CN109325626A - Method based on apple feedstock specifications prediction dried product integrated quality - Google Patents
Method based on apple feedstock specifications prediction dried product integrated quality Download PDFInfo
- Publication number
- CN109325626A CN109325626A CN201811147985.5A CN201811147985A CN109325626A CN 109325626 A CN109325626 A CN 109325626A CN 201811147985 A CN201811147985 A CN 201811147985A CN 109325626 A CN109325626 A CN 109325626A
- Authority
- CN
- China
- Prior art keywords
- dried product
- evaluation index
- apple
- index
- raw material
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Preparation Of Fruits And Vegetables (AREA)
Abstract
The invention discloses the methods based on apple feedstock specifications prediction dried product integrated quality, comprising: determines dried product evaluation index, establishes the corresponding relationship of the dried product evaluation index and integrated quality;Determine apple raw material evaluation index relevant to the dried product evaluation index;Obtain the apple raw material evaluation index data of training sample, and the dried product evaluation index data of training sample are obtained, the integrated quality of each training sample is calculated, using the apple raw material evaluation index data of training sample as input layer, using integrated quality as output layer, training obtains neural-network learning model;Using the apple raw material evaluation index data of the neural-network learning model and dried product to be measured, the integrated quality of dried product to be measured is predicted.The present invention is without being determined in advance mapping relations between input and output, and predetermined speed is fast, precision is high.
Description
Technical field
The present invention relates to apples to process related fields.It is more particularly related to be predicted based on apple feedstock specifications
The method of dried product integrated quality.
Background technique
Apple occupies first of the big fruit in the world four, and China is maximum apple production state and country of consumption in the world, China's apple
Plantation is extensively and kind is also relatively abundant.Apple can also be squeezed into juice and dried product is made in addition to being used as fresh food, but fresh food supplies at present
It should be saturated, the process change motive force of development is insufficient, needs diversification processing technology.Dry apple is as a kind of novel leisure
Food, delicious health, can enrich apple secondary industry, drive Raw material processing conversion, have preferable development prospect.At present
The methods of step analysis, grey correlation independent analysis, such as Wu's thickness nine et al. are mostly used to the research of product integrated quality evaluation
The quantitative evaluation method of sweet orange kind processing suitability is differentiated with percentage, Wang Yongyang et al. is based on grape quality apparent characteristic
Predict that Table Grape organoleptic quality, such method can only evaluate the processing characteristics of studied kind, unpredictable unknown raw material
The processing performance of sample.The method of existing association property of raw material and product property is also merely with discriminant function, linear regression etc.
Linear model, such as Song Jie et al. handle achievement data with range transformation method, obtain the quality evaluation result that pork rinses food, this
Class model can not handle complicated nonlinear correspondence relation between index, and there are limitation, the scope of application are small, prediction unknown sample adds
The low problem of work performance accuracy rate.Therefore a kind of quick, accurate dry apple integrated quality prediction technique need to be sought, from source
Screening is suitable for the dry kind of system on head raw material, produces the dried product of high-quality, optimizes Apple Industry structure, promotes China's apple
Industry development increases agricultural benefit, improves farmers' income.
Summary of the invention
It is an object of the present invention to provide it is a kind of based on apple feedstock specifications prediction dried product integrated quality method,
It is not necessary that mapping relations between input and output are determined in advance, predetermined speed is fast, precision is high.
In order to realize these purposes and other advantages according to the present invention, provide drying based on the prediction of apple feedstock specifications
The method of product integrated quality, comprising:
It determines dried product evaluation index, establishes the corresponding relationship of the dried product evaluation index and integrated quality;
Determine apple raw material evaluation index relevant to the dried product evaluation index;
The apple raw material evaluation index data of training sample are obtained, and obtain the dried product evaluation index number of training sample
According to, quality grade or the scoring of each training sample are calculated, using the apple raw material evaluation index data of training sample as input layer, with
Quality grade is output layer, and training obtains neural-network learning model;
Using the apple raw material evaluation index data of the neural-network learning model and dried product to be measured, predict to be measured dry
The integrated quality of product.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, determines drying judge
Valence refers to that calibration method includes: to choose several initial dried product evaluation indexes for being used to evaluate product quality, to several initial dry
Product evaluation index is analyzed using factor analysis, and selected characteristic value is greater than 1 factor, and will respectively represent each factor
Each initial dried product evaluation index as dried product evaluation index.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, integrated quality definition
For the weighted average of dried product evaluation index.Preferably, described based on the comprehensive product of apple feedstock specifications prediction dried product
The method of matter determines the weight of dried product evaluation index according to analytic hierarchy process (AHP).
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, it is described drying to judge
Valence index includes lustre index, texture index and flavor index, with the comprehensive product of lustre index, texture index and flavor index definition
Matter.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, it is determining to be done with described
The method of the relevant apple raw material evaluation index of product evaluation index includes: to choose several initial apples for being used to evaluate apple raw material
Fruit raw material evaluation index carries out correlation analysis to initial apple raw material evaluation index and dried product evaluation index, chooses related
Coefficient is higher than the initial apple raw material evaluation index of setting value as apple raw material evaluation index.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, further includes:
Verifying sample is chosen, the apple raw material evaluation index data of verifying sample is obtained, utilizes neural-network learning model
The quality grade of prediction verifying sample increases training sample if accuracy is lower than 90%, updates neural-network learning model,
Or it improves setting value, more new apple raw material evaluation index, and re -training and obtains neural-network learning model.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, the neural network
The training parameter of learning model is as follows: maximum cycle is 1000~20000, and learning rate is 0.1~1, factor of momentum 0.1
~0.5, error amount is 0.00005~0.1.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, it is initially drying to judge
It is many that valence index includes at least sense organ class index, processing class index and nutrition class index, the quantity of initial dried product evaluation index
In 17 kinds.
Preferably, the method based on apple feedstock specifications prediction dried product integrated quality, initial apple raw material
Evaluation index includes at least physics class index, sense organ class index, processing class index and nutrition class index, initial apple raw material evaluation
The quantity of index is no less than 15 kinds.
The present invention is include at least the following beneficial effects:
The present invention establishes the quick predict model of dry apple integrated quality with artificial neural network, can screen suitable
It is processed into the apple variety of dried apple slices, the quality of dry apple is promoted from source, dry apple can be greatly promoted
The development of industry.The present invention has found apple raw material and dried product comprehensive score or quality grade from apple property of raw material
Association, construct model, according to model quickly, accurately predict dried product integrated quality, without be determined in advance input and output it
Between mapping relations, and when using comprehensive score, can objective quantitative reflection apple raw material difference, when using quality grade,
It intuitively can qualitatively reflect the difference of apple raw material.
Further advantage, target and feature of the invention will be partially reflected by the following instructions, and part will also be by this
The research and practice of invention and be understood by the person skilled in the art.
Specific embodiment
The present invention will be further described in detail below with reference to the embodiments, to enable those skilled in the art referring to specification
Text can be implemented accordingly.
It should be appreciated that such as " having ", "comprising" and " comprising " term used herein do not allot one or more
The presence or addition of a other elements or combinations thereof.
In a kind of technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, comprising:
It determines dried product evaluation index, establishes the corresponding relationship of the dried product evaluation index and integrated quality;
Determine apple raw material evaluation index relevant to the dried product evaluation index;
The apple raw material evaluation index data of training sample are obtained, and obtain the dried product evaluation index number of training sample
According to the integrated quality of each training sample being calculated, using the apple raw material evaluation index data of training sample as input layer, to integrate product
Matter is output layer, and training obtains neural-network learning model;
Using the apple raw material evaluation index data of the neural-network learning model and dried product to be measured, predict to be measured dry
The integrated quality of product.
In the above-mentioned technical solutions, the quality of dried product mainly considers its organoleptic quality, processing quality and nutritional quality, sense
Official's quality includes Color Quality (L*, a*, b* value), texture quality (hardness, brittleness), puffed degree, processing quality include yield rate,
Rehydration ratio, moisture content, nutritional quality include soluble solid, titratable acid, total phenol content, pectin content, soluble sugar,
Sugar-acid ratio, crude protein, crude fibre.Several indexs are chosen from the above index of quality as dried product evaluation index.It is comprehensive
Quality is the quality of reflection product quality, for example can be product quality grade or dried product comprehensive score, quality grade
Or comprehensive score is determined according to dried product evaluation index, such as rule of thumb or statistical method establishes single or multiple drying judge
The mathematical relationship of valence index and quality grade or comprehensive score, make it possible to be determined according to dried product evaluation index quality grade or
Comprehensive score.Apple raw material evaluation index mainly considers physical qualities, organoleptic quality, processing quality and nutritional quality, as quality,
It is volume, density, fruit shape index, skin L* value, meat L* value, titratable acid, soluble sugar, crude fibre, crude protein, polyphenol, pectin, more
Phenoloxidase activity, microelement etc. rule of thumb or statistical method are chosen to evaluate with dried product from these index of quality and are referred to
The biggish several indexs of correlation are marked as apple raw material evaluation index.Training sample is chosen from all parts of the country, obtains above-mentioned step
Suddenly determining apple raw material evaluation index data and and corresponding integrated quality.It is nerve with apple raw material evaluation index data
The input layer of network, using integrated quality as output layer, training obtains neural-network learning model, for sample to be tested, inputs apple
Fruit raw material evaluation index data, can be obtained the integrated quality of prediction.Here dried product is conventional method acquisition, such as a kind of
Preparation method includes cleaning, removes stalk, stoning, removes the peel, slice, predrying, hot-air seasoning, and another preparation method includes cleaning,
Stalk, stoning are removed, is removed the peel, slice, predrying, hot wind predry, all wet, pressure difference flash distillation.As can be seen that the technical program can basis
Apple raw material predicts the quality of dried product, and it is not necessary that mapping relations between input and output are determined in advance, process is simple, time-consuming short.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, really
The method for determining dried product evaluation index includes: to choose several initial dried product evaluation indexes for being used to evaluate product quality, right
Several initial dried product evaluation indexes are analyzed using factor analysis, and selected characteristic value is greater than 1 factor, and will distinguish
Each initial dried product evaluation index of each factor is represented as dried product evaluation index.It is dry that the technical program provides a kind of determination
The method of product evaluation index, chooses initial dried product evaluation index first, and initial dried product evaluation index needs to cover sense organ
The index of quality, processing quality index and Nutrition quality indicator, and every class index needs to choose multiple types, to initial dried product
Evaluation is analyzed with factor analysis, such as establishes factorial analysis rotation component matrix, choose all characteristic values greater than 1 because
Then son finds out the corresponding representative initial dried product evaluation index of each factor, combines these initial dried product evaluation indexes
For dried product evaluation index.In this way, the grade of dried product can be preferably determined by the dried product evaluation index simplified,
And reduce subsequent calculation amount.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality is comprehensive
Close the weighted average that quality definition is dried product evaluation index.The technical program utilizes the weighted average of dried product evaluation index
Value defines integrated quality, and weight can determine by experience or statistical method, and the integrated quality that this method determines is more objective, accurate,
Further dried product can be classified using weighted average, or weighted average is directly used as comprehensive score.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, root
The weight of dried product evaluation index is determined according to analytic hierarchy process (AHP).The technical program provides the determination method of weight, that is, uses layer
Fractional analysis such as establishes judgment matrix, and compared to experience, analytic hierarchy process (AHP) is more acurrate.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, institute
Stating dried product evaluation index includes lustre index, texture index and flavor index, with lustre index, texture index and flavor index
Define integrated quality.Here there is provided by organoleptic quality determine integrated quality in the way of, can be rule of thumb by color, matter
Ground and flavor are classified, every grade determine a fraction range, then according to subjective feeling to the color of dried product, quality and
Flavor is given a mark, by the sum of score be used as final score, then may further using final score as comprehensive score, or
Multiple quality grades are divided into dried product using final score.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, really
The method of fixed apple raw material evaluation index relevant to the dried product evaluation index includes: that selection is several for evaluating apple original
The initial apple raw material evaluation index of material carries out correlation point to initial apple raw material evaluation index and dried product evaluation index
Analysis chooses related coefficient and is higher than the initial apple raw material evaluation index of setting value as apple raw material evaluation index.This technology side
Case provides the method for determining apple raw material evaluation index, initial apple raw material evaluation index need to cover description physical qualities,
The index of organoleptic quality, processing quality and nutritional quality, and every class index needs to choose multiple types, to initial apple raw material
Evaluation index and dried product evaluation index carry out correlation analysis, are such as analyzed using SPSS software, and it is big to choose related coefficient
In the initial apple raw material evaluation index of setting value, group is combined into apple raw material evaluation index, and setting value is empirically determined, passes through
The apple raw material evaluation index that correlation analysis obtains can preferably react dried product evaluation index, and then can be improved prediction
Accuracy rate.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, also
Include:
Verifying sample is chosen, the apple raw material evaluation index data of verifying sample is obtained, utilizes neural-network learning model
The integrated quality of prediction verifying sample increases training sample if accuracy is lower than 90%, updates neural-network learning model,
Or it improves setting value, more new apple raw material evaluation index, and re -training and obtains neural-network learning model.The technical program
By setting accuracy probabilistic index come optimization neural network learning model, when accuracy is lower than 90%, then increase training sample, to mind
It is updated through network, or simplifies apple raw material evaluation index, so that apple raw material evaluation index can more reflect integrated quality.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, institute
The training parameter for stating neural-network learning model is as follows: maximum cycle is 1000~20000, and learning rate is 0.1~1, is moved
Measuring the factor is 0.1~0.5, and error amount is 0.00005~0.1.The technical program provides training parameter, which can
Preferably adapt to the prediction of dried product integrated quality.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, just
Beginning dried product evaluation index includes at least sense organ class index, processing class index and nutrition class index, initial dried product evaluation index
Quantity be no less than 17 kinds.Here there is provided the selection mode and quantity of preferred initial dried product evaluation index, sense organ class refers to
Mark such as L* value, a* value, b* value, processing class index such as rehydration ratio, yield rate, moisture content, puffed degree, nutrition class index such as pH,
Titratable acid, soluble solid, total phenol, pectin, soluble sugar, crude fibre, crude protein, sugar-acid ratio can preferably reflect
Integrated quality.
In another technical solution, the method based on apple feedstock specifications prediction dried product integrated quality, just
Beginning apple raw material evaluation index includes at least physics class index, sense organ class index, processing class index and nutrition class index, initial apple
The quantity of fruit raw material evaluation index is no less than 15 kinds.Here there is provided the selecting partys of preferred initial apple raw material evaluation index
Formula and quantity, physics class index such as quality, volume, density, fruit shape index, fruit stone ratio, Rind hardness, flesh firmness, sense organ
Class index such as L* value (skin), a* value (skin), b* value (skin), L* value (meat), a* value (meat), b* value (meat), processing class index such as contain
Water rate, nutrition class index such as pH, titratable acid, soluble solid, crude fibre, crude protein, Vc, reduced sugar, total reducing sugar, can
Preferably reflect dried product integrated quality.
Embodiment 1
(1) apple sample is selected
34 apple varieties from various parts of the country are chosen as experimental raw, part variety name, the place of production are shown in Table 1.Fruit
Real maturity period sampling, has no mechanical damage, no disease and pests harm.
1 part apple variety title of table and the place of production
(2) dry apple quality grade is determined
Dried product integrated quality mainly considers that its organoleptic quality, processing quality and nutritional quality, organoleptic quality include color
Quality (L*, a*, b* value), texture quality (hardness, brittleness), puffed degree, processing quality include that yield rate, rehydration ratio, moisture contain
Amount, nutritional quality includes soluble solid, titratable acid, total phenol content, pectin content, soluble sugar, sugar-acid ratio, thick egg
White, crude fibre amounts to 17 indexs (initial dried product evaluation index).Dimensionality reduction (Factor minute is carried out to 17 indexs of dried product
Analysis), 6 factors can be obtained, the 77% of all information can be accounted for, the results are shown in Table 2.As shown in Table 2, the characteristic value of preceding 6 factors is big
In 1, adding up variance contribution ratio is 77.402%.The factor 1 mainly combines the information of titratable acid and sugar-acid ratio, and the two is shown
Extremely significant correlation (R=-0.849), and the weighted value of titratable acid is higher, therefore screens generation of the titratable acid as the factor 1
Table index.The factor 2 mainly combines the information of b* value, brittleness and total phenol, wherein brittleness and the significant related (R=of total phenol content
0.426), extremely significant related (R=-0.472) to b* value, and brittleness is that the important sense organ of measuring apple dried product processing quality refers to
Mark, therefore screen representative index of the brittleness as the factor 2.The factor 4 mainly combines the information of L* value, a* value, and L* value represents
Bright-dark degree, a* value represent red green degree, embodiment dried product Color Quality, and show extremely significant correlation (R=-
0.865) representative index of the higher L* value of weighted value as the factor 4, is screened herein.Meanwhile the factor 3, the factor 5 and the factor 6
The weighted value of middle puffed degree, soluble sugar and crude protein is apparently higher than other indexs, therefore screens puffed degree, soluble sugar respectively
Representative index with crude protein as the factor 3, the factor 5 and the factor 6.To sum up, it is filtered out in 17 index of quality titratable
Acid, brittleness, puffed degree, L* value, soluble sugar and crude protein are as dried product evaluation index.
2 dried product index factor of table analysis rotation component matrix
Note: PC1-PC6 respectively indicates the 1st to the 6th main gene
According to the obtained 6 dried product evaluation indexes of screening to the significance level of dried product integrated quality, using 1-9 scale
Method is established Y-P judgment matrix (table 3), and its consistency ratio (consistency ratio, CR) is 0.03, less than 0.1, it is believed that
Judgment matrix approach is acceptable, to obtaining the weight of dried product evaluation index after matrix characteristic vector normalized.Therefore
Dry apple quality comprehensive evaluation model is represented by YScore=L* value × 0.3724+ brittleness × 0.2665+ puffed degree ×
0.1583+ titratable acid content × 0.0890+ soluble sugar content × crude protein content × 0.0569 0.0569+.
3 judgment matrix Y-P of table
Note: Y represents dried product integrated quality, and P represents dried product evaluation index, and P1-P6 respectively indicates the L* of dried product
Value, brittleness, puffed degree, titratable acid, soluble sugar and crude protein
According to the dry apple quality comprehensive evaluation model of building, calculates 34 dry apple sample qualities synthesis and comment
Point, and it is classified, score 0.7-0.8 is level-one, and 0.6-0.7 is second level, and 0.5-0.6 is three-level, 0.4-0.5 tetra-
Grade is Pyatyi less than 0.4, the results are shown in Table 4.
The sequence of 4 dried product integrated quality of table and score
(3) apple raw material evaluation index is screened
Correlation analysis is carried out to dried product evaluation index and initial apple raw material evaluation index (5 first row of table), as a result
As shown in table 5.Obtain fruit shape index, pulp a* value, pH value, titratable acid content, VCContent, fruit stone ratio, crude protein content,
12 pulp b* value, density, soluble solid content, crude fiber content, total sugar content indexs refer to as the evaluation of apple raw material
Mark.
The initial apple raw material evaluation index of table 5 and dried product evaluation index correlation analysis
(4) learning model is constructed
This research 29 samples of random screening from 34 apple samples establish learning model, and remaining 5 samples are done
Product property grade forecast.Wherein mode input layer is 12 apple raw materials such as Apple fruit shape index, pulp a* value, pH value
Evaluation index value, model output layer are the corresponding product quality grade of Apple.The optimal implicit number of plies of model by software from
Dynamic to generate, the number of plies is 9.Remaining each training parameter selection is as follows: maximum cycle 5000, learning rate 0.3, factor of momentum
0.2, error amount 0.001.
(5) forecast sample is verified
Applied Learning model, to the product quality grade forecast of 5 kind apples, the results are shown in Table 6.
6 neural network prediction result of table
Embodiment 2
(1) apple sample is selected
34 apple varieties from various parts of the country are chosen as experimental raw, part variety name, the place of production are shown in Table 7.Fruit
Real maturity period sampling, has no mechanical damage, no disease and pests harm.
7 part apple variety title of table and the place of production
(2) dry apple comprehensive score is determined
Dried product sensory scores mainly consider its color, quality, flavor score, and sense organ evaluating meter is as shown in table 8, and synthesis is commented
Divide as shown in table 9.
8 dry apple sense organ evaluating meter of table
9 dry apple comprehensive score of table
(3) apple raw material evaluation index is screened
By dried product color score, quality score, flavor score respectively with initial apple raw material evaluation index (table 10 the 1st
Column) correlation analysis is carried out, as shown in table 10, filtering out apple raw material evaluation index is pericarp L* value, pulp a* value, pulp b*
Value, pH, titratable acid, soluble solid, crude fibre, Vc, reduced sugar, magnesium.
10 apple raw material evaluation index of table and dried product evaluation index correlation analysis
(4) learning model is constructed
This research 28 samples of random screening from 34 apple samples establish learning model, and remaining 6 samples are done
The prediction of product comprehensive score.Wherein mode input layer be Apple pericarp L* value, it is pulp a* value, pulp b* value, pH, titratable
10 acid, soluble solid, crude fibre, Vc, reduced sugar, magnesium apple raw material evaluation index values, model output layer are Apples
Real corresponding dried product comprehensive score.The optimal hidden layer number of plies of model is 9.Remaining each training parameter selection is as follows: maximum is followed
Ring number 8000, learning rate 0.2, factor of momentum 0.1, error amount 0.01.
(5) forecast sample is verified
Applied Learning model predicts the comprehensive score of the dried product of 5 kind apples, as a result as shown in table 11.
11 prediction result of table
Although the embodiments of the present invention have been disclosed as above, but its is not only in the description and the implementation listed
With it can be fully applied to various fields suitable for the present invention, for those skilled in the art, can be easily
Realize other modification, therefore without departing from the general concept defined in the claims and the equivalent scope, the present invention is simultaneously unlimited
In specific details and embodiment shown and described herein.
Claims (10)
1. the method based on apple feedstock specifications prediction dried product integrated quality characterized by comprising
It determines dried product evaluation index, establishes the corresponding relationship of the dried product evaluation index and integrated quality;
Determine apple raw material evaluation index relevant to the dried product evaluation index;
The apple raw material evaluation index data of training sample are obtained, and obtain the dried product evaluation index data of training sample, are counted
The integrated quality for calculating each training sample is with integrated quality using the apple raw material evaluation index data of training sample as input layer
Output layer, training obtain neural-network learning model;
Using the apple raw material evaluation index data of the neural-network learning model and dried product to be measured, dried product to be measured is predicted
Integrated quality.
2. the method as described in claim 1 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that really
The method for determining dried product evaluation index includes: to choose several initial dried product evaluation indexes for being used to evaluate product quality, right
Several initial dried product evaluation indexes are analyzed using factor analysis, and selected characteristic value is greater than 1 factor, and will distinguish
Each initial dried product evaluation index of each factor is represented as dried product evaluation index.
3. the method as described in claim 1 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that comprehensive
Close the weighted average that quality definition is dried product evaluation index.
4. the method as claimed in claim 3 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that root
The weight of dried product evaluation index is determined according to analytic hierarchy process (AHP).
5. the method as described in claim 1 based on apple feedstock specifications prediction dried product integrated quality, described drying to judge
Valence index includes lustre index, texture index and flavor index, with the comprehensive product of lustre index, texture index and flavor index definition
Matter.
6. the method as described in claim 1 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that really
The method of fixed apple raw material evaluation index relevant to the dried product evaluation index includes: that selection is several for evaluating apple original
The initial apple raw material evaluation index of material carries out correlation point to initial apple raw material evaluation index and dried product evaluation index
Analysis chooses related coefficient and is higher than the initial apple raw material evaluation index of setting value as apple raw material evaluation index.
7. the method based on apple feedstock specifications prediction dried product integrated quality stated such as claim 1, which is characterized in that also wrap
It includes:
Verifying sample is chosen, the apple raw material evaluation index data of verifying sample is obtained, is predicted using neural-network learning model
The quality grade of sample is verified, if accuracy is lower than 90%, increases training sample, updates neural-network learning model, or
It improves setting value, more new apple raw material evaluation index, and re -training and obtains neural-network learning model.
8. the method as described in claim 1 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that institute
The training parameter for stating neural-network learning model is as follows: maximum cycle is 1000~20000, and learning rate is 0.1~1, is moved
Measuring the factor is 0.1~0.5, and error amount is 0.00005~0.1.
9. the method as claimed in claim 2 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that just
Beginning dried product evaluation index includes at least sense organ class index, processing class index and nutrition class index, initial dried product evaluation index
Quantity be no less than 17 kinds.
10. the method as claimed in claim 6 based on apple feedstock specifications prediction dried product integrated quality, which is characterized in that
Initial apple raw material evaluation index includes at least physics class index, sense organ class index, processing class index and nutrition class index, initially
The quantity of apple raw material evaluation index is no less than 15 kinds.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811147985.5A CN109325626A (en) | 2018-09-29 | 2018-09-29 | Method based on apple feedstock specifications prediction dried product integrated quality |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811147985.5A CN109325626A (en) | 2018-09-29 | 2018-09-29 | Method based on apple feedstock specifications prediction dried product integrated quality |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109325626A true CN109325626A (en) | 2019-02-12 |
Family
ID=65266122
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811147985.5A Pending CN109325626A (en) | 2018-09-29 | 2018-09-29 | Method based on apple feedstock specifications prediction dried product integrated quality |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109325626A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110320333A (en) * | 2019-05-07 | 2019-10-11 | 西北农林科技大学 | A kind of visualization assessment method based on a variety of apple quality indexs |
CN111292006A (en) * | 2020-02-25 | 2020-06-16 | 武汉轻工大学 | Method and device for obtaining raw material quality range based on quality range of yellow rice wine product |
CN111291496A (en) * | 2020-02-25 | 2020-06-16 | 武汉轻工大学 | Data-driven model analysis method and device for solving index range of rice dumpling raw materials |
CN111340361A (en) * | 2020-02-25 | 2020-06-26 | 武汉轻工大学 | Data-driven model analysis method and device for solving index range of yellow rice wine raw materials |
CN111695743A (en) * | 2020-06-23 | 2020-09-22 | 信阳农林学院 | Industrial food safety production monitoring method, device, equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105675539A (en) * | 2016-01-07 | 2016-06-15 | 北京市农林科学院 | Comprehensive evaluation method of quality of agricultural products |
CN106202267A (en) * | 2016-06-30 | 2016-12-07 | 四川农业大学 | The Citrus chachiensis Hort. integrated quality evaluation methodology of a kind of yellow fruit |
CN107464020A (en) * | 2017-08-03 | 2017-12-12 | 中南林业科技大学 | A kind of rice made products processes raw material rapid screening method |
US10031118B1 (en) * | 2017-06-20 | 2018-07-24 | Institute of Food Science and Technology, CAAS | Method of determining and evaluating quality of peanut raw material suitable for protein processing |
-
2018
- 2018-09-29 CN CN201811147985.5A patent/CN109325626A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105675539A (en) * | 2016-01-07 | 2016-06-15 | 北京市农林科学院 | Comprehensive evaluation method of quality of agricultural products |
CN106202267A (en) * | 2016-06-30 | 2016-12-07 | 四川农业大学 | The Citrus chachiensis Hort. integrated quality evaluation methodology of a kind of yellow fruit |
US10031118B1 (en) * | 2017-06-20 | 2018-07-24 | Institute of Food Science and Technology, CAAS | Method of determining and evaluating quality of peanut raw material suitable for protein processing |
CN107464020A (en) * | 2017-08-03 | 2017-12-12 | 中南林业科技大学 | A kind of rice made products processes raw material rapid screening method |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110320333A (en) * | 2019-05-07 | 2019-10-11 | 西北农林科技大学 | A kind of visualization assessment method based on a variety of apple quality indexs |
CN111292006A (en) * | 2020-02-25 | 2020-06-16 | 武汉轻工大学 | Method and device for obtaining raw material quality range based on quality range of yellow rice wine product |
CN111291496A (en) * | 2020-02-25 | 2020-06-16 | 武汉轻工大学 | Data-driven model analysis method and device for solving index range of rice dumpling raw materials |
CN111340361A (en) * | 2020-02-25 | 2020-06-26 | 武汉轻工大学 | Data-driven model analysis method and device for solving index range of yellow rice wine raw materials |
CN111340361B (en) * | 2020-02-25 | 2023-04-28 | 武汉轻工大学 | Data-driven model analysis method and device for solving yellow wine raw material index range |
CN111291496B (en) * | 2020-02-25 | 2023-05-12 | 武汉轻工大学 | Data-driven model analysis method and device for solving range of raw material indexes of rice dumplings |
CN111695743A (en) * | 2020-06-23 | 2020-09-22 | 信阳农林学院 | Industrial food safety production monitoring method, device, equipment and storage medium |
CN111695743B (en) * | 2020-06-23 | 2024-04-05 | 信阳农林学院 | Industrial food safety production monitoring method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109325626A (en) | Method based on apple feedstock specifications prediction dried product integrated quality | |
Nyasordzi et al. | Utilizing the IAD index to determine internal quality attributes of apples at harvest and after storage | |
Jha et al. | Modeling of color values for nondestructive evaluation of maturity of mango | |
CN107228924B (en) | A kind of adequate proteins processing peanut raw material quality determination and its evaluation method | |
CN103954734B (en) | A kind of mensuration of lichee quality of suitable fruit juice production and evaluation method | |
Igbari et al. | Morphological characterization of mango (Mangifera indica L.) cultivars from south-west Nigeria | |
Khadivi-Khub et al. | Phenotypic characterization and relatedness among some Iranian pomegranate (Punica granatum L.) accessions | |
CN106596453A (en) | Method for discriminating wild and cultured sea bass based on near infrared spectroscopy technology | |
CN112163779A (en) | Method for evaluating quality of pear fruits | |
Antimiani et al. | Agri-food exports in the enlarged European Union. | |
CN109242341A (en) | Method based on apple feedstock specifications prediction fruit juice integrated quality | |
Timkhum et al. | Non-destructive classification of durian maturity of ‘Monthong’cultivar by means of visible spectroscopy of the spine | |
CN112465366A (en) | Comprehensive evaluation method for persimmon quality based on entropy weight TOPSIS model | |
Bilgin et al. | Analysis of Fruit Properties of’Hacıhaliloğlu’x’Boccucia’Hybrid Population | |
CN113780845A (en) | Sweet potato germplasm resource evaluation method based on multi-criterion decision | |
Nantongo et al. | End‐user preferences to enhance prospects for varietal acceptance and adoption in potato breeding in Uganda | |
Kök et al. | Physical and biochemical traits of selected grape varieties cultivated in Tekirdağ, Turkey | |
Polinar et al. | Near-infrared spectroscopy for non-destructive prediction of maturity and eating quality of ‘Carabao’mango (Mangifera indica L.) fruit | |
Cosmulescu et al. | Some pomological characteristics of medlar (Mespilus germanica L.) genotypes | |
CN109409687A (en) | Method based on apple feedstock specifications prediction dried product Color Quality | |
Letaief et al. | Sensory and instrumental texture measurements for assessing grape seed parameters during fruit development | |
CN109242342A (en) | Method based on apple feedstock specifications prediction apple fresh food organoleptic quality | |
Huang et al. | Neural network for classification of Chinese zither panel wood via near-infrared spectroscopy | |
CN109409580A (en) | Method based on apple feedstock specifications prediction fruit juice turbidity | |
Guo et al. | Determination of storage period of harvested plums by near‐infrared spectroscopy and quality attributes |
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 |