CN111383722A

CN111383722A - Data-driven model analysis method and device for solving wine raw material index range

Info

Publication number: CN111383722A
Application number: CN202010118187.0A
Authority: CN
Inventors: 尚雪纯; 杨华; 周康; 刘朔; 刘江蓉; 高婧; 周坚; 镇依婷
Original assignee: Wuhan Polytechnic University
Current assignee: Wuhan Polytechnic University
Priority date: 2020-02-25
Filing date: 2020-02-25
Publication date: 2020-07-07
Anticipated expiration: 2040-02-25
Also published as: CN111383722B

Abstract

The invention relates to the technical field of wine processing, and discloses a data-driven model analysis method and a data-driven model analysis device for solving a wine raw material index range, wherein the method comprises the steps of obtaining a sample database constructed according to wine grape raw material indexes and wine product indexes; the method comprises the steps of extracting a wine grape raw material sample set and a wine product sample set from the raw material sample set, preprocessing data in a sample database based on the wine grape raw material sample set and the wine product sample set, predicting a wine product quality index based on processed target data, constructing a wine grape raw material quality standard optimization model according to calculation data and a preset optimization target in the prediction process, and determining the wine grape raw material quality index range by constructing the wine grape raw material quality standard optimization model.

Description

Data-driven model analysis method and device for solving wine raw material index range

Technical Field

The invention relates to the technical field of wine processing, in particular to a data-driven model analysis method and device for solving a wine raw material index range.

Background

In the field of wine processing, for brewing of wine, due to the fact that various wine products have different physicochemical properties and different requirements for the quality of wine-making grape raw materials, the wine products meeting national standards, provincial standards, execution standards and enterprise standards need to be manufactured, and the optimal quality range of the wine-making grape raw materials needs to be predicted according to the quality of the wine products, so that a basis can be provided for enterprises to collect and handle the raw materials, the proper wine-making grape raw materials are collected, the qualified wine products are brewed under the given production process conditions, the qualification rate of the wine products is improved, the waste of the wine-making grape raw materials is reduced, the loss of the enterprises is reduced, and the profits of the enterprises are improved.

At present, enterprises have a plurality of methods for determining which wine grape raw material is used for brewing production, and generally, the wine products brewed by various wine grape raw materials can be compared according to the sensory perception of wine brewing experts, so that what wine product can be obtained by brewing which wine grape raw material is put into the wine. However, the method for making a decision on the wine brewing process only through the sensory perception of the wine brewing expert has certain risks in efficiency and effect, and also has certain lifting space. For the production planning of enterprises, how to scientifically find the quality range of wine brewing raw materials most suitable for brewing wine is an urgent problem to be solved.

The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.

Disclosure of Invention

The invention mainly aims to provide a data-driven model analysis method and a data-driven model analysis device for solving the index range of wine raw materials, and aims to solve the technical problem of how to rapidly and scientifically determine the optimal range of the quality index of wine grape raw materials under the given production process condition so that the wine grape raw materials can produce qualified wine products.

To achieve the above object, the present invention provides a data-driven model analysis method for solving an index range of wine raw materials, the method comprising the steps of:

acquiring a sample database constructed according to raw material indexes of wine grapes and indexes of wine products;

extracting a wine grape raw material sample set and a wine product sample set from the sample database, setting indexes in the wine grape raw material sample set as prediction variables, and setting indexes in the wine product sample set as response variables;

preprocessing the data in the sample database based on the predictive variable and the response variable to obtain processed target data;

predicting the quality index of the wine product based on the target data, and acquiring calculation data in the prediction process;

constructing a wine grape raw material quality standard optimization model according to the calculated data and a preset optimization target;

and determining the quality index range of the wine grape raw materials according to the wine grape raw material quality standard optimization model.

Preferably, the step of preprocessing the data in the sample database based on the predictor variable and the response variable to obtain the processed target data specifically includes:

standardizing the sample set in the sample database based on the predictive variable and the response variable to obtain a standardized wine brewing grape raw material sample set and a standardized wine product sample set;

performing principal component dimensionality reduction on the standard wine grape raw material sample data in the standard wine grape raw material sample set to obtain target wine grape raw material sample data and a principal component factor load matrix;

extracting target wine grape raw material indexes from the target wine grape raw material sample data;

mapping the target wine grape raw material index into a wine grape raw material quality index according to the principal component factor load matrix;

and taking the target wine grape raw material sample data, the standardized wine product sample set and the wine grape raw material quality index as target data.

Preferably, the step of predicting the wine product quality index based on the target data and obtaining the calculation data in the prediction process specifically includes:

generating a training set of a multiple linear regression model according to the target wine making grape raw material sample data and the standardized wine product sample set;

training the multiple linear regression model according to the training set to obtain a prediction model corresponding to the quality index of the wine product;

extracting a target wine product index from the standardized wine product sample set, and determining a regression function corresponding to the target wine product index based on a preset coefficient matrix;

determining goodness of fit according to the regression function;

and taking the prediction model and the goodness-of-fit as calculation data in the prediction process.

Preferably, the step of constructing a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target specifically includes:

searching physicochemical property data of the quality of the wine grape raw material corresponding to the wine grape raw material quality index, and determining wine grape raw material limit constraint based on the physicochemical property data;

determining a wine product limit constraint based on the prediction model and a preset target food quality requirement;

correcting the wine product limit constraint according to the goodness-of-fit to obtain a target wine product limit constraint;

constructing a high-dimensional target space according to a plurality of target wine-making grape raw material indexes, and setting diffusion factors in the high-dimensional target space;

determining diffusion factor constraint according to the diffusion factors and the requirements of a preset target area;

taking the wine grape raw material limit constraint, the wine product limit constraint and the diffusion factor constraint as constraint conditions;

and constructing a wine grape raw material quality standard optimization model according to the constraint conditions and a preset optimization target.

Preferably, the step of determining the quality index range of the wine grape raw materials according to the wine grape raw material quality standard optimization model specifically includes:

converting the wine brewing grape raw material quality standard optimization model into a single-target wine brewing grape raw material quality standard optimization model based on a linear weighting method;

performing quartile calculation on the standard sample data of the standardized wine grape raw materials according to the single-target wine grape raw material quality standard optimization model to obtain quartile calculation result data;

determining a regulation and control model according to the quartile calculation result data and the wine grape raw material quality standard optimization model;

and determining the quality index range of the wine grape raw material according to the regulation and control model.

Preferably, the step of determining a regulation and control model according to the quartile calculation result data and the wine grape raw material quality standard optimization model specifically includes:

selecting lower quarter data and upper quarter data from the quartile calculation result data;

calculating a difference value according to the lower four-quarter data and the upper four-quarter data;

determining a constraint weight according to the difference and a preset standardization method;

optimizing the diffusion factor constraint according to the constraint weight and the diffusion factor to obtain a target diffusion factor constraint;

and regulating and controlling the wine grape raw material quality standard optimization model according to the target diffusion factor constraint to obtain the regulation and control model.

Preferably, the step of determining the quality index range of the wine grape raw material according to the regulation and control model specifically comprises:

calculating a lower threshold value of the quality index of the wine grape raw material and an upper threshold value of the quality index of the wine grape raw material according to the regulation and control model;

performing anti-standardization treatment on the lower bound value of the wine grape raw material quality index and the upper bound value of the wine grape raw material quality index;

and determining the quality index range of the wine grape raw material according to the processing result.

In addition, in order to achieve the above object, the present invention further provides a data-driven model analysis apparatus for solving an index range of wine raw materials, comprising:

the data acquisition module is used for acquiring a sample database constructed according to the raw material indexes of the wine grapes and the indexes of the wine products;

the variable setting module is used for extracting a wine grape raw material sample set and a wine product sample set from the sample database, setting indexes in the wine grape raw material sample set as prediction variables, and setting the indexes in the wine product sample set as response variables;

the data processing module is used for preprocessing the data in the sample database based on the predictive variable and the response variable to obtain processed target data;

the index prediction module is used for predicting the quality index of the wine product based on the target data and acquiring calculation data in the prediction process;

the model construction module is used for constructing a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target;

and the standard determining module is used for determining the quality index range of the wine grape raw materials according to the wine grape raw material quality standard optimization model.

Preferably, the data processing module is further configured to perform standardization processing on the sample set in the sample database based on the predictor variable and the response variable, so as to obtain a standardized wine grape raw material sample set and a standardized wine product sample set;

the data processing module is also used for performing principal component dimensionality reduction on the standardized wine grape raw material sample data in the standardized wine grape raw material sample set to obtain target wine grape raw material sample data and a principal component factor load matrix;

the data processing module is also used for extracting a target wine grape raw material index from the target wine grape raw material sample data;

the data processing module is further used for mapping the target wine brewing grape raw material index into a wine brewing grape raw material quality index according to the principal component factor load matrix;

the data processing module is also used for taking the target wine grape raw material sample data, the standardized wine product sample set and the wine grape raw material quality index as target data.

Preferably, the index prediction module is further configured to generate a training set of a multiple linear regression model according to the target wine grape raw material sample data and the standardized wine product sample set;

the index prediction module is further used for training the multiple linear regression model according to the training set to obtain a prediction model corresponding to the quality index of the wine product;

the index prediction module is also used for extracting a target wine product index from the standardized wine product sample set and determining a regression function corresponding to the target wine product index based on a preset coefficient matrix;

the index prediction module is further used for determining goodness of fit according to the regression function;

and the index prediction module is also used for taking the prediction model and the goodness of fit as calculation data in the prediction process.

The data-driven model analysis method for solving the wine raw material index range, provided by the invention, comprises the steps of obtaining a sample database constructed according to wine grape raw material indexes and wine product indexes; extracting a wine grape raw material sample set and a wine product sample set from the sample database, setting indexes in the wine grape raw material sample set as prediction variables, and setting the indexes in the wine product sample set as response variables; preprocessing data in the sample database based on the predictive variable and the response variable to obtain processed target data; predicting the quality index of the wine product based on the target data, and acquiring calculation data in the prediction process; constructing a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target; and determining the quality index range of the wine grape raw materials according to the standard optimization model of the wine grape raw material quality. By the mode, data in the sample database are preprocessed to obtain target data, calculation data in the process of predicting the quality index of the wine product based on the target data are obtained, the transfer of the color tone and the taste of the wine product to the raw material components and sweetness of the wine grape is completed, a wine grape raw material quality standard optimization model is constructed, the wine grape raw material quality index range is determined by constructing the wine grape raw material quality standard optimization model, the potential of the selected wine grape raw material input for improving the quality of the wine product is exerted, the technical problem that how to rapidly and scientifically determine the optimal range of the wine grape raw material quality index under the given production process condition is solved, and the qualified wine product can be produced by the grape raw materials.

Drawings

FIG. 1 is a schematic flow chart of a first embodiment of a data-driven model analysis method for solving an index range of wine material according to the present invention;

FIG. 2 is a representation of the wine product constraint in two dimensions in a first embodiment of the data-driven model analysis method of the present invention for solving the wine material index range;

FIG. 3 is a schematic flow chart of a second embodiment of the data-driven model analysis method for solving the index range of wine raw materials according to the present invention;

FIG. 4 is a schematic flow chart diagram of a third embodiment of the data-driven model analysis method for solving the index range of wine raw materials according to the present invention;

FIG. 5 is a functional block diagram of a first embodiment of the data-driven model analysis apparatus for solving the index range of wine material according to the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further described with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

An embodiment of the present invention provides a data-driven model analysis method for solving an index range of a wine raw material, and referring to fig. 1, fig. 1 is a flowchart illustrating a first embodiment of the data-driven model analysis method for solving the index range of the wine raw material.

In this embodiment, the data-driven model analysis method for solving the wine raw material index range includes the following steps:

and step S10, acquiring a sample database constructed according to the raw material indexes of the wine grapes and the indexes of the wine products.

It should be noted that, the execution main body of this embodiment may be a computing service device with data processing, program running and network communication functions, such as a smart phone, a tablet computer, a personal computer, etc., and may also be other data-driven model analysis devices capable of implementing the same or similar functions to solve the wine raw material index range, which is not limited in this embodiment. And solving the wine raw material index range, namely solving the wine grape raw material index range.

It should be noted that, a sample database constructed according to the raw material index of wine grape and the index of wine product is known, and the sample database is a high-quality database and includes { x }₁,x₂,…x_m，y₁,y₂,…y_nIn which { x }₁,x₂,…x_mIs a raw material index of wine grape which can evaluate the raw material of wine grape, { y₁,y₂,…y_nIs a wine product index by which a wine product can be evaluated.

The sample database D is:

D_Ais basic information data of wine grape raw material, D_MIs an index data set of wine grape raw material, i.e. a sample set of wine grape raw material, D_NThe index data set is an index data set of the wine product, namely a wine product sample set, wherein t is t wine grape raw materials, k is basic information data of the kth wine grape raw material, m is index data of the mth wine grape raw material, and n is index data of the nth wine product.

Specifically, A₁Is a number, A₁₁Is the 1 st basic attribute, x, of the 1 st wine grape raw material_tmIs the mth basic attribute of the t wine grape raw material. The basic attributes of the wine grape raw materials can include the varieties and producing areas of the wine grapes, wherein the varieties of the wine grape raw materials can include: cabernet sauvignon, Cabernet Gernischt, Pinelizhu, MEILE, XILAI, XIANTIAN indigo naturalis, etc.; the source of the wine grape material may include: langedog, longhuogu, burgundy, boldo, crodilya, california, bohai, eastern bohaiBay, Tianjin coast, etc.

Further, the wine product was evaluated for wine product indices (the following are included but not all indices): sensory requirements ═ clarity, hue, purity, concentration, quality, purity, concentration, durability, quality }, physicochemical requirements ═ alcoholic strength, total sugar, dry extract, volatile acid }, ingredients ═ anthocyanin, tannin, total phenols, wine total flavonoids, resveratrol, trans-polydatin, cis-polydatin, trans-resveratrol, cis-resveratrol, DPPH half-inhibited volume }. The wine product index dataset may consist of all or part of the wine product indices described above for the wine product being evaluated.

Evaluating wine grape raw material indexes (the following indexes are included but not all indexes) of wine grape raw materials: amino acid total amount, single amino acid content, protein, VC content, anthocyanin, fresh weight, tartaric acid, malic acid, citric acid, polyphenol oxidase activity, browning degree, DPPH free radical, total phenol, tannin, total flavonoids of glucose, resveratrol, trans-polydatin, cis-polydatin, trans-resveratrol, cis-resveratrol, flavonol, myricetin, quercetin, kaempferol, isorhamnetin, total sugar, reduced sugar, fructose, glucose, soluble solid matter, pH value, titratable acid, solid acid ratio, dry matter content, ear mass, hundred grains mass, fruit stem ratio, juice yield, pericarp mass and pericarp color. The index dataset for the wine grape raw materials may consist of all or a portion of the above wine grape raw material indices for evaluating the wine grape raw materials.

It should be understood that the manner of obtaining the sample database may be obtaining from a local storage, or obtaining from a network, or obtaining in other manners, which is not limited in this embodiment.

And step S20, extracting a wine grape raw material sample set and a wine product sample set from the sample database, setting indexes in the wine grape raw material sample set as predictive variables, and setting indexes in the wine product sample set as response variables.

It should be noted that, by acquiring the wine making grape raw material sample set and the wine product sample set in the sample database, setting the indexes in the wine product sample set as response variables and setting the indexes in the wine making grape raw material sample set as prediction variables, the constraint on the wine product standard can be converted into the constraint on the wine making grape raw material standard.

And step S30, preprocessing the data in the sample database based on the predictive variable and the response variable to obtain processed target data.

It should be understood that, the step of preprocessing the data in the sample database based on the predictor variable and the response variable to obtain the processed target data specifically includes:

firstly, standardizing a sample set in the sample database based on the predictive variable and the response variable to obtain a standardized wine brewing grape raw material sample set and a standardized wine product sample set. Specifically, the wine making raw material sample set and the wine product sample set are extracted from the sample database based on the predictive variable and the response variable, and then the standardized wine making raw material sample set and the standardized wine product sample set are respectively subjected to standardization processing, so that the standardized wine making raw material sample set and the standardized wine product sample set can be obtained, and the sample set is subjected to standardization processing firstly, so that the data in the sample set after the standardization processing is more convenient for subsequent use.

And secondly, performing principal component dimensionality reduction on the standardized wine grape raw material sample data in the standardized wine grape raw material sample set to obtain target wine grape raw material sample data and a principal component factor load matrix. Specifically, standardized wine grape raw material sample data in a standardized wine grape raw material sample set is obtained, principal component dimensionality reduction is conducted on the standardized wine grape raw material sample data to obtain target wine grape raw material sample data and a principal component factor load matrix, and the purpose of the principal component dimensionality reduction is to eliminate inappropriate data, so that dimensionality of the data is reduced, and the data after dimensionality reduction meets training requirements of a subsequent relation model and reduces training difficulty.

And extracting target wine grape raw material indexes from the target wine grape raw material sample data, mapping the target wine grape raw material indexes into wine grape raw material quality indexes according to the principal component factor load matrix, and taking the target wine grape raw material sample data, the standardized wine product sample set and the wine grape raw material quality indexes as target data for subsequent calculation and use, namely the target data comprises the target wine grape raw material sample data, the standardized wine product sample set and the wine grape raw material quality indexes.

And step S40, predicting the quality index of the wine product based on the target data, and acquiring calculation data in the prediction process.

The step of predicting the wine product quality index based on the target data is specifically as follows:

firstly, generating a training set of a multiple linear regression model according to the target wine grape raw material sample data and the standardized wine product sample set, then training the multiple linear regression model according to the training set to obtain a prediction model corresponding to the quality index of the wine product, wherein the prediction model is used for predicting the quality index of the wine product according to the wine grape raw material quality index, and in the subsequent steps, the wine product limit constraint is determined through the prediction model.

Secondly, extracting a target wine product index from the standardized wine product sample set, determining a regression function corresponding to the target wine product index based on a preset coefficient matrix, and determining the goodness of fit according to the regression function.

And thirdly, taking the prediction model and the goodness of fit obtained in the prediction step process as calculation data in the prediction process for subsequent calculation and use, namely the calculation data in the prediction process comprises the prediction model and the goodness of fit.

It should be understood that, in order to solve the accuracy problem, a correction factor based on regression prediction accuracy is introduced for the constraint of the wine grape raw material standard to reform the constraint condition, so that the optimized wine grape raw material maximally ensures to obtain a product meeting the wine quality standard, and the reliability of optimization decision is improved.

And step S50, constructing a wine grape raw material quality standard optimization model according to the calculated data and a preset optimization target.

It should be noted that the preset optimization target is set to optimize the raw material quality index standard.

It should be noted that the constraint conditions of the wine grape raw material quality standard optimization model are mainly divided into two types:

a first type of constraint: and limiting and restricting the raw materials of the wine grapes. This constraint limits the solution range according to the theoretical reasonable range of index values of the wine grape raw materials.

The second type of constraint: wine product restriction. The constraint is realized by transferring the limit of the wine product to the limit of the wine raw material through the transfer of a relation model of the wine raw material and the wine product, wherein the problem of precision correction of the relation model is also considered. The principle of constructing the limitation constraint of the wine product is researched by taking two wine making grape raw material indexes and two wine product indexes as examples, and the principle can be popularized to the situation of higher dimensional space.

Establishing wine product limit constraints:

wherein x is₁、x₂Two predictive variables (composition and sweetness of wine grape material), y₁、y₂For two response variables (hue and taste of wine product), the prediction model trained was y₁＝f₁(x₁，x₂)、y₂＝f₂(x₁，x₂) (ii) a Wherein the content of the first and second substances,

the lower limit of the standard of the color tone and the taste index of the wine product,

Indicating the upper limits of the color tone and taste index standards of the wine product.

Four solid oblique straight lines as shown in FIG. 2 form the boundary of the wine product restriction constraint, and the area enclosed by the four oblique straight lines is the wine product restriction constraint domain, i.e. the parallelogram area, and the coordinate point (x) in the area₁，x₂) The expressed ingredients and sweetness quality indexes of the wine grape raw materials are mapped into wine product indexes through a relation model f under the condition that first class constraints of the wine grape raw materials are not considered, and the wine product indexes obtained after mapping theoretically meet the limitation constraints of wine products.

The wine product restriction constraint domain shown in fig. 2, considering the first type of constraint, obtains the quality of wine brewing grape material according with the standard of wine brewing grape material quality, i.e. the target area of the required wine quality range is selected in the area, and the required target area is a rectangle in the quality range area of wine brewing grape material. The target area should be a rectangular area containing as many varieties of raw material of wine grape as possible in the quality area of raw material of wine grape that meets the standards for quality of raw material of wine grape.

The rectangular area may be determined by two points: the point of the rectangular region farthest from the origin and the nearest point, and accordingly the other vertices of the rectangular region can be determined. The optimization model aims to ensure that the point farthest from the original point reaches the farthest and the point nearest to the original point reaches the nearest on the premise that all vertexes of the target area are in the area established according to the quality standard of the wine grape raw materials.

For example, two points a and B in fig. 2, coordinate values of these two points are the upper and lower bounds of the quality index of the raw material for brewing grape as the optimization modeling result. The rectangular area drawn by the points A and B is the dotted area in FIG. 2, and the index value of the wine grape raw material represented by the coordinate points included in the area is theoretically preferable. The objective of the optimization model is to find coordinates of two points a and B, and set the coordinates of the two points as an optimization target. The set requirements of the optimization objective are described in conjunction with the rectangles in FIG. 2:

(1) each side represents the value range of the wine quality index represented by the side, so each side needs to be as long as possible.

(2) The sum of the sides of the rectangles should also seek maximum within the feasible domain.

(3) Therefore, when the multi-target is subjected to weighted solving, the optimization weight of each index is set according to the discrete characteristic of the quality index, and the proportion of each weight is visually expressed as the ratio of the side length of a dotted line rectangle.

And reducing the solving space dimension to the index number of the wine grape raw material index set, wherein the wine grape raw material limit constraint and the wine product limit constraint form feasible domains defined by the hyperplanes, and the modeling optimization solving target is to search a hypercube in the feasible domains.

And step S60, determining the quality index range of the wine grape raw material according to the wine grape raw material quality standard optimization model.

It should be noted that, the raw material quality standard optimization model of the wine grapes may be optimized to obtain the regulation and control model, and then the raw material quality index range of the wine grapes is determined according to the regulation and control model, and it should be understood that the index range, i.e., the index standard, represents the same meaning in this embodiment.

It should be noted that the main means of optimization is to adjust the weight in the optimization function, so that the sample for solving the conclusion after regulation and control covers as much as possible, and the weight can be adjusted by methods such as quartile in the discrete measurement index. And calculating the length of the quartile, carrying out normalization processing to obtain a weight, and substituting the weight into the model in the modeling step to carry out secondary modeling solution. The proportion of each side of the rectangle obtained by secondary modeling is the ratio of the dispersion of each index, and then sensitivity analysis is carried out to explore the sum of each index range (sum of each side length) so as to observe whether the probability of improvement exists or not, thereby obtaining the most appropriate solution.

The secondary modeling is regulated and controlled on the basis of the primary modeling as follows:

(1) substituting the maximum side length in the primary modeling solution conclusion into the constraint, and setting the optimization weight of each side length as the dispersion of each index of the sample data.

(2) And (4) setting a reduction step length for the maximum side length in the step (1), and carrying out optimization solution for multiple times.

(1) And the step is to carry out parameter optimization on the basis of a primary modeling result and redefine the proportion constraint of each side length. (2) Setting a contrast experiment to carry out sensitivity analysis, analyzing the change condition of the solution range and obtaining the optimal solution. The optimization solution regulated by the two steps can meet the setting requirement of the optimization target in the modeling step, and the optimal range of each index after optimization is obtained.

In the embodiment, a sample database constructed according to indexes of wine grape raw materials and indexes of wine products is obtained; extracting a wine grape raw material sample set and a wine product sample set from the sample database, setting indexes in the wine grape raw material sample set as prediction variables, and setting indexes in the wine product sample set as response variables; preprocessing data in the sample database based on the predictive variable and the response variable to obtain processed target data; predicting the quality index of the wine product based on the target data, and acquiring calculation data in the prediction process; constructing a wine grape raw material quality standard optimization model according to the calculated data and a preset optimization target; and determining the quality index range of the wine grape raw materials according to the wine grape raw material quality standard optimization model. By the mode, data in the sample database are preprocessed to obtain target data, calculation data in the process of predicting the quality index of the wine product based on the target data are obtained, the transfer of the color tone and the taste of the wine product to the components and sweetness of the wine grape raw materials is completed, a wine grape raw material quality standard optimization model is constructed, the wine grape raw material quality index range is determined by constructing the wine grape raw material quality standard optimization model, the potential of the selection type of the wine grape raw materials to improve the quality of the wine product is exerted, the technical problem that how to rapidly and scientifically determine the optimal range of the wine grape raw material quality index under the given production process condition is solved, and the wine grape raw materials can produce qualified wine products is solved.

As shown in fig. 3, a second embodiment of the data-driven model analysis method for solving the index range of the wine raw material according to the present invention is proposed based on the first embodiment, and the step S30 specifically includes:

step S301, based on the prediction variable and the response variable, carrying out standardization processing on the sample set in the sample database to obtain a standardized wine brewing grape raw material sample set and a standardized wine product sample set.

It should be noted that, the raw material sample sets D of the wine grapes are respectively set_MAnd wine product sample set D_NCarrying out standardization treatment to obtain a standardized wine brewing grape raw material sample set and a standardized wine product sample set, wherein the calculation formula is as follows:

wherein x is_i＝(x_1i；x_2i；…；x_ti) (i ═ 1, 2, …, m) is a sample set of grapevine raw materials;

y_j＝(y_1j；y_2j；…；y_tj) (j ═ 1, 2, …, n) is the wine product sample set; t kinds of wine grape raw materials;

mean value

Standard deviation of

The normalized sample set remains labeled D_M、D_N。

And S302, performing principal component dimension reduction on the standard wine grape raw material sample data in the standard wine grape raw material sample set to obtain target wine grape raw material sample data and a principal component factor load moment.

It should be noted that the specific steps of the dimensionality reduction of the specific principal components are as follows:

(1) and performing Pearson correlation analysis according to the evaluation indexes of the wine products and the evaluation indexes of the wine grape raw materials. If the indexes are strongly correlated, the principal component dimensionality reduction can be considered, and the quality of the wine product under a given production scene is predicted by using a multivariate linear model.

The evaluation indices of wine products are (the following are included but not all indices): sensory requirements ═ clarity, hue, purity, concentration, quality, purity, concentration, durability, quality }, physicochemical requirements ═ alcoholic strength, total sugar, dry extract, volatile acid }, ingredients ═ anthocyanin, tannin, total phenols, wine total flavonoids, resveratrol, trans-polydatin, cis-polydatin, trans-resveratrol, cis-resveratrol, DPPH half-inhibitory volume }.

The evaluation indexes of the wine grape raw materials are as follows (including but not all indexes): amino acid total amount, single amino acid content, protein, VC content, anthocyanin, fresh weight, tartaric acid, malic acid, citric acid, polyphenol oxidase activity, browning degree, DPPH free radical, total phenol, tannin, grape total flavone, resveratrol, trans-resveratrol glycoside, cis-resveratrol glycoside, trans-resveratrol, cis-resveratrol, flavonol, myricetin, quercetin, kaempferol, isorhamnetin, total sugar, reducing sugar, fructose, glucose, soluble solid matter, pH value, titratable acid, solid acid ratio, dry matter content, fruit ear quality, hundred grain quality, fruit stem ratio, juice yield, fruit peel quality and fruit peel color.

(2) And (5) performing principal component dimensionality reduction. In the PCA process of principal component analysis, m is the dimensionality of a wine grape raw material sample set, d is the dimensionality of data after dimensionality reduction, d is specified by observing the total variance quantity of the data after dimensionality reduction, which can explain the original data, and d is the explained variable quantity put into training of a second-stage regression model. And after the PCA process of principal component analysis is finished, discarding the eigenvectors with smaller corresponding eigenvalues, wherein the discarding rule is that if the accumulated variance of the first eigenvalues exceeds 95%, the subsequent eigenvectors can be discarded, and the dimensionality of the principal component for reducing the dimensionality is m-d. The purpose of dimension reduction is to reduce the consumption of training calculation cost and simultaneously make the input training data conform to one of basic assumptions of a multivariate linear model, namely that no correlation exists between interpretation variables.

Finally, the dimension of the standard wine grape raw material sample data is reduced to be the target wine grape raw material sample data D_d：

And obtaining a principal component factor load matrix:

and S303, extracting target wine grape raw material indexes from the target wine grape raw material sample data.

And S304, mapping the target wine grape raw material index into a wine grape raw material quality index according to the principal component factor load matrix.

It should be noted that the principal component factor load matrix is used for calculating s principal component expressions, and the principal component expressions are used for mapping the target raw material indexes of the wine grapes to the raw material quality indexes of the wine grapes. The expression of the main components is as follows:

where i is 1, 2, …, m,

denotes the ith raw material quality index, j is 1, 2, …, d, and the coefficient of the main component with respect to each raw material quality index

，λ＝(λ₁，λ₂，…，λ_j) Is d selected eigenvalues.

And S305, taking the target wine grape raw material sample data, the standardized wine product sample set and the wine grape raw material quality index as target data.

Further, the step S40 includes:

and S401, generating a training set of a multiple linear regression model according to the target wine brewing grape raw material sample data and the standardized wine product sample set.

It should be understood that the target wine grape raw material sample data D after dimensionality reduction is adopted_dBased on the quality index of the wine brewing grape raw material, the quality index of the grape wine product is predicted.

In the training of the multiple linear regression model for each quality index of the wine product, D is set_dAnd standardized wine product sample set D_NTraining set (D) for constructing a multiple linear regression model_d|D_N)。

And S402, training the multiple linear regression model according to the training set to obtain a prediction model corresponding to the quality index of the wine product.

Step S403, extracting a target wine product index from the standardized wine product sample set, and determining a regression function corresponding to the target wine product index based on a preset coefficient matrix.

And S404, determining the goodness of fit according to the regression function.

It should be noted that, since each multiple linear regression model is used to predict univariates, the sample set put into each solution using the least square method is:

wherein D is_dIs the target wine grape raw material sample data after dimensionality reduction, y_j＝(y_1j；y_2j；…；y_tj) (j ═ 1, 2, …, n) is the wine product index set.

And (5) finishing the algorithm flow to obtain a prediction model of the quality index of each wine product. According to coefficient matrix (theta | omega)^*)_n*(m+1)Give the ith wine productThe target regression function:

where i is 1, 2, …, n, j is 1, 2, …, m. Theta is the principal component factor load matrix, the multiple linear regression function f_i(x) Goodness of fit R_i ²Value of [0,1 ]]。

Step S405, the prediction model and the goodness-of-fit are used as calculation data in the prediction process.

In the embodiment, the data in the sample database is subjected to data processing to ensure the accuracy of the prediction model, and then the quality index of the wine product is predicted according to the quality index of the wine grape raw material, so that the transfer of the quality index of the wine product to the quality index of the wine grape raw material is completed, and the model is directly, effectively and reasonably established.

As shown in fig. 4, a third embodiment of the data-driven model analysis method for solving the index range of wine raw material according to the present invention is proposed based on the first embodiment or the second embodiment, and in this embodiment, the step S50 is explained based on the second embodiment, and includes:

and S501, searching physicochemical property data of the wine grape raw material quality corresponding to the wine grape raw material quality index, and determining wine grape raw material limit constraints based on the physicochemical property data.

It should be understood that the lower and upper bounds x of the criteria for determining the various quality indicators of the wine grape raw material^low，x^upVectors composed of decision variables;

wherein the content of the first and second substances,

it is to be noted that, based on the physicochemical properties of the quality of the wine grape raw material, the first type of constraint is determined: and limiting and restricting the raw materials of the wine grapes.

The restriction of the raw material of the wine grape is the index range of the raw material of the wine grape which is input in the actual production and meets the requirement, and the vector space expressed by the restriction is as follows:

X_limit＝{x|l_α≤x_α≤u_α，α＝1，2，…，m)

wherein l_αAnd u_αRespectively is the index x of the raw material of wine grape_αThere are m such indices, lower and upper bounds.

And step S502, determining the limitation constraint of the wine product based on the prediction model and the preset target food quality requirement.

And S503, correcting the wine product limit constraint according to the goodness-of-fit to obtain a target wine product limit constraint.

It should be noted that, determining the wine product limit constraint based on the prediction model and the preset target food quality requirement, and correcting the wine product limit constraint according to the goodness-of-fit to obtain the target wine product limit constraint specifically includes:

(1) determination of quality standard of wine product

Assuming that the feasible range of physicochemical property, sensory score or TPA index in the quality index of the wine product is L ═ L (L ═ L)₁，L₂，…，L_n)，U＝(U₁，U₂，…，U_n) Respectively representing the lower and upper bounds of the quality index of the target wine product.

(2) Determination of correction factors

To improve the accuracy of the restriction of the wine product, the degree of fit R is used²And correcting the constraint of the quality standard of the wine product. Goodness of fit obtained for the prediction stage

The specific correction method comprises the following steps:

a. calculating Delta_k＝L_k-U_k，Δ_kIs the initial value of the range.

b. Calculating goodness of fit R of each regression function_i ²"correction factor":

c. calculating the upper and lower limits of the quality index of the wine product, and reducing and increasing the same size according to the correction quantity of the upper and lower limits:

[L′_j，U′_j]＝[L_j+0、5*Δ_j*∈_j，U_j-0.5*Δ_j*∈_j]

where j is 1, 2, …, n, which represents the quality standard in the k-th product index, and 0.5 represents the sharing of the upper and lower limit correction amounts in the upper and lower limits.

(3) Wine product restriction

The wine product limit constraint is set by two steps of (1) and (2):

wherein k is 1, 2, …, n, X_processA feasible domain limited by a "wine product limit constraint",

is a regression function of the kth wine product index, L ═ L₁，L₂，…，L_n)，U＝(U₁，U₂，…，U_n) Respectively representing the lower and upper bounds of the quality index of the target wine product.

And step S504, constructing a high-dimensional target space according to the indexes of the target wine brewing grape raw materials, and setting diffusion factors in the high-dimensional target space.

And step S505, determining diffusion factor constraint according to the diffusion factor and the requirement of a preset target area.

It should be noted that the preset target area requirement is a requirement that the target area is maximized.

It should be noted that, based on the requirement that the target area is maximized as the target, the third type of constraint is determined: and (3) diffusion factor constraint, namely setting a diffusion factor in a high-dimensional target space consisting of a plurality of target wine-making grape raw material indexes, and substituting the diffusion factor into a linear weighting method target to meet the following requirements:

(1) the range is reasonable, so that the calculation of the range of the wine grape raw materials is guaranteed to be a high-reliability range solution given under the condition of comprehensively considering the prediction errors of various products.

(2) The range is the widest, and the coverage degree as much as possible in the restriction of wine grape raw materials and the restriction of wine products is pursued to be finally solved.

Let x^low，x^upThe lower and upper bounds for the final solution of each material, where δ is the diffusion factor, and if there are m such indices, then there are

And S506, taking the wine grape raw material limit constraint, the wine product limit constraint and the diffusion factor constraint as constraint conditions.

And step S507, constructing a standard optimization model of the quality of the wine grape raw materials according to the constraint conditions and a preset optimization target.

The quality index range of the wine grape raw materials solved by the multi-objective optimization model is wide in range under the premise of meeting constraint conditions, and the maximization of delta is met firstly, which is a primary objective:

max f₁＝δ

and the final solution upper and lower bounds are to satisfy the maximum and minimum, respectively, with the following secondary objectives:

in conclusion, the wine grape raw material quality standard optimization model is a multi-objective optimization mathematical model, and is specifically expressed as follows:

wherein x is^low，x^upIs the lower and upper limit values of the standard of each quality index of wine grape raw material, delta is diffusion factor, f is objective function, and L ═ L₁，L₂，…，L_n)，U＝(U₁，U₂，…，U_n) Respectively representing the lower and upper bounds of the target wine product quality index, X_limitVector space, y, expressed in index ranges of wine grape raw materials meeting requirements_kIs an index set of wine products.

Further, the step S60 includes:

and S601, converting the wine grape raw material quality standard optimization model into a single-target wine grape raw material quality standard optimization model based on a linear weighting method.

It should be noted that, by using a linear weighting method, a weight value with magnitude difference is set according to the priority of a target and the importance of a same-level target, and a wine grape raw material quality standard optimization model is converted into a single-target wine grape raw material quality standard optimization model:

wherein, β_αAre the weights of the indexes, each weight is equal and

β' is diffusion factor weight, and the ratio β is the general ratio when finding a proper solution_αOne or two orders of magnitude larger.

In a high dimensional space constructed from index variables of the whole wine grape raw material, X_limitLimit the space eachThe basic value range of the variable of the dimension forms a hypercube space, and X_processA super-dimensional space with an irregular shape is constructed by the dimensional variables and the linear function of L, U. The two spaces are included or partially overlapped in a high-dimensional space. The objective function is set to find a hypercube with 2 in the overlapping region of the two high dimensional spaces^mA vertex in which two points have respective coordinates of

And

the two coordinates already contain all vertex coordinate information of the hypercube, so that the standard range of the quality of all wine grape raw materials can be determined only by determining the two coordinates.

And step S602, performing quartile calculation on the standardized wine grape raw material sample data according to the single-target wine grape raw material quality standard optimization model to obtain quartile calculation result data.

It should be understood that the third type of constraint is optimized by using the optimal solution calculated by the single-target spelling standardized model, and the quartile of the raw material standardized data is calculated, namely the quartile is used for arranging all values from small to large and dividing all values into four equal parts, and the values are positioned at the positions of three dividing points.

And S603, determining a regulation and control model according to the quartile calculation result data and the standard optimization model for the quality of the wine grape raw materials.

Further, in step S603, the method specifically includes:

selecting lower quarter data and upper quarter data from the quartile calculation result data; calculating a difference value according to the lower four-quarter data and the upper four-quarter data; determining a constraint weight according to the difference and a preset standardization method; optimizing the diffusion factor constraint according to the constraint weight and the diffusion factor to obtain a target diffusion factor constraint; and regulating and controlling the wine grape raw material quality standard optimization model according to the target diffusion factor constraint to obtain a regulation and control model.

It should be noted that a value at a position of 25% (lower quartile data) and a value at a position of 75% (upper quartile data) are selected from the calculation result data, and a difference value R ═ R (R) is calculated from the lower quartile data and the upper quartile data₁，r₂，…，r_α)。

Calculating new solving weight and constraint weight of each decision variable by a min-max standardization method:

β_α＝r_α/∑_αr_α

at this time, the weights still satisfy

But not equal. According to the diffusion factor delta₀Adjusting third type constraints

It is modified into

Where ρ is the relaxation factor used to reduce the diffusion factor such that the coordinates of the final solution space (hypercube) are labeled with δ₀The relaxation space of ρ, in which each vertex can adjust for variations. And gamma is a floating variable and is used for controlling the final solution of each index to float according to the discrete characteristics of each quality index of the actual wine grape raw material so as to obtain different solution ranges.

The regulated and controlled multi-objective optimization model with the priority is as follows:

wherein x is^low，x^upIs the lower and upper limit values of the standard of each quality index of wine grape raw material, delta is diffusion factor, f is objective function, and L ═ L₁，L₂，…，L_n)，U＝(U₁，U₂，…，U_n) Respectively representing the lower and upper bounds of the target wine product quality index, X_limitVector space, y, expressed in index ranges of wine grape raw materials meeting requirements_kIs an index set of wine products, and gamma is a floating variable.

Wherein P1 is far larger than P2, the value range of the relaxation factor rho is [0, 0.5], namely the value of at most half of the maximum diffusion factor is used as the floating range of the hypercube coordinate, the step length of 0.05 which is rho is set, 10 times of solving is carried out, and the result is compared to obtain the most reasonable solution range.

And S604, determining the quality index range of the wine grape raw materials according to the regulation and control model.

Further, in step S604, the method specifically includes:

calculating a lower threshold value of the quality index of the wine grape raw material and an upper threshold value of the quality index of the wine grape raw material according to the regulation and control model; performing anti-standardization treatment on the lower bound value of the wine grape raw material quality index and the upper bound value of the wine grape raw material quality index; and determining the quality index range of the wine grape raw material according to the processing result.

It should be noted that, when the linear weighting method is used for solving, new solving weights of each decision variable are substituted, and the multiple objectives are converted into a single objective:

wherein γ is a "floating variable", x^low，x^upThe lower limit value (lower limit value of quality index of wine grape raw material) and the upper limit value (upper limit value of quality index of wine grape raw material) of the standards of all quality indexes of wine grape raw material, m indexes, β_αβ' is order of magnitude ratio β for updated optimization weights_αIs high.

It should be understood that the standard deviation σ is normalized by the data normalization process_iAnd mean value

i represents the quality index of the first material, and the solving result x^low、x^upPerforming anti-standardization treatment to obtain the optimized range of each raw material index

The denormalization formula is:

in the embodiment, the raw material quality index standard of the wine grapes is determined by constructing a standardized raw material quality optimization model of the wine grapes, the structure and parameters of the raw material quality index model are optimized and adjusted in order to contain as many raw material varieties as possible in the established variation range of the raw material quality index of the wine grapes, the standardized raw material quality optimization model of the wine grapes is improved to obtain a regulation and control model, and the shape of a hypercube is optimized, so that the most appropriate raw material quality index range of the target product of the wine grapes is obtained.

In addition, referring to fig. 5, an embodiment of the present invention further provides a data-driven model analysis apparatus for solving a wine raw material index range, including:

and the data acquisition module 10 is used for acquiring a sample database constructed according to the raw material indexes of the wine grapes and the indexes of the wine products.

And the variable setting module 20 is configured to extract a wine grape raw material sample set and a wine product sample set from the sample database, set the indexes in the wine grape raw material sample set as prediction variables, and set the indexes in the wine product sample set as response variables.

And the data processing module 30 is configured to pre-process the data in the sample database based on the predictive variable and the response variable, and obtain processed target data.

And the index prediction module 40 is used for predicting the quality index of the wine product based on the target data and acquiring calculation data in the prediction process.

And the model building module 50 is used for building a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target.

And the standard determining module 60 is used for determining the quality index range of the wine brewing grape raw material according to the wine brewing grape raw material quality standard optimization model.

In the embodiment, the data acquisition module 10 is used for acquiring a sample database constructed according to the raw material indexes of wine grapes and the indexes of wine products; a variable setting module 20, configured to extract a wine grape raw material sample set and a wine product sample set from the sample database, set an index in the wine grape raw material sample set as a prediction variable, and set an index in the wine product sample set as a response variable; a data processing module 30, configured to pre-process the data in the sample database based on the predictor variable and the response variable, and obtain processed target data; the index prediction module 40 is used for predicting the quality index of the wine product based on the target data and acquiring calculation data in the prediction process; the model building module 50 is used for building a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target; and the standard determining module 60 is used for determining the quality index range of the wine grape raw materials according to the wine grape raw material quality standard optimization model.

In an embodiment, the model building module 50 is further configured to search physicochemical property data of the wine grape raw material quality corresponding to the wine grape raw material quality index, and determine a wine grape raw material limitation constraint based on the physicochemical property data; determining a wine product limit constraint based on the prediction model and a preset target food quality requirement; correcting the wine product limit constraint according to the goodness-of-fit to obtain a target wine product limit constraint; constructing a high-dimensional target space according to a plurality of target wine-making grape raw material indexes, and setting diffusion factors in the high-dimensional target space; determining diffusion factor constraints according to the diffusion factors and the requirements of a preset target area; taking the wine grape raw material limit constraint, the wine product limit constraint and the diffusion factor constraint as constraint conditions; and constructing a wine grape raw material quality standard optimization model according to the constraint conditions and a preset optimization target.

In an embodiment, the standard determination module 60 is further configured to convert the wine brewing grape raw material quality standard optimization model into a single-target wine brewing grape raw material quality standard optimization model based on a linear weighting method; performing quartile calculation on the standard sample data of the standardized wine grape raw materials according to the single-target wine grape raw material quality standard optimization model to obtain quartile calculation result data; determining a regulation and control model according to the quartile calculation result data and the wine grape raw material quality standard optimization model; and determining the quality index range of the wine grape raw material according to the regulation and control model.

In an embodiment, the criterion determining module 60 is further configured to select a lower quartile data and an upper quartile data from the quartile calculation result data; calculating a difference value according to the lower four-quarter data and the upper four-quarter data; determining a constraint weight according to the difference and a preset standardization method; optimizing the diffusion factor constraint according to the constraint weight and the diffusion factor to obtain a target diffusion factor constraint; and regulating and controlling the raw material quality standard optimization model of the wine grapes according to the target diffusion factor constraint to obtain the regulation and control model.

In an embodiment, the standard determination module 60 is further configured to calculate a wine brewing grape raw material quality index lower bound value and a wine brewing grape raw material quality index upper bound value according to the regulation and control model; carrying out anti-standardization treatment on the lower threshold value of the wine grape raw material quality index and the upper threshold value of the wine grape raw material quality index; and determining the quality index range of the wine grape raw material according to the processing result.

Other embodiments or specific implementation methods of the data-driven model analysis device for solving the wine raw material index range can refer to the above method embodiments, and are not described herein again.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.

Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solution of the present invention can be embodied in the form of software product, which is stored in an estimator readable storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling an intelligent device (such as a mobile phone, an estimator, a data-driven model analysis device for solving the index range of wine ingredients, an air conditioner, or a data-driven model analysis device for solving the index range of wine ingredients on a network) to execute the method according to the embodiments of the present invention.

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the present specification and drawings, or used directly or indirectly in other related fields, are included in the scope of the present invention.

Claims

1. A data-driven model analysis method for solving an index range of wine raw materials, comprising the steps of:

preprocessing data in the sample database based on the predictive variable and the response variable to obtain processed target data;

constructing a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target;

2. The method of claim 1, wherein the step of preprocessing the data in the sample database based on the predictor variables and the response variables to obtain processed target data comprises:

3. The method for analyzing a data-driven model for solving the index range of wine ingredients as set forth in claim 2, wherein the step of predicting the wine product quality index based on the target data and obtaining the calculation data in the prediction process comprises:

determining goodness of fit according to the regression function;

4. The method for analyzing a data-driven model for solving the index range of wine raw materials according to claim 3, wherein the step of constructing the wine grape raw material quality standard optimization model according to the calculated data and the preset optimization objective specifically comprises:

searching physicochemical property data of the wine grape raw material quality corresponding to the wine grape raw material quality index, and determining wine grape raw material limit constraint based on the physicochemical property data;

5. The data-driven model analysis method for solving wine raw material index range according to claim 4, wherein the step of determining a wine grape raw material quality index range according to the wine grape raw material quality standard optimization model specifically comprises:

converting the wine grape raw material quality standard optimization model into a single-target wine grape raw material quality standard optimization model based on a linear weighting method;

6. The method for analyzing a data-driven model for solving the index range of wine raw materials as recited in claim 5, wherein the step of determining a regulation model based on the quartile calculation result data and the wine grape raw material quality standard optimization model specifically comprises:

7. The data-driven model analysis method for solving wine raw material index range according to claim 5, wherein the step of determining wine grape raw material quality index range according to the regulation and control model specifically comprises:

8. A data-driven model analysis device for solving an index range of wine raw materials, the data-driven model analysis device for solving the index range of wine raw materials comprising:

the model building module is used for building a wine grape raw material quality standard optimization model according to the calculation data and a preset optimization target;

9. The apparatus of claim 8, wherein the data processing module is further configured to normalize the sample set in the sample database based on the predictor variable and the response variable to obtain a normalized wine grape raw material sample set and a normalized wine product sample set;

the data processing module is also used for extracting target wine grape raw material indexes from the target wine grape raw material sample data;

the data processing module is also used for mapping the target wine grape raw material index into a wine grape raw material quality index according to the principal component factor load matrix;

10. The data-driven model analysis apparatus for solving the wine feedstock index range of claim 9, wherein the index prediction module is further configured to generate a training set of a multiple linear regression model from the target wine brewing grape feedstock sample data and the set of standardized wine product samples;