow fermentation process shown in the attached diagram of 2 in example 1 is shown as the data schematic diagram of the near infrared spectrum of the temperature-controlled slow fermentation process, namely:
step 2: rejecting abnormal or damaged data in the near infrared spectrum data set; calculating tiTime instant near infrared spectrum data (i ═ 1,2, K, n)
R element of (2)
Absolute value of residual error of
If it is
The measurement is bad, where σ is the standard deviation σ of the normal distribution (σ, μ) of the variables to which the measurement belongs;
filling missing data; let tj(j is more than or equal to 1 and less than or equal to n) time point near infrared spectrum
Deletion of the mth (M is more than or equal to 1 and less than or equal to M) dimension element
Based on a nearest neighbor method, in a sample space formed by a near infrared spectrum data set D, based on the Euclidean distance D:
wherein
Means a certain ti(ti≠tj) The temporal near-infrared spectral vector of the light,
is tiTemporal near infrared spectroscopy
The (n) th element of (a),
is tjTime of day and presence of missing data
Near infrared spectrum of
The nth element of (1). According to the above formula, determining
Euclidean distance of
Is recorded as the nearest neighbor point corresponding to the minimum value of
And use
Middle mth dimension element
As
Missing data
Filling all missing data in the near infrared spectrum data set by analogy;
correcting error data, namely correcting error data caused by measurement errors, misoperation and the like according to artificial priori knowledge;
vector normalization is carried out on the near infrared spectrum data set of the ice grape juice in the temperature-controlled slow fermentation link,
wherein
Is tiTemporal near infrared spectroscopy
The m-th-dimensional element of (2),
is the mean value of the m-dimension near infrared spectrum wave number-absorbance variable, | xm| is the square root of the sum of squares of the m-dimension near infrared spectrum wave number-absorbance variable of the n pieces of near infrared spectrum data;
the first derivative smoothing filter is:
wherein w is a coefficient determining the width of the filter window,
finger tiTemporal near infrared spectroscopy
B refers to the moving sequence number of the filtering window. The optional preprocessing methods include elimination of constants, elimination of a straight line, maximum-minimum normalization, multivariate scattering correction, second derivative smoothing filtering, etc., according to the data characteristics of the near infrared spectrum.
And step 3: performing feature extraction, establishing a feature extraction model, wherein the feature extraction algorithm is partial least square analysis and setting feature variablesThe quantity dimension is 10, the maximum covariance between the maximum near infrared spectrum variable and the regression label value is the target, and the first main component of X is u1The corresponding weight matrix is p1The first major component of Y is v1The corresponding weight matrix is q1Is provided with
u1＝Xp1,v1＝Yq1
Var(u1)→max，Var(v1)→max
Corr(u1,v1)→max
The optimization goals are as follows:
after the optimization goal is reached, a second main component u is searched from X2And the second major component of Y is v2And optimizing according to the optimizing method, and repeating the steps until the number of the main components reaches the set requirement.
Reducing the number of near infrared spectrum variables including original spectrum variables 2203 to 10 dimensions by a partial least square algorithm, and according to the relationships among the near infrared spectrum variables and the relationships among the variables, the sugar degree and the alcohol potential in the temperature-controlled slow fermentation process of the ice grape juice, the optional feature extraction algorithm can also be principal component analysis, linear discriminant analysis, a deep learning algorithm based on L1 and L2 regular extraction methods, correlation coefficient extraction methods, typical automatic encoders, deep belief networks and the like;
and 4, step 4: establishing a regression monitoring model; the method comprises the following steps:
step 41: dividing a data set; a feature data set Dp composed of 10-dimensional feature vectors is used as sample input and is divided into a training set Dp according to the proportion of 3:2Train＝{XpTrain,YpTrainAnd test set DpTest＝{XpTest,YpTest}；
Step 42: selecting a regression monitoring model class; selecting regression monitoring model class, selecting multiple linear regression as regression monitoring algorithm, training-basedExercise and Collection DpTrain＝{XpTrain,YpTrainAnd (4) setting a multivariate linear regression equation with a parameter domain of theta as follows:
hθ(X)＝θTX＝θ0x0+θ1x1+L+θnxn,n＝10
the loss function is established as:
and (3) optimizing parameters by adopting a gradient descent method until the loss function converges to a smaller value:
where α is the learning rate. Applying the regression model establishment process to a near infrared spectrum characteristic variable training set XpTrainTo sugar degree and alcohol potential YpTrainIn the regression modeling, the characteristic variable Xp extracted from partial least squares characteristics can be obtainedtTo sugar degree TtAnd alcohol potency AtF and g.
hθThe quantitative model of the mapping relation between the sugar degree of the ice grape juice and the alcohol potential in the low-dimensional characteristic variable and temperature-controlled slow fermentation link is selected, and the regression monitoring model can be selected and used, such as polynomial regression, ridge regression, Lasso regression, deep neural network and the like according to the correlation (linear or nonlinear) between the near infrared spectrum characteristic variable and the sugar degree and the alcohol potential in the temperature-controlled slow fermentation process of the ice grape juice;
step 43: test set DpTest＝{XpTest,YpTestInputting a regression monitoring model, and taking a mean square error and a goodness of fit as performance evaluation indexes:
wherein n istestFor the number of samples in the test set, MSETIs the mean square error corresponding to the sugar degree,
representing near infrared spectral data
Obtaining the sugar degree estimated value under the characteristic extraction and regression model mapping f,
namely the corresponding real sugar degree label value, MSEAIs the mean square error corresponding to the potential of alcohol,
representing near infrared spectral data
Obtaining an alcohol potential estimation value under the conditions of feature extraction and regression model mapping g,
namely the corresponding real alcohol potential label value. The mean square error MSE is the average of the distances of each data from the average, and can reflect the degree of dispersion between the estimated values obtained by the regression detection model on the test set and the true tag values, and the smaller the value, the better the value.
Wherein the content of the first and second substances,
is the average of the brix label values, RT 2Is the goodness of fit corresponding to the brix, goodness of fit RA 2Refers to the fitting degree of the regression monitoring model to the sugar degree and the alcohol potential label value,
is the average value of the alcohol potential label value, RA 2Maximum value of 1, RA 2The closer the value of (1) is, the better the fitting degree of the regression straight line to the observed value is; otherwise, RA 2The smaller the value of (A) is, the worse the fitting degree of the regression straight line to the observed value is;
step 44: after the model training is finished, determining and storing regression monitoring model description and parameters;
and 5: monitoring the temperature-controlled slow fermentation process of the ice wine in real time based on the near infrared spectrum; the method comprises the following steps:
step 51: picking, sorting and squeezing frozen grapes, comprising: obtaining a raw material of the ice grape juice, namely picking, sorting and squeezing the ice grapes;
step 52: container, concentrated grape juice sterilization comprising:
step 521: cleaning the inner wall of a fermentation vessel (barrel) with clear water;
step 522: sterilizing in a container, adding a potassium metabisulfite solution, dissolving in water according to the concentration of 5g/L, and shaking a container to make the container fully contact with the inner wall;
step 523: putting the raw material of the ice grape juice into a fermentation vessel, wherein the distance between the liquid surface and an opening is kept about 10cm, so that a transmission type optical fiber probe configured by a near-infrared spectrometer can be conveniently inserted;
step 524: adding a potassium metabisulfite solution according to the amount of the ice grape juice in each vessel to enable the concentration of the potassium metabisulfite solution to reach 50ppm, and uniformly stirring to kill the mixed bacteria in the ice grape juice raw material;
step 525: standing the sealing cover without sealing, and waiting for the potassium metabisulfite solution to completely exert a sterilization effect;
step 53: controlling the temperature and slowly fermenting; the method comprises the following steps:
step 531: adding wine yeast according to the proportion of 10g/25kg of the raw material of the ice grape juice, and properly stirring;
step 532: sealing and standing without sealing, controlling the fermentation temperature to be higher than 22 ℃, and making the temperature of the fermentation vessel slightly higher than the ambient temperature by fermentation heat release;
step 533: inserting a transmission type optical fiber probe configured in a near-infrared spectrometer from a central hole position of a top cover of a vessel to ensure that the tip of the transmission type optical fiber probe is immersed into the ice grape juice raw material to a sufficient depth;
step 534: setting sampling parameters of a near-infrared spectrometer according to the step 1, setting an expected target sugar degree to be 150g/L and setting the alcohol potential to be 10.5%;
step 535: the near infrared spectrometer sets sampling parameters according to the step 1, starts to collect near infrared spectrum data consisting of wave number-absorbance element vectors, and based on the previously established feature extraction model and regression monitoring model, the sugar degree and alcohol potential of the raw material of the ice grape juice in the temperature-controlled slow fermentation process are monitored and recorded in real time, the schematic diagram of the real-time monitoring effect of the sugar degree of the temperature-controlled slow fermentation process of the ice wine based on the near infrared spectrum detection technology shown in the attached figure 3 of the embodiment 1 and the schematic diagram of the real-time monitoring effect of the alcohol potential of the temperature-controlled slow fermentation process of the ice wine based on the near infrared spectrum detection technology shown in the attached figure 4 can be obtained, setting sampling parameters according to the step 1, realizing real-time monitoring of sugar degree and alcohol potential once in about 4 minutes and drawing a trace point shown in a schematic diagram, the progress of the temperature-controlled slow fermentation process of the ice wine is monitored in real time, and the change trends of the sugar degree and the alcohol potential are observed;
step 536: continuously observing the sugar degree and alcohol potential in the temperature-controlled slow fermentation process until the sugar degree and alcohol potential reach the expectation;
step 54: removing the grape yeast, including: stopping fermentation, adding a potassium metabisulfite solution to make the concentration of the potassium metabisulfite solution reach 150ppm, and fully stirring to remove wine yeast;
step 55: clarifying the product, comprising: adding chitosan according to the mass ratio of 1:100 and fully stirring; standing the sealing cover, and completely sealing;
step 56: filtering and filling into wine, comprising: preserving food, adding potassium sorbate to make its concentration reach 100ppm, and stirring; filtering and filling to obtain the finished product.
The scope of the present invention is not limited to the above embodiments, and any modifications, equivalent substitutions, improvements, etc. that can be made by those skilled in the art within the spirit and principle of the inventive concept should be included in the scope of the present invention.
Claims (6)
1. A real-time monitoring method for an ice wine temperature control slow fermentation process based on near infrared spectrum is characterized by comprising the following steps:
step 1: installing a transmission type liquid phase optical fiber probe of a near infrared spectrometer on a hole position of a cover of the ice grape juice temperature-controlled slow fermentation vessel, and collecting spectral data in the ice grape juice temperature-controlled slow fermentation process; the spectral data is near infrared spectral data consisting of vectors of wave number-absorbance elements in the temperature-controlled slow fermentation process of the ice wine, and the near infrared spectral data acquired at the time t is set as Xt，Xt＝{x1,x2,x3,…,xM}; adding a hydrometer into a temperature-controlled slow fermentation vessel, reading at certain time intervals, converting and recording the sugar degree T at the T moment based on an empirical tabletAnd alcohol potential AtN times in total; will be read by a hydrometer and record the resulting sugar degree TtAnd alcohol potency AtAs regression label vector YtThe regression label vector is prior index information required in the training process of the regression model; will YtNear infrared spectral data X corresponding to timetMatching is performed based on the time t correspondence relationship, thereby forming a time sequence consisting of n bars and { X }t,Yt}＝{x1,x2…,xM,Tt,AtD, i.e.:
step 2: rejecting abnormal or damaged data in the near infrared spectrum data set, filling missing data and correcting error data; performing vector normalization and first derivative smoothing filtering on a near infrared spectrum data set of the ice grape juice in the temperature-controlled slow fermentation link;
and step 3: carrying out feature extraction and establishing a feature extraction model;
and 4, step 4: establishing a regression monitoring model, wherein the regression monitoring model comprises data set division and selection, inputting a test set into the regression monitoring model, and taking a mean square error and goodness of fit as performance evaluation indexes;
the data set dividing process comprises the following steps: the method comprises the steps of taking an original data set D subjected to near infrared spectrum data preprocessing as a sample input, and dividing the original data set D into a training set Dp according to a certain proportionTrain＝{XpTrain,YpTrainAnd test set DpTest＝{XpTest,YpTest}；
And 5: monitoring the temperature-controlled slow fermentation process of the ice wine in real time according to the trained model;
the missing data filling process comprises the following steps: let tjTemporal near infrared spectroscopy
Missing the mth dimension element
J is more than or equal to 1 and less than or equal to n, and M is more than or equal to 1 and less than or equal to M; the Euclidean distance D in a sample space formed by a near infrared spectrum data set D based on a nearest neighbor method is as follows:
wherein the content of the first and second substances,
means a certain tiTime of day near infrared spectral vector, ti≠tj，
Is tiTemporal near infrared spectroscopy
The (n) th element of (a),
is tjTime of day and presence of missing data
Near infrared spectrum of
The nth element of (1); according to the above formula, determining
Euclidean distance of
Is recorded as the nearest neighbor point corresponding to the minimum value of
And use
Middle mth dimension element
As
Missing data
And filling all missing data in the near infrared spectrum data set by analogy.
2. The method for monitoring the temperature-controlled slow fermentation process of the ice wine in real time based on the near infrared spectrum as claimed in claim 1, wherein the process of rejecting abnormal or damaged data in the near infrared spectrum data set comprises the following steps: calculating tiTemporal near infrared spectral data
Absolute value of residual error of the r-th element of (1)
i is 1,2, K, n; if it is
The measurement is bad, where σ is the standard deviation σ of the normal distribution (σ, μ) of the variables to which the measurement belongs.
3. The method for monitoring the temperature-controlled slow fermentation process of ice wine in real time based on the near infrared spectrum as claimed in claim 1, wherein the error correction data is as follows: and correcting error data caused by measurement errors and misoperation according to the artificial priori knowledge.
4. The method for monitoring the temperature-controlled slow fermentation process of ice wine in real time based on the near infrared spectrum as claimed in claim 1, wherein the first derivative smoothing filtering is as follows:
wherein w is a coefficient determining the width of the filter window,
finger tiTemporal near infrared spectroscopy
B refers to the moving sequence number of the filtering window.
5. The method for real-time monitoring of temperature-controlled slow fermentation process of ice wine based on near infrared spectrum as claimed in claim 1, wherein the characteristic extraction method is partial least squares analysis, the dimension of characteristic variable is set to 10 to maximize the maximum covariance between near infrared spectrum variable and regression label value, and the first main component of X is u1The corresponding weight matrix is p1The first major component of Y is v1The corresponding weight matrix is q1Is provided with
u1＝Xp1,v1＝Yq1
Var(u1)→max，Var(v1)→max
Corr(u1,v1)→max
The optimization goals are as follows:
after the optimization goal is reached, a second main component u is searched from X2And the second major component of Y is v2And optimizing according to the optimizing method, and repeating the steps until the number of the main components reaches the set requirement.
6. The method for monitoring the temperature-controlled slow fermentation process of the ice wine in real time based on the near infrared spectrum as claimed in claim 1, wherein the process of inputting the test set into the regression monitoring model and taking the mean square error and the goodness of fit as performance evaluation indexes comprises the following steps:
test set DpTest＝{XpTest,YpTestInputting a regression monitoring model, and taking a mean square error and a goodness of fit as performance evaluation indexes:
wherein n istestFor the number of samples in the test set, MSETIs the mean square error corresponding to the sugar degree,
representing near infrared spectral data
Obtaining the sugar degree estimated value under the characteristic extraction and regression model mapping f,
namely the corresponding real sugar degree label value, MSEAIs the mean square error corresponding to the potential of alcohol,
representing near infrared spectral data
Obtaining an alcohol potential estimation value under the conditions of feature extraction and regression model mapping g,
that is, the corresponding real alcohol potential label value is obtained, the mean square error MSE is the average number of the distances of each data from the average number, and the inverseMapping the discrete degree between the estimated value and the real label value on the test set by the regression detection model;
wherein the content of the first and second substances,
is the average of the brix label values, RT 2Is the goodness of fit corresponding to the brix, goodness of fit RA 2Refers to the fitting degree of the regression monitoring model to the sugar degree and the alcohol potential label value,
the average of the alcohol potential label values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010106463.1A CN111141704B (en) | 2020-02-21 | 2020-02-21 | Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010106463.1A CN111141704B (en) | 2020-02-21 | 2020-02-21 | Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111141704A CN111141704A (en) | 2020-05-12 |
| CN111141704B true CN111141704B (en) | 2021-03-02 |
Family
ID=70527676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010106463.1A Active CN111141704B (en) | 2020-02-21 | 2020-02-21 | Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111141704B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112396344A (en) * | 2020-11-30 | 2021-02-23 | 天津大学 | Chemical process reliability online evaluation method based on product quality |
| CN113052271B (en) * | 2021-05-14 | 2022-02-15 | 江南大学 | Biological fermentation data prediction method based on deep neural network |
| CN114839332A (en) * | 2022-04-17 | 2022-08-02 | 山东富添源生物科技有限公司 | Fermentation state monitoring system for fermenting pear wine |
| CN116735848B (en) * | 2023-07-20 | 2024-05-28 | 上海拜谱生物科技有限公司 | Sample detection device and control method thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102621092A (en) * | 2012-03-17 | 2012-08-01 | 浙江大学 | Method for detecting Danhong injection ethanol precipitation process on line |
| CN103234923A (en) * | 2013-04-09 | 2013-08-07 | 江苏大学 | On-line monitoring method of total sugar content during yellow wine fermentation process |
| CN108827905A (en) * | 2018-04-08 | 2018-11-16 | 江南大学 | A kind of near-infrared model online updating method based on local weighted Lasso |
-
2020
- 2020-02-21 CN CN202010106463.1A patent/CN111141704B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102621092A (en) * | 2012-03-17 | 2012-08-01 | 浙江大学 | Method for detecting Danhong injection ethanol precipitation process on line |
| CN103234923A (en) * | 2013-04-09 | 2013-08-07 | 江苏大学 | On-line monitoring method of total sugar content during yellow wine fermentation process |
| CN108827905A (en) * | 2018-04-08 | 2018-11-16 | 江南大学 | A kind of near-infrared model online updating method based on local weighted Lasso |
Non-Patent Citations (5)
| Title |
|---|
| 冰葡萄酒酿造过程中糖分与乙醇变化的研究;王运照 等;《食品工业科技》;20151231;第36卷(第17期);摘要 * |
| 基于近红外光谱实时监测乙醇发酵过程多组分参数;王旭东 等;《生物工程学报》;20191231;第35卷(第8期);摘要，第1493-1498页，图1-图2 * |
| 干红葡萄酒中苹果酸一乳酸发酵的研究 I苹果酸一乳酸发酵工艺条件的选择;安丹玫;《食品与发酵工业》;19881231(第05期);第2页 * |
| 新疆慕萨莱思葡萄酒的加糖酿制及其酒质检验;黄英 等;《食品工业》;20181231;第39卷(第7期);第90页 * |
| 葡萄酒发酵过程主要参数近红外光谱分析;张树明 等;《农业机械学报》;20130131;第44卷(第1期);第153页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111141704A (en) | 2020-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111141704B (en) | Near infrared spectrum-based real-time monitoring method for temperature-controlled slow fermentation process of ice wine | |
| Saeys et al. | Multivariate calibration of spectroscopic sensors for postharvest quality evaluation: A review | |
| Zude et al. | Non-destructive tests on the prediction of apple fruit flesh firmness and soluble solids content on tree and in shelf life | |
| JP6339244B2 (en) | Method for predicting sugar content and acidity of fruit using multivariate statistical analysis of FT-IR spectrum data | |
| JP5441703B2 (en) | Visible / Near-Infrared Spectroscopy and Grape Brewing Method | |
| León et al. | Detection of apple juice adulteration using near-infrared transflectance spectroscopy | |
| Sánchez et al. | Internal and external quality assessment of mandarins on-tree and at harvest using a portable NIR spectrophotometer | |
| Tsuta et al. | Near-infrared imaging spectroscopy based on sugar absorption band for melons | |
| He et al. | Nondestructive determination of tomato fruit quality characteristics using VIS/NIR spectroscopy technique | |
| Yue et al. | A smart data-driven rapid method to recognize the strawberry maturity | |
| CN101413885A (en) | Near-infrared spectrum method for rapidly quantifying honey quality | |
| Peng et al. | Monitoring of alcohol strength and titratable acidity of apple wine during fermentation using near-infrared spectroscopy | |
| CN107703097B (en) | Method for constructing model for rapidly predicting crude oil property by using near-infrared spectrometer | |
| CN102876816A (en) | Fermentation process statue monitoring and controlling method based on multi-sensor information fusion | |
| CN110320165A (en) | The Vis/NIR lossless detection method of banana soluble solid content | |
| Paz et al. | Evaluating NIR instruments for quantitative and qualitative assessment of intact apple quality | |
| Wang et al. | A feasibility study on monitoring residual sugar and alcohol strength in kiwi wine fermentation using a fiber‐optic FT‐NIR spectrometry and PLS regression | |
| Taira et al. | Direct sugar content analysis for whole stalk sugarcane using a portable near infrared instrument | |
| CN110144280A (en) | A kind of white liquor method and its online simulation based near infrared spectrum on-line checking plucks wine device | |
| CN109946246B (en) | Detection method and device for soluble solid of apple | |
| Wang et al. | Development of near‐infrared online grading device for long jujube | |
| Bai et al. | Model robustness in estimation of blueberry SSC using NIRS | |
| Yang | Nondestructive prediction of optimal harvest time of cherry tomatoes using VIS-NIR spectroscopy and PLSR calibration | |
| CN111157484A (en) | Near infrared spectrum model transfer method for fruit sugar degree detection equipment | |
| CN114112978A (en) | Wavelet function calibration modeling method for detecting concentration of solution in crystallization process by using in-situ infrared spectroscopy |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |