CN105866223A - Electronic tongue detection information-based Wuyi rock tea taste grade discrimination method - Google Patents
Electronic tongue detection information-based Wuyi rock tea taste grade discrimination method Download PDFInfo
- Publication number
- CN105866223A CN105866223A CN201610186782.1A CN201610186782A CN105866223A CN 105866223 A CN105866223 A CN 105866223A CN 201610186782 A CN201610186782 A CN 201610186782A CN 105866223 A CN105866223 A CN 105866223A
- Authority
- CN
- China
- Prior art keywords
- electrode
- grade
- detection information
- variable
- grades
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/26—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating electrochemical variables; by using electrolysis or electrophoresis
- G01N27/416—Systems
- G01N27/48—Systems using polarography, i.e. measuring changes in current under a slowly-varying voltage
Abstract
The invention relates to an electronic tongue detection information-based Wuyi rock tea taste grade discrimination method. The method comprises the following steps: preparing a liquid sample, acquiring detection information, constructing a Logistic regression discriminant connection function, and calculating the centroid distance of various grades of samples. The discrimination method effectively solves the problem of unable long-time examination of artificial examination due to sensory fatigue, well eliminates discrimination result errors among examination personnel, improves the stability and the consistence of taste grade examination results, and has the advantages of convenient and fast examination process, short time, and realization of completion of a whole process from sample preparation to discrimination completion in 5min; and a model analysis result obtained through the method has good stability, the consistency of the model analysis result with an artificial examination result is higher than 90%. The method has very good practical values, and is of great significance to application of automatic control means in the production process of Wuyi rock tea and quality evaluation description, sales and circulation of final tea products.
Description
Technical field
The present invention relates to a kind of tea grades method of discrimination, be specifically related to a kind of based on electronic tongues detection information
Wuyi cliff tea flavour Grade Judgment, belongs to food technology field.
Background technology
Wuyi cliff tea belongs to the north of Fujian Province oolong tea, originates in Wuyi Mountain Area, the north, Fujian Province, China's well-known tea it
One, world-renowned because of " the petrous bone fragrance of a flower " of its uniqueness, it is exactly the essence in China's tradition well-known tea all the time
Product.Wuyi cliff tea quality constitutes complexity, and its grade evaluates " first weight rock rhythm ", with flavour " clear, fragrant, sweet,
Living " person be that upper, traditional quality evaluation needs " sense organ " of the dependence person of evaluating to complete, and not only evaluates the cycle relatively
Long, and it is easily subject to the interference of various subjectivity and extraneous factor, " concordance " and " quantization " of data
All not enough, it is difficult to objective and stably sign rock tea actual quality, it is therefore necessary to find a kind of standardization
And easy-operating means realize the stabilizing determination of Wuyi cliff tea green tea grade.
Electronic tongues is that it is by low for a kind of typical Artifical Taste system (Artificial Taste System, ATS)
Selectivity, nonspecific mutual sensitive sensor array, be equipped with suitable pattern recognition or multivariate statistics side
Method, and then realize analyzing.What electronic tongues was measured is not the quantitative and qualitative result of a certain chemical composition, but
Obtain the Global Information of the surveyed material sense of taste.The composition of electronic tongues mainly has: taste sensor array, signal
Excite acquisition system and PRS.Taste sensor array therein is the core component of electronic tongues,
It is to be made up of several even tens different sensors, " taste bud " of each sensor people seemingly equally,
A class chemical substance can be experienced, and then realize the simulation to people's sense of taste.According to the operation principle of sensor,
Electronic tongue sensor can be divided into: Copper diethlydithiocarbamate, impedance spectral pattern sensor and volt-ampere type sensor three major types.
Volt-ampere electronic tongue sensor is to be made up of some noble metal working electrodes and 1 reference electrode.Pass through
Control system controls to provide electric pulse, acquisition system measure and store response current multifrequency pulse, measuring
Journey combines the routine significantly pulse electromotive force of several different pulse intervals one as different frequency bands
Rising, as pulse electromotive force excitation signal, along with increasing of frequency, pulse interval is reduced to a certain degree,
Due to reciprocal action between former and later two pulse electromotive forces in same frequency range so that excitation signal deviation is original
Routine significantly pulse voltammetry, close to small size pulse voltammetry and step pulse voltammetry, so different
Electrode presents different differentiation effects at different frequencies to detected solution sample, and then achieves taste
The collection of taste fingerprint.
The detection information utilizing electronic tongues carries out the differentiation of rock tea flavour grade, and its mathematics essence is i.e. to electronics
" finger print data " that tongue is obtained carries out classification prediction, and the analysis method of such problem is had a lot by statistics,
Common solution mainly has: " Logistic regression analysis " and " techniques of discriminant analysis ".Work as output variable
When being two classified variables, use " binary logistic regression analysis ", when output variable is many classified variables,
Generally use " multinomial Logistic regression analysis ".When " multinomial Logistic regression analysis " is for solution 1
When individual or multiple covariants are with an ordered categorization dependent variable relation, also referred to as " grade returns point the method
Analysis ".
Summary of the invention
It is an object of the invention to provide a kind of Wuyi cliff tea flavour grade discrimination side based on electronic tongues detection information
Method.
The purpose of the present invention is achieved through the following technical solutions.
A kind of Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information, it is characterised in that differentiate
The step of method is as follows:
(1) prepared by sample liquid: weighs 5.0g Tea Samples to be checked, adds boiling water 110mL, adds a cover immersion 3min
After, leach millet paste, be cooled to room temperature, be sample liquid；
(2) acquisition testing information: select volt-ampere electronic tongue, immerses its combination electrode in sample liquid；With 0.2v
For current potential stepped voltage, gathering each electrode at 1Hz, 10Hz and 100Hz, three power up frequency and correspondence
Stepping forcing voltage under pulse peak, peak flex point, minimum, minimum flex point, it is thus achieved that electricity
Double tongue detection information, after this detection information being derived, chooses " the electricity of max p10v palladium electrode 1Hz therein
Place value " it is variable X1；" potential value of min p10v palladium electrode 1Hz " is variable X2；" d2p10v palladium
The potential value of electrode 1Hz " it is variable X3；" potential value of min p2v palladium electrode 1Hz " is variable X4；
" potential value of max p9v palladium electrode 100Hz " is variable X5；" the electricity of max p7v gold electrode 100Hz
Place value " it is variable X6；" potential value of max p6v Ti electrode 100Hz " is variable X7, above variable
Unit is " millivolt "；The combination electrode of described volt-ampere electronic tongue be platinum electrode, gold electrode, tungsten electrode,
Silver electrode, Ti electrode and palladium electrode；Described acquisition testing information, the forcing voltage in combination electrode be+10V～
In the range of-10V, choose its pulse value critical potential, build electronic tongues flavour detection information；Described " max
The potential value of p10v palladium electrode 1Hz " refer to that this variate-value is under the forcing voltage of 10v and the electric frequency of 1Hz,
Gather the maximum obtaining palladium electrode response pctential；" potential value of min p10v palladium electrode 1Hz " refers to this change
Value is under the forcing voltage of 10v and the electric frequency of 1Hz, gathers the palladium electrode response pctential obtained
Little value；" potential value of d2p10v palladium electrode 1Hz " refers to that this variable is at the forcing voltage of 10v and 1Hz
Electric frequency under, gather the corresponding potential value of palladium electrode potential change flex point obtained.
(3) utilize the sample of known grades, structure Logistic to return and differentiate connectivity function:
Logistic returns and differentiates that the expression formula of connectivity function is respectively as follows:
Link1 level
=3.128-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)
Link2 level
=6.330-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)
Described " Link1 level " representative sample flavour grade is the connectivity function of 1 grade；" Link2 level " representative sample
Product flavour grade is the connectivity function of 2 grades；
(4) each rating sample differentiation probability of outcome:
Each rating sample differentiates that the formula of probability of outcome is:
P1 grade=1/ 1+Exp (-Link1 level)
P2 grades=1/ 1+Exp (-Link2 level)-P1 grade
P3 grades=1-P1 grade-P2 grades
Described P1 grade, P2 grades, P3 gradesNumerical value, each flavour grade judged for " electronic tongues detection information " ownership
Probability, the grade of the P corresponding to its maximum, be the flavour grade of tea sample.
Advantages of the present invention and beneficial effect:
1, the Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information of the present invention, effectively solves
The problem that " manually evaluating " process can not be carried out for a long time due to " sensory fatigue "；
2, the relative immobility of discrimination standard, preferably eliminates the differentiation resultant error that difference is evaluated between personnel,
Improve stability and the concordance of flavour grade assessment result；
3, evaluating overall process convenient and swift, elapsed time is short, each sample from sample preparation to complete differentiate full mistake
Journey only needs 5min；
4, the large sample amount tea sample using the senior person of evaluating of multidigit to carry out under kilter due to discrimination model is examined
Completing based on commenting data to build, meanwhile, modal analysis results shows that model has good stability, with
The concordance of artificial assessment result is also above 90%, and therefore this differentiation result has good practical value, energy
Enough auxiliary well realize the concordance of Wuyi cliff tea flavour grade scale, objectivity and versatility, to Folium Camelliae sinensis
The application of Automated condtrol means in production process, and the evaluation table of final tea products quality addresses sale
Circulation is cleared away the obstacles.
Detailed description of the invention
A kind of based on electronic tongues detection information Wuyi cliff tea flavour grade discrimination for the openest present invention
Method, is illustrated below in conjunction with embodiment.
Embodiment 1, a kind of based on electronic tongues detection information Wuyi cliff tea flavour Grade Judgment, including with
Lower step:
(1) prepared by sample liquid: weighing 5.0g Wuyi cliff tea is Tea Samples to be checked, adds boiling water 110mL, adds a cover
After soaking 3min, leach millet paste, be cooled to room temperature, be sample liquid；
(2) acquisition testing information: select Smartouge volt-ampere electronic tongue, by electronic tongues combination electrode,
I.e. platinum electrode, gold electrode, tungsten electrode, silver electrode, Ti electrode and palladium electrode, immerses step (1) respectively
In sample liquid, gather each electrode powering up frequency and be 1Hz, 10Hz and 100Hz and forcing voltage is
With 0.2v for current potential stepped voltage in the range of+10V～-10V, it is thus achieved that the X corresponding to electronic tongues detection information1
～X7Value be respectively as follows: X1=4.242 × 10-4、X2=3.639 × 10-4、X3=5.261 × 10-5、X4=3.207 × 10-4、
X5=3.412 × 10-4、X6=9.946 × 10-5、X7=2.843 × 10-5；
(3) structure Logistic recurrence discriminant function:
Utilize the detection information of electronic tongues, carry out Logistic regression analysis through Statistics Application software, it is thus achieved that grow
Taste grade is the logit connectivity function Link Function of 1,2 rating sample, and its expression formula is respectively as follows:
Link1 level
=3.128-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)；
Link2 level
=6.330-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)；
By electronic tongues testing result, i.e. X1～X7Value substitute into respectively two connectivity functions " Link1 level " and
" Link2 level ", result of calculation is Link1 level=-0.777, Link2 level=2.425；
(4) distribution probability of each rating sample:
By calculate obtain " Link1 level " and " Link2 level " numerical value bring into its grade belong to new probability formula:
P1 grade=1/ (1+Exp (-Link1 level)；
P2 grades=1/ (1+Exp (-Link2 level)-P1 grade；
P3 grades=1-P1 grade-P2 grades；
Result of calculation is, P1 grade=0.685, P2 grades=0.234, P3 grades=0.081, because 0.685 > 0.234 > 0.081,
So P1 grade>P2 grades>P3 grades, therefore, this tea sample flavour grade is judged as the maximum probability of 1 grade, reaches 68.5%,
Therefore determine that this millet paste flavour grade is 1 grade.
Claims (4)
1. a Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information, it is characterised in that sentence
Other method is as follows:
(1) prepared by sample liquid: weighs 5.0g Tea Samples to be checked, adds boiling water 110mL, adds a cover immersion 3min
After, leach millet paste, be cooled to room temperature, be sample liquid；
(2) acquisition testing information: select volt-ampere electronic tongue, immerses its combination electrode in sample liquid；With 0.2v
For current potential stepped voltage, gathering each electrode at 1Hz, 10Hz and 100Hz, three power up frequency and correspondence
Stepping forcing voltage under pulse peak, peak flex point, minimum, minimum flex point, it is thus achieved that electricity
Double tongue detection information, after this detection information being derived, chooses " the electricity of max p10v palladium electrode 1Hz therein
Place value " it is variable X1；" potential value of min p10v palladium electrode 1Hz " is variable X2；" d2p10v palladium
The potential value of electrode 1Hz " it is variable X3；" potential value of min p2v palladium electrode 1Hz " is variable X4；
" potential value of max p9v palladium electrode 100Hz " is variable X5；" the electricity of max p7v gold electrode 100Hz
Place value " it is variable X6；" potential value of max p6v Ti electrode 100Hz " is variable X7；
(3) structure Logistic returns and differentiates connectivity function:
Logistic returns and differentiates that the expression formula of connectivity function is respectively as follows:
Link1 level
=3.128-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)；
Link2 level
=6.330-(17539.404X1-772.891X2-187291.898X3-82947.127X4+107598.920X5-36914
.754X6+5663.373X7)；
Described " Link1 level " representative sample flavour grade is the connectivity function of 1 grade；" Link2 level " representative sample
Product flavour grade is the connectivity function of 2 grades；
(4) each rating sample differentiation probability of outcome:
Each rating sample differentiates that the formula of probability of outcome is:
P1 grade=1/ 1+Exp (-Link1 level)；
P2 grades=1/ 1+Exp (-Link2 level)-P1 grade；
P3 grades=1-P1 grade-P2 grades；
Described P1 grade、P2 grades、P3 gradesNumerical value, each flavour grade judged for " electronic tongues detection information " ownership
Probability, the grade of the P corresponding to its maximum, for the flavour grade of tea sample.
A kind of Wuyi cliff tea flavour grade based on electronic tongues detection information the most according to claim 1 is sentenced
Other method, it is characterised in that described acquisition testing information, in electronic tongues combination electrode forcing voltage be+10V～
In the range of-10V, choose its pulse value critical potential, build electronic tongues flavour detection information.
A kind of Wuyi cliff tea flavour grade based on electronic tongues detection information the most according to claim 1 is sentenced
Other method, it is characterised in that the unit of variable is " millivolt ".
A kind of Wuyi cliff tea flavour grade based on electronic tongues detection information the most according to claim 1 is sentenced
Other method, it is characterised in that described electronic tongues is volt-ampere electronic tongue, its combination electrode is platinum electrode, gold electricity
Pole, tungsten electrode, silver electrode, Ti electrode and palladium electrode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610186782.1A CN105866223B (en) | 2016-03-29 | 2016-03-29 | A kind of Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610186782.1A CN105866223B (en) | 2016-03-29 | 2016-03-29 | A kind of Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105866223A true CN105866223A (en) | 2016-08-17 |
| CN105866223B CN105866223B (en) | 2018-06-29 |
Family
ID=56626299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610186782.1A Active CN105866223B (en) | 2016-03-29 | 2016-03-29 | A kind of Wuyi cliff tea flavour Grade Judgment based on electronic tongues detection information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105866223B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106560702A (en) * | 2016-10-20 | 2017-04-12 | 中国计量大学 | Wuyi rock tea production place identification method through combination of electronic tongue and chromatographic separation technology |
| CN110632162A (en) * | 2019-09-11 | 2019-12-31 | 山西大学 | Method for identifying wild astragalus and cultivated astragalus in ground |
| CN110646267A (en) * | 2019-10-29 | 2020-01-03 | 福建农林大学 | Method for judging grade and baking degree of Wuyi cinnamon tea |
| CN111220658A (en) * | 2020-01-06 | 2020-06-02 | 广东博智林机器人有限公司 | Tea product identification method and device and electronic tongue |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000283958A (en) * | 1999-03-31 | 2000-10-13 | Anritsu Corp | Method and device for detecting poison |
| CN101487825A (en) * | 2009-03-03 | 2009-07-22 | 中国农业大学 | Method for recognizing tea kinds and/or grades |
| CN103134850A (en) * | 2013-03-01 | 2013-06-05 | 河南农业大学 | Tea quality rapid detection apparatus and detection method based on characteristic fragrance |
| US20130240255A1 (en) * | 2012-03-15 | 2013-09-19 | Hui-Ping Chiang | Test piece and manufacturing method thereof |
| CN103499615A (en) * | 2013-07-30 | 2014-01-08 | 中国标准化研究院 | Selection method of sensors in tree species model for quality detection of Longjing tea based on genetic algorithm |
| CN104122308A (en) * | 2014-07-15 | 2014-10-29 | 浙江工商大学 | Difference degree calculation method and tea quality identification method based on electronic tongue detection |
| CN104543064A (en) * | 2015-01-28 | 2015-04-29 | 四川中测量仪科技有限公司 | Intelligent control mode for tidying tea strips |
-
2016
- 2016-03-29 CN CN201610186782.1A patent/CN105866223B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000283958A (en) * | 1999-03-31 | 2000-10-13 | Anritsu Corp | Method and device for detecting poison |
| CN101487825A (en) * | 2009-03-03 | 2009-07-22 | 中国农业大学 | Method for recognizing tea kinds and/or grades |
| US20130240255A1 (en) * | 2012-03-15 | 2013-09-19 | Hui-Ping Chiang | Test piece and manufacturing method thereof |
| CN103134850A (en) * | 2013-03-01 | 2013-06-05 | 河南农业大学 | Tea quality rapid detection apparatus and detection method based on characteristic fragrance |
| CN103499615A (en) * | 2013-07-30 | 2014-01-08 | 中国标准化研究院 | Selection method of sensors in tree species model for quality detection of Longjing tea based on genetic algorithm |
| CN104122308A (en) * | 2014-07-15 | 2014-10-29 | 浙江工商大学 | Difference degree calculation method and tea quality identification method based on electronic tongue detection |
| CN104543064A (en) * | 2015-01-28 | 2015-04-29 | 四川中测量仪科技有限公司 | Intelligent control mode for tidying tea strips |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106560702A (en) * | 2016-10-20 | 2017-04-12 | 中国计量大学 | Wuyi rock tea production place identification method through combination of electronic tongue and chromatographic separation technology |
| CN110632162A (en) * | 2019-09-11 | 2019-12-31 | 山西大学 | Method for identifying wild astragalus and cultivated astragalus in ground |
| CN110632162B (en) * | 2019-09-11 | 2021-07-27 | 山西大学 | Method for identifying wild astragalus and cultivated astragalus in ground |
| CN110646267A (en) * | 2019-10-29 | 2020-01-03 | 福建农林大学 | Method for judging grade and baking degree of Wuyi cinnamon tea |
| CN111220658A (en) * | 2020-01-06 | 2020-06-02 | 广东博智林机器人有限公司 | Tea product identification method and device and electronic tongue |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105866223B (en) | 2018-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102084794B (en) | Method and device for early detecting crop pests based on multisensor information fusion | |
| CN105866223A (en) | Electronic tongue detection information-based Wuyi rock tea taste grade discrimination method | |
| CN108875913B (en) | Tricholoma matsutake rapid nondestructive testing system and method based on convolutional neural network | |
| CN102621118B (en) | Early warning method of greenhouse vegetable diseases and insect pests | |
| CN108037256B (en) | The rapid assay methods of rice eating-quality | |
| CN108960315A (en) | A kind of processed meat products quality intelligent evaluation system and method | |
| CN110161194A (en) | It is a kind of based on odiferous information BP fuzzy neuron identification the recognition methods of fruit freshness, apparatus and system | |
| CN103399050B (en) | Method for rapidly evaluating ginseng-adulterated American ginseng based on mouth feel information | |
| CN107505358B (en) | A kind of detection method of the fermentation of black tea index of quality based on electrical characteristic parameter | |
| CN107045013A (en) | Beef taste quality automatic classification detector and its detection method | |
| CN101620195A (en) | Method for detecting internal quality of jirou sweet persimmon by smell sensor | |
| CN102914584A (en) | Rapid detection system and detection method for lactogenesis mixing | |
| CN107490778B (en) | A kind of multifactor quantitative analysis method of measurement error influence factor | |
| CN105866222A (en) | Electronic tongue detection information-based white peony taste grade discrimination method | |
| CN103376282A (en) | Taste information based method for rapid evaluation of ginsengs of different ages | |
| CN105891130B (en) | A method of the different spectral informations of correction determine material information error | |
| CN105758897B (en) | A kind of white peony fragrance Grade Judgment based on electronic nose detection information | |
| CN105699304B (en) | A kind of method of material information representated by acquisition spectral information | |
| CN105866203A (en) | Electronic tongue detection information-based Baihao Yinzhen taste grade discrimination method | |
| CN207336424U (en) | A kind of pears class fruit maturity nondestructive testing instrument | |
| CN105445333B (en) | The method of Fast nondestructive evaluation wheat seed viability | |
| Sarkar et al. | Taste recognizer by multi sensor electronic tongue: a case study with tea quality classification | |
| CN111638306A (en) | Crop dynamic monitoring method, device, equipment and system | |
| CN110687166A (en) | Real-time nondestructive evaluation method and device for life vitality of arbor and electronic equipment | |
| CN103412005B (en) | Detection method for storage time of sweet oranges |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200713 Address after: 350018 No. 84, Chengshan mountain, Cangshan District, Fujian, Fuzhou Patentee after: FUJIAN CHUNLUN GROUP Co.,Ltd. Address before: Cangshan District of Fuzhou City, Fujian province 350002 to build a new town, Jinshan District Patentee before: FUJIAN AGRICULTURE AND FORESTRY University |
|
| TR01 | Transfer of patent right |