CN207227388U - White wine based on expert system blends equipment automatically - Google Patents
White wine based on expert system blends equipment automatically Download PDFInfo
- Publication number
- CN207227388U CN207227388U CN201720616420.1U CN201720616420U CN207227388U CN 207227388 U CN207227388 U CN 207227388U CN 201720616420 U CN201720616420 U CN 201720616420U CN 207227388 U CN207227388 U CN 207227388U
- Authority
- CN
- China
- Prior art keywords
- tank
- liquor
- base liquor
- expert system
- blent
- 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.)
- Expired - Fee Related
Links
Landscapes
- Distillation Of Fermentation Liquor, Processing Of Alcohols, Vinegar And Beer (AREA)
Abstract
It the utility model is related to the white wine based on expert system and blend equipment automatically, white wine characteristic odor peak value measurement can be carried out, choose suitable base liquor automatically according to the white wine specification to be blent, carry out the automatic control for blending and completing related executing agency of feature based smell.White wine based on expert system blends equipment automatically to be included：Frequency converter, base liquor tank, pump, blend tank, industrial personal computer, capture card, electronic nose, channel switching device, fluid level transmitter.The utility model can be in production process be blent on a large scale, traditional taster is substituted to be detected with expert system, the work of decision-making, so as to fulfill continuity, the automation for blending production process, the influence of taster, base liquor quality factor to blended liquor is reduced at the same time, ensure that the uniformity of product quality.
Description
Technical field
It the utility model is related to the automatic of white wine and blend equipment, more particularly to the white wine based on expert system is blent automatically sets
It is standby.
Background technology
Liquor-making enterprises are according to technique and store sth. in a cellar the difference of time, and it is different that each white wine is divided into style, quality and characteristic
Multiple specifications.Although the white wine of each specification each have their own advantage in terms of mouthfeel, flavor, more or less there are mouth
The shortcomings of sense is not plentiful, flavor is boring, entrance is not soft, sour and astringent miscellaneous taste.Therefore, the white wine that stores sth. in a cellar of each specification usually will not be direct
Filling sale, and be used as base liquor and blent by rational proportion, make perfect wine body, style protrusion, sense organ by learning from other's strong points to offset one's weaknesses
Quality balances, and is finally reached the purpose for improving white wine quality.
However, presently used tradition is blent in equipment, it is necessary to according to the sensory evaluation of taster, it is suitable subjectively to choose
The base liquor of conjunction, and set each base liquor blends ratio.After the factors such as batch, raw material, time, weather, pit change, together
The quality of one specification base liquor also can accordingly change, therefore in order to keep the uniformity of blended liquor quality, it is necessary to which taster exists
Constantly evaluated in production throughout the year, select base liquor, and the ratio of base liquor is adjusted with the quality fluctuation of base liquor.But due to
Subjectively sensory evaluation standard has differences each taster, and the subjective factor meeting such as the mood of taster, sensory evaluation ability
Fluctuated with change of external conditions, can cause the quality of different batches blended liquor that slight change occurs.In order to solve this
One problem, need to develop the automatic of the white wine based on expert system and blend equipment, be set to replace blending based on artificial tradition
It is standby.
The content of the invention
The applicant is directed to the above-mentioned limitation of traditional blending method, is studied and is explored, there is provided one kind is based on expert
The white wine of system blends equipment automatically mainly to be included：Base liquor tank, pump, frequency converter, blend tank, electronic nose, capture card, industrial personal computer,
Fluid level transmitter etc.；
Industrial personal computer, into row data communication, realizes the control of frequency converter by capture card and frequency converter, electronic nose, fluid level transmitter
System and the collection of e-nose signal, liquid level signal；
The equipment that white wine based on expert system is blent automatically includes multiple base liquor tanks, and different bases is housed in base liquor tank
Wine, base liquor tank are connected by pipeline and pump with blending tank, and base liquor is blent tank by pumping injection by base liquor tank, blend pot bottom dress
There is agitating device guarantee to be sufficiently mixed；
Blend pot bottom and liquid level sensor is housed, the liquid level of blended liquor in tank is blent for measuring；Liquid level sensor
Data-signal is sent into industrial personal computer by capture card, and industrial personal computer can blend tank internal diameter according to the combination of this signal and be hooked to calculate to blend in tank
Convert the volume of wine；
The power that frequency converter is pumped by data line traffic control, can adjust the flow of pump in the range of 0~100%；
It is preinstalled with industrial personal computer and blends expert system automatically.Automatically expert system is blent to be chosen automatically by database, base liquor
Algorithm, blend Algorithm constitution automatically；Wherein prestored in database：The characteristic odor peak value of optional blended liquor, each blended liquor pair
Multiple base liquors for answering and its initial flow, the characteristic odor peak value of each base liquor.
Industrial personal computer can obtain this and blend corresponding base liquor and its characteristic peak according to target blended liquor from pre-stored data storehouse
Value, each base liquor tank characteristic odor peak value of electronic nose collection is obtained by capture card, is passed through the automatic Algorithms of Selecting of base liquor and is calculated institute
There is the characteristic odor peak error of base liquor and the base liquor that prestores in base liquor tank, and choose in the base liquor tank with the base liquor error minimum that prestores
The base liquor that base liquor is blent as this；
The power that industrial personal computer is pumped according to the initial flow blent automatically in Data In Expert System storehouse by Frequency Converter Control
The beginning first round blends；By blending algorithm automatically, using the characteristic odor peak value of target blended liquor as target peak, based on linear
The next knots modification blent round and blend base liquor flow of solving equations, the power pumped by Frequency Converter Control change the hook of base liquor
Ratio is converted to complete next to blend round.After being spaced at every fixed time, above-mentioned work is repeated, each wheel is completed and blends；Liquid level becomes
Send that device gathers blend in tank blended liquor liquid level reach needed for the process of blending terminates automatically when blending capacity for liquor.
The automatic Algorithms of Selecting of base liquor that industrial personal computer prestores in the utility model is to reach following purpose：Selection target is blent
After drinking, industrial personal computer can obtain the blended liquor corresponding prestore base liquor and initial flow (by taking 3 kinds of base liquors as an example) from database, electricity
Sub- nose detects the characteristic odor peak value of base liquor in all base liquor tanks, the automatic Algorithms of Selecting of base liquor calculate in all base liquor tanks base liquor with
The base liquor that prestores corresponds to the error of target blended liquor characteristic odor peak value, chooses error minimum base liquor and is blent for this.
Using blend expert system prestore this blend specification correspond to base liquor quantity as 3, this blend corresponding blended liquor feature
Exemplified by smell peak-to-peak value quantity is 3, base liquor and the error of the characteristic odor peak value for the base liquor that prestores are in all base liquor tanks：
Wherein：Askj(k=1,2,3, j=1,2,3) be that kth kind prestores j-th of feature gas of base liquor in expert system database
Taste peak value；
Abij(i=1,2,3 ..., j=1,2,3) be j-th of characteristic odor peak value in i-th base liquor tank；
Eik(i=1,2,3 ..., k=1,2,3) be the jth kind base liquor feature gas that prestores in i-th of base liquor tank base liquor and database
Error between taste peak value；
Calculation error Eik, choose Ei1Base liquor is used to hook in corresponding i-th of base liquor tank of minimum value in (i=1,2,3 ...)
Convert, the initial flow of the base liquor is set to the initial flow of the 1st kind of base liquor that prestores；And so on, complete all base liquor correspondences that prestore
Base liquor and initial flow determine.
What industrial personal computer prestored in the utility model blends algorithm automatically, each round base liquor flow during being blent for calculating
Knots modification, to adjust the ratio of blending, make to blend blended liquor and target in tank blend smells of wine peak value as close possible to.
3 kinds of base liquors (a, b, c), target blended liquor characteristic odor peak A are corresponded to target blended liquori(i=1,2,3) exemplified by：
When containing certain certain density class solute in blended liquor according to Raoult's law, in blended liquor tank, ullage should
The vapor partial pressure power of solute, the saturated vapour pressure equal to the solute pure material under Current Temperatures are multiplied by mole of solute in solution
Concentration, i.e.,：
Wherein, piFor the partial pressure of the i-th class solute vapor above liquid level of solution,It is pure for the i-th class solute under Current Temperatures
The saturated vapour pressure of material, ciFor the molar concentration of the i-th class solute in solution.
According to Dalton's law (of partial pressures), the vapor partial pressure power of a kind of material of certain in the mixed gas above liquid level of solution and its
The amount of material is directly proportional, i.e.,：
Wherein, V be liquid level of solution superjacent air space volume, niFor the amount of the material of the i-th class solute vapor in space, R is
Ideal gas constant, T are environment temperature.
Therefore：
Wherein, k1=RT, Ci=ni/ V is the molar concentration of the i-th class solute vapor.
The smell peak value extracted above liquid level of solution that electronic nose detects in the smell collection of illustrative plates of mixed gas with it is corresponding molten
The molar concentration of matter steam is directly proportional, i.e.,：
Ai=k2Ci (4)
Wherein, Ai：The smell peak value (i.e. ith feature smell peak value) for the i-th class solute vapor that electronic nose measures；
k2：The ratio between electronic nose characteristic constant, characteristic feature smell peak value and molar concentration of corresponding solute vapor, its numerical value
It is unrelated with solute species, determined by electronic nose characteristic.
Therefore：
Wherein, Ki：With the relevant coefficient of solute.
According to (5) linear relationship, if the gas of the i-th class flavor substance (i.e. the i-th class solute) in base liquor a and base liquor b
Taste peak value is respectively Aai、Abi, by La、LbTwo flows are blent, then the molar concentration of the i-th class flavor substance is in blended liquor：
It can be expressed as in the form of smell peak value：
LaAai+LbAbi=(La+Lb)Ani (9)
LaΔt·Aai+Δt·LbAbi=Δ t (La+Lb)Ani (10)
Wherein, cai、cbiThe molar concentration of the i-th class flavor substance respectively in base liquor a, b, AniFor the i-th class in blended liquor
The smell peak value of flavor substance, Δ t is each blends the Fixed Time Interval that round pumping takes base liquor to be undergone.
According to (10), if it is a certain blend round after blend existing blended liquor volume in tank and be multiplied by its some characteristic odor
Peak value, plus it is next blend round and newly inject the volume of each base liquor for blending tank be multiplied by respective features described above smell peak value, i.e.,
Equal to it is next blend round after blend blended liquor volume in tank and be multiplied by its features described above smell peak value；Therefore：
Wherein, Aai、Abi、Aci(i=1,2,3) are 3 characteristic odor peak values of a, b, c base liquor；
La、Lb、LcFor current time a, b, c base liquor flow；
ΔLa、ΔLb、ΔLcFor the flow knots modification of next a, b, c base liquor for blending round；
After V is currently blends round, the total amount of white wine in tank is blent；
An1、An2、An3After currently blending round, 3 characteristic odors for blending blended liquor in tank of electronic nose measurement
Peak value；
At1、At2、At3Target peak for the 3 characteristic odor peaks of target blended liquor to prestore in database；
Solve above-mentioned system of linear equations and can obtain Δ La、ΔLb、ΔLc, i.e., next flow knots modification for blending round base liquor.
White wine based on expert system blends equipment automatically, and application method step is as follows：
1st step：The white wine specification to be blent is selected to input objective body accumulated amount as target blended liquor in industrial personal computer；
2nd step：Industrial personal computer control passage switching device, electronic nose gather the characteristic odor peak of base liquor in each base liquor tank successively
Value Data, is sent into industrial personal computer；
3rd step：Industrial personal computer determines characteristic odor peak value according to the automatic Algorithms of Selecting of the base liquor blent automatically in expert system
The base liquor that the base liquor of error minimum is blent as this；
4th step：Industrial personal computer blends base liquor according to selected this and initial flow controls corresponding frequency converter to adjust pump work
Rate, completes the first round and blends；
5th step：After Fixed Time Interval, electronic nose gathers this successively to be blent in base liquor tank base liquor and blends in tank and hook
Convert the characteristic odor peak-data of wine, fluid level transmitter collection blends blended liquor liquid level data in tank and is sent into industrial personal computer, industrial personal computer
Blend algorithm automatically according to expert system and calculate this and blend base liquor flow knots modification, control this to blend the corresponding change of base liquor tank
Frequency device adjusts pump power, completes next round and blends；
After being spaced at every fixed time, above-mentioned work is repeated, each wheel is completed and blends；
The process of blending terminates automatically when what fluid level transmitter gathered blends blended liquor liquid level arrival objective body accumulated amount in tank.
The utility model has technical effect that：
The utility model is established blends equipment automatically based on expert system, is set to replace being blent by the tradition manually participated in
It is standby.
Liquor-making enterprises are usually blent in proportion using storing sth. in a cellar white wine as base liquor, and learning from other's strong points to offset one's weaknesses makes the perfection of wine body, style
Prominent, organoleptic quality balance, to lift white wine quality.Base liquor in base liquor tank batch, raw material, time, weather, pit etc. because
After element changes, base liquor quality can also change.Equipment is blent during automatic blend based on expert system automatically,
With the automatic Algorithms of Selecting of base liquor select and prestore from all base liquors the error minimum of the characteristic odor peak value of base liquor base liquor make
For the base liquor that this is blent, ensure that the base liquor quality of selection is most suitable for this and blends；Equipment is blent based on expert system automatically
In engineering is blent automatically, flow is blent to blend algorithm automatic adjustment base liquor automatically, makes blended liquor convergence target blended liquor product
Matter.Automatic based on expert system blend equipment can during blending on a large scale with expert system substitute traditional taster into
Row detection, decision-making, so as to realize the continuity, automatic for blending production process under the continually changing objective fact of base liquor quality
Change, while reduce taster, base liquor quality factor to blending the influence of quality, ensure that the uniformity of liquor production quality.
Brief description of the drawings
Fig. 1 is that the white wine based on expert system blends equipment schematic diagram automatically.
In Fig. 1：1 frequency converter；2 base liquor tanks；3. pump；4 blend tank；5 industrial personal computers；6 capture cards；7 electronic noses；8 passages switch
Device；9 fluid level transmitters；Double solid line represents the pipeline that base liquor flows through；Dotted line represents collection olfactory signal and sends control signal
Data cable；Slash filling line is the gas circuit for electronic nose detection.
Fig. 2 is the flow chart that the white wine based on expert system blends equipment automatically.
Fig. 3 is the white wine smell collection of illustrative plates that electronic nose gathers in example, wherein during corresponding sampling time 1.8s, 2.2s, 3.6s
The peak gone out, that is, indicate three peaks of " 1 ", " 2 ", " 3 ", to be tested three characteristic odor peaks of white wine；What is prestored in database can
Select blended liquor and the corresponding multiple base liquors of each blended liquor characteristic odor peak value, it is automatic blend during online acquisition base liquor and
Blended liquor characteristic odor peak value, is the peak value above three characteristic odor peak；
Fig. 4 is that 1 characteristic odor peak value relative error of peak is blent in tank between blended liquor and target blended liquor in example with hook
Convert the variation diagram of round.
Embodiment
Specific embodiment of the present utility model is described further below in conjunction with the accompanying drawings.
As shown in Figure 1, the equipment of the automatic blending method based on expert system of this example includes：1 frequency converter；2 base liquors
Tank；3. pump；4 blend tank；5 industrial personal computers；6 capture cards；7 electronic noses；8 channel switching devices；9 fluid level transmitters.
It is preinstalled with industrial personal computer 5 and blends expert system automatically；Industrial personal computer 5 passes through capture card 6 and electronic nose 7, multiple frequency conversions
Device 1, fluid level transmitter 9 connect, into row data communication；Each base liquor tank 2 is connected by pump 3 and pipeline with blending tank 4, can be by base
Tank 2 is blent in wine injection；Frequency converter 1 adjusts 3 power of pump to adjust base liquor flow.
The automatic application method for blending equipment based on expert system of the present embodiment, key step：
1st step：The white wine specification to be blent is selected to input objective body accumulated amount as target blended liquor in industrial personal computer；
2nd step：Industrial personal computer 5 is gathered in each base liquor tank 2 successively by 6 control passage switching device 8 of capture card, electronic nose 7
The characteristic odor peak-data of base liquor, is sent into industrial personal computer 5；
3rd step：Industrial personal computer 5 determines characteristic odor peak value according to the automatic Algorithms of Selecting of the base liquor blent automatically in expert system
The base liquor that the base liquor of error minimum is blent as this；
4th step：Industrial personal computer 5 blends each base liquor according to selected this and initial flow controls corresponding frequency converter 1 to adjust
Pump power, completes the first round and blends；
5th step：After Fixed Time Interval, electronic nose 7 gathers this and blends base liquor in each base liquor tank 2 and blend tank successively
The characteristic odor peak-data of blended liquor, the collection of fluid level transmitter 9 blend blended liquor liquid level data in tank 4 and are sent into industrial personal computer in 4
5, industrial personal computer 5 blends algorithm according to expert system and calculates this automatically blends base liquor flow knots modification, controls this to blend each base
The corresponding frequency converter 1 of wine tank 2 adjusts pump power, starts next round and blends；After being spaced at every fixed time, above-mentioned work is repeated
Make, complete each wheel and blend；
What fluid level transmitter 9 gathered blends blended liquor liquid level in tank, and the process of blending terminates automatically when reaching objective body accumulated amount.
The present embodiment white wine blends flow chart such as Fig. 2 of equipment automatically.
Electronic nose collection odor characteristics collection of illustrative plates such as Fig. 3 in the present embodiment.
1 characteristic odor peak value relative error of peak is blent in tank between blended liquor and target blended liquor in the present embodiment with blending
The change of round such as Fig. 4.
It is special in Solving Linear, characteristic peaks application condition, electronic nose smell collection of illustrative plates involved in the utility model
Collection, the transmission of control signal for levying smell peak value are the prior art, its detailed process is not described further.
Example provided above is the better embodiment of the utility model, is only used for convenient explanation the utility model, and
The non-limitation made to the utility model in any form, any those of ordinary skill in the art, if not departing from
In the range of the utility model carries technical characteristic, make part using the utility model disclosed technology content and change or repair
The equivalent example of decorations, and without departing from the technical characteristics of the utility model content, still fall within the utility model technical characteristic
In the range of.
Claims (1)
1. a kind of white wine based on expert system blends equipment automatically, it is characterised in that：White wine based on expert system hooks automatically
The equipment converted mainly includes：Frequency converter, base liquor tank, pump, blend tank, industrial personal computer, capture card, electronic nose, channel switching device, liquid
Position transmitter；
Industrial personal computer, into row data communication, realizes e-nose signal, liquid by capture card and frequency converter, electronic nose, fluid level transmitter
The position collection of signal and the control of frequency converter；
The equipment that white wine based on expert system is blent automatically includes multiple base liquor tanks, and different base liquors, base are housed in base liquor tank
Wine tank is connected with blending tank by pipeline and pump, and base liquor is blent tank by pumping injection by base liquor tank, is blent pot bottom and is equipped with and stirs
Device guarantee is mixed to be sufficiently mixed；
Blend pot bottom and liquid level sensor is housed, the liquid level of blended liquor in tank is blent for measuring；
Liquid level sensor data-signal by capture card be sent into industrial personal computer, industrial personal computer can according to this signal combine blend tank internal diameter come
Calculate the volume for blending blended liquor in tank；
Frequency converter can control the power of pump to realize the flow in the range of 0~100%；
It is preinstalled with industrial personal computer and blends expert system automatically.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201720616420.1U CN207227388U (en) | 2017-05-27 | 2017-05-27 | White wine based on expert system blends equipment automatically |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201720616420.1U CN207227388U (en) | 2017-05-27 | 2017-05-27 | White wine based on expert system blends equipment automatically |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN207227388U true CN207227388U (en) | 2018-04-13 |
Family
ID=61839815
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201720616420.1U Expired - Fee Related CN207227388U (en) | 2017-05-27 | 2017-05-27 | White wine based on expert system blends equipment automatically |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN207227388U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109444359A (en) * | 2018-10-22 | 2019-03-08 | 戴建设 | A kind of white spirit process is with blending detection device |
-
2017
- 2017-05-27 CN CN201720616420.1U patent/CN207227388U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109444359A (en) * | 2018-10-22 | 2019-03-08 | 戴建设 | A kind of white spirit process is with blending detection device |
| CN109444359B (en) * | 2018-10-22 | 2021-04-27 | 山东百脉泉酒业股份有限公司 | Blending detection device for white spirit process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107090392A (en) | A kind of vodka automatic intelligent blends system | |
| CN107177458A (en) | A kind of automatic intelligent Production of Medicinal Liqueur method | |
| WO2018218868A1 (en) | Method for smart fully-automatic rum blending | |
| CN110540918B (en) | Artificial intelligence-based liquor blending control method and system | |
| CN104450463A (en) | Automatic white wine blending system based on odour control | |
| Shylaja et al. | Real-time monitoring of soil nutrient analysis using WSN | |
| CN106487875A (en) | Culinary art parameter generation method for intelligent cooking terminal and device | |
| CN101374443A (en) | Method and apparatus for pump control | |
| CN101949747B (en) | Calibration measuring system and measuring method of dynamic response characteristic of humidity sensing element | |
| CN207227388U (en) | White wine based on expert system blends equipment automatically | |
| RU2012126089A (en) | AUTOMATED DRINKS | |
| CN107189915A (en) | The automatic blending method of white wine based on expert system | |
| CN109507872A (en) | A kind of novel automatic disturbance rejection controller of model insertion formula | |
| CN109299830A (en) | A kind of hydrologic forecast system and method based on Soil Moisture Monitoring | |
| JP4600884B2 (en) | Fruit sugar content control system and fruit sugar content control method | |
| CN201983839U (en) | Combined-type standard weighing container used for liquid flow standard calibrating device | |
| CN107216978A (en) | Brandy blending method based on Intelligent Control Strategy | |
| CN107216967A (en) | A kind of intelligent control Full-automatic red wine blending method | |
| CN107189916A (en) | A kind of full-automatic yellow rice wine blending method based on on-line continuous observation and control technology | |
| CN107245418A (en) | Gin automatic manufacturing method based on Intelligent Control Strategy | |
| CN107523465A (en) | Whiskey automatic intelligent production method | |
| CN105699243B (en) | Internet of Things water-fertilizer integral lysimeter measuring system | |
| CN104804961A (en) | Automatic blending system based on fusion detection of electronic nose and near infrared spectrum | |
| CN104833779B (en) | A kind of high-quality pickles bittern detection method | |
| CN105004407B (en) | A kind of Instantaneous Fuel Consumption Monitor calibration system and its scaling method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180413 Termination date: 20190527 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |