CN106773698B - A kind of beverage filling control method based on on-line parameter identification - Google Patents
A kind of beverage filling control method based on on-line parameter identification Download PDFInfo
- Publication number
- CN106773698B CN106773698B CN201611236149.5A CN201611236149A CN106773698B CN 106773698 B CN106773698 B CN 106773698B CN 201611236149 A CN201611236149 A CN 201611236149A CN 106773698 B CN106773698 B CN 106773698B
- Authority
- CN
- China
- Prior art keywords
- filling
- beverage
- control
- period
- control amount
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Filling Of Jars Or Cans And Processes For Cleaning And Sealing Jars (AREA)
Abstract
The invention discloses a kind of beverage filling control methods based on on-line parameter identification, for the time-varying and non-linear behavior of Beverage filling machine, this method adjusts the servo stroke of filling servomechanism according to set filling amount and the filling data intelligence of history, achievees the purpose that improve filling accuracy.The invention patent is when carrying out pour-in controller design aiming at the problem that pouring process different phase, it designs intelligent control algorithm and carrys out the filling error of automatic straightening, in order to more reasonably set the parameter in intelligent controller, the present invention picks out the filling sensitivity of beverage filling system according to the filling online data of history, according to the penalty coefficient of the bulking system sensitivity real time modifying human simulated intelligent controller picked out, and then filling accuracy is improved, and prevent from adjusting and excessively occur shaking or adjusting the case where deficiency causes disqualification rate to increase.
Description
Technical field
The present invention relates to beverage filling production lines automatic control technology fields, are related to a kind of drink based on on-line parameter identification
Expect filling control method.
Background technique
In the food industry, Beverage filling machine is a kind of widely used automated production equipment, in beverage processing, encapsulation
Play highly important role in the process.The raising of beverage filling precision not only can reduce and generate in production process
Waste is also related to the experience of consumer and then influences the brand benefit of beverage businessman.However, improving production equipment by extensive
Level is huge again come the method cost for reaching raising filling accuracy.Therefore, it at present in industrial production, is substantially and is not changing
Become under conditions of original production equipment and improves filling accuracy.
Currently, the method for commonly improving filling accuracy mostly uses greatly opened loop control, by way of manual fine-tuning, into
The filling calibration of row, which reaches, proposes high-precision purpose.This method according to physical relation, derives beverage filling system servo first
Then relationship between control amount and material filling amount is established filling unit volume according to physical quantitys such as the relationship and beverage densities and is drunk
Expect the control amount size needed.In production, operator also rule of thumb can manually be finely tuned control amount, this is manually
Although adjusting method can play certain adjusting purpose, filling accuracy is improved, this method is to operator's skill requirement ratio
It is higher, and with the variation of filling environment, it may be necessary to continuous manual fine-tuning, which can just play, continues the filling effect of high-precision
Fruit.Since servo-system inherently has nonlinear characteristic, when there is same filling error, different servo bulking systems exists
Different moments, the degree of adjusting be also it is different, rely solely on manually adjust be also it is difficult to ensure that adjust validity.
Therefore, it is necessary to design a kind of beverage filling control method for capableing of quick, stable raising filling accuracy.
Summary of the invention
The technical problem to be solved by the present invention is to not for beverage filling control method in existing Beverage filling machine
Foot provides a kind of beverage filling control method based on on-line parameter identification, can effectively improve the filling accuracy of Beverage filling machine.
In order to solve the above technical problems, the technical scheme adopted by the invention is that: a kind of drink based on on-line parameter identification
Expect bottle placer pour-in controller design method, comprising the following steps:
1) it obtains Beverage filling machine self-starting system and starts the filling data of history, including filling servo stroke array u [1,
2 ..., n-1] and actual filling quantity deviation array e [1,2 ..., n]=set-y [1,2 ..., n], and it is averaging, it obtains
Current filling period servo stroke and filling amount are respectively as follows: after to filteringWith
Wherein set is setting filling amount, and h is to be averaging array length.
2) by above-mentioned steps 1) obtain data analyzed, then select different control mode.
3) the control mode according to selected by step 2) finds out filling control amount u (k) in current filling period
4) the filling control amount u (k) in the filling period that production line is found out according to step 3) carries out real-time control, and incite somebody to action this
In secondary control amount and filling error log to history array.
In step 1), it is averaging the given method of array length h are as follows: established standards are averaging array first on host computer
Length Nr, will practical filling frequency n and N with the increase of filling numberrBe compared, select that relatively small parameter as
The averaging array length in the control period, i.e. h=min { n, Nr}.
In step 2), the selection rule for controlling mode is as follows:
(1) if error is greater than some threshold value, i.e.,Show that error filling in the past is very big, need to be reduced rapidly error,
Using first order compensating proportion pmaxControl amount increment is calculated, using following control mode one:
Wherein, whereinFor the error upper limit of setting；pmaxMaximum compensating proportion is set for host computer；λ0Indicate filling amount with
Servo stroke scale factor, is determined by physical relation.
(2) whenController increment is determined according to the degree of error size, using second level compensating proportion
Control amount increment is calculated, using following control mode two:
WhereinE is the error floor of setting；pminMinimum compensation is set for host computer
Ratio.
(3) when error be less than some threshold value, i.e.,Filling error is in the lesser error threshold of third, due to this
When error it is very small, control amount is very close to optimal injection stroke, using third level compensating proportion pminCalculate control amount
Increment, and use following control mode three:
(4) if filling error continuous l times or error history value are both less than the allowable error range set, it is believed that be
It unites stable in present period, control amount is constant, causes system unstable to prevent from frequently changing control amount, prevents accidental mistake
Difference causes system to adjust excessively, that is, works as
max{e(1),e(2),…,e(k)}≤es,k≤l
Or
max{e(k-l+1),e(k-l+2),…,e(k)}≤es,k>l
When, using control mode four:
U (k)=u (k-1)
Wherein esFor the allowable error range of setting；L is that the stabilization of setting judges array length.
In step 2), when carrying out filling control to system, in order to prevent because beverage filling production lines non-linear factor causes
Regulating effect is undesirable, and using the filling sensitivity of principle of least square method on-line identification system, and on this basis, building is filled
Dress system locally linear model carries out on-line analysis to perfusion system sensitivity, design be able to solve because of system time-varying and
The pour-in controller of non-linear effects.Identification process is as follows:
(1) filling amount output expression formula is defined first:
V=λ u+ δ (u, v, μ, T)
Wherein δ (u, v, μ, T) indicates nonlinear function
The linear approximate relationship of filling amount and control amount in sampling instant k is indicated are as follows:
Bulking system sensitivity indicates that when controlling increment has minor change, filling amount variation exports, therefore defines filling
System sensitivity are as follows:
It (2), can be according to the filling servo stroke array obtained in step 1) and practical filling according to principle of least square method
Loading amount deviation array on-line identification goes out bulking system sensitivity, and expression formula is as follows:
Wherein, m is the identification array length of setting, V (k-1), V (k-2) ..., and V (k-m) is respectively (k-1) to (k-
M) secondary filling for actual filling quantity；Actual filling quantity measures in the following manner: after expendable, being measured by scale
Weight of drink bottle after filling out, the difference of weight of drink bottle and beverage bottle empty bottle weight that scale measures is this filling beverage
Then weight calculates this filling beverage volume according to density, as this filling actual filling quantity.
(3) result of identification is brought into control one~tri- formula of mode, replaces λ with the s (k) that real-time identification goes out0, and add
Upper adjusting coefficient gamma obtains current period control amount expression formula are as follows:
Wherein, γ indicates adjustment factor.
The utility model has the advantages that
For the present invention by analyzing the filling data of history, designing being capable of control that is quick, stablizing raising filling accuracy
Algorithm processed, has great importance and vast market prospect.
Using the present invention come when being transformed to beverage filling system, without increasing or changing the machinery of Beverage filling machine
Structure and hardware circuit, it is only necessary to add upper control algolithm proposed by the present invention in the PLC controller of beverage filling lines, then adjust
Good parameter, compared to it is general by change canning line some mechanical structure for, cost substantially reduces.In the present invention
The control mode is the filling situation after being started according to system come fixed, almost every kind of condition provide one it is very excellent
The control amount of change exports: for starting soon, historical data it is insufficient in the case of, being averaging array can fill according to all history
Data are filled, rather than standard set by host computer is averaging array length NrIt is handled；It is when continuous l times or all filling
When error is respectively less than the allowable error range set, it is believed that system is stable in present period, and control amount is constant, to prevent
Frequently changing control amount causes system unstable, overcomes disturbance and interference is caused to cause to control not normal problem；For working as beverage
The problem that bulking system causes systematic comparison sensitive because of environment or self reason, at this time smaller controlling increment also results in tune
Section excessively, easily causes concussion, and the present invention goes out system sensitivity s (k) by on-line identification, when system sensitivity is higher, control
Increment processed is decayed, and when system sensitivity is lower, controlling increment is reinforced, and mission nonlinear can be overcome to draw well
The beverage filling error risen.It is at low cost the present invention is based on the Beverage filling machine pour-in controller design method of on-line parameter identification
And control precision is high, reduces the disqualification rate of beverage filling process, practical value with higher and good application prospect,
This method also has reference function to other quantitative liquid bulking systems.
Detailed description of the invention
Fig. 1 is Beverage filling machine structure diagram.
Description of symbols:
In figure, 1 is beverage bottle, and 2 be to fill needle, and 3 be beverage filling pump, and 4 be filling servo motor, and 5 be encoder, and 6 be to watch
Driver is taken, 7 be master controller PLC, and 8 be host computer PC machine, and 9 be scale, and 10 be main belt.
Specific embodiment
Controlled device of the present invention-Beverage filling machine structure is as shown in Figure 1,1 is beverage bottle, and 2 be to fill needle, and 3 be beverage
Filling pump, 4 be filling servo motor, and 5 be encoder, and 6 be servo-driver, and 7 be master controller PLC, and 8 be host computer PC machine, 9
It is scale, 10 be main belt.Beverage filling system work process of the present invention are as follows: beverage bottle 1 is on main belt 10
Carry out translational motion；When beverage bottle, which moves to, fills 2 lower section of needle, the servo stroke that servo-driver 6 is calculated according to PLC,
Drive filling servomechanism 4 that beverage filling pump 3 is driven to complete to draw and inject the movement of beverage；Wherein encoder 5 is watched for measuring
The angle and angular speed of device are taken, it and servo-drive, servomechanism three constitute motion control closed loop；Drink after filling
Expect bottle under the action of main belt, through weighed 9 measure it is filling after weight of drink bottle, the weight of drink bottle and drink that scale measures
The difference for expecting bottle empty bottle weight is this filling beverage weight, then calculates this filling beverage volume according to density, i.e.,
For this filling actual filling quantity.Data exchange is carried out using ethernet communication mode between PLC and host computer PC machine 8,
The parameters of control system can be configured in monitor supervision platform, it can also be to the filling error and servo of Beverage filling machine
Stroke carries out real-time monitoring.
In order to make technological means of the invention, creation characteristic, workflow, application method, reach purpose and effect is easy to
Understand and understand, below with reference to specific filling process, the present invention is further explained.
1. preparation before filling starting
Before progress is filling, do not start main belt first, choose emptying mode, beverage is packed into holding vessel and to filling
Dress system is emptied；Beverage filling mode is then switched to, and sets the on-line parameter that is based on host computer and recognizes
Pour-in controller relevant parameter: setting filling amount set, standard be averaging array length Nr, error bound e and e, it is maximum most
Small compensating proportion pmaxAnd pmin, stabilization judge array length l, allowable error range es, identification array length m, adjustment factor γ
Deng；Then activation system carries out beverage and automates filling production process.
2. obtaining and analyzing filling data
After activation system, beverage filling of every completion calculates this filling result by scale measurement result and deposits
Enter in array.Obtain the filling data of history, including filling servo stroke array u [1,2 ..., n-1] and actual filling quantity deviation
Array e [1,2 ..., n]=set-y [1,2 ..., n], and it is averaging, current filling period servo row after being filtered
Journey and filling amount are respectively as follows:With
The initial several wheels of activation system are filling, since the history array length n of preservation is smaller, usually select n as average
Array length h；When the filling number of system reaches NrAfter secondary, N is selectedrAs averaging array length h；On-line parameter is distinguished
The selection for knowing data, as n < m without parameter identification (λ at this time0Value according to setting filling amount, pump parameter etc. determines, tool
Body is related with technical parameter, and such as every filling unit volume beverage needs much servo strokes)；As n >=m, with m times saved recently
Filling data carry out on-line identification beverage filling system sensitivity as Identification Data.
3. system automatically selects control mode
The parameter of filtered mean error combination host computer setting, selection control mode, mode are calculated according to step 2
Selection rule is as follows:
(1) if error is greater than some threshold value, i.e.,Show that error filling in the past is very big, need to be reduced rapidly error,
Using first order compensating proportion pmaxControl amount increment is calculated, using following control mode one:
(2) whenController increment is determined according to the degree of error size, using second level compensating proportion p
(e) control amount increment is calculated, compensating proportion basic principle is the bigger selected compensating proportion of errorIt is bigger, using as follows
Control mode two:
(3) when error be less than some threshold value, i.e.,Filling error is in the lesser error threshold of third, due to this
When error it is very small, control amount is very close to optimum servo stroke, it is only necessary to minor adjustments is carried out, using third
Grade compensating proportion pminControl amount increment is calculated, and uses following control mode three:
(4) if filling error continuous l times or error history value are both less than the allowable error range set, it is believed that be
It unites stable in present period, control amount is constant, causes system unstable to prevent from frequently changing control amount.I.e. as max { e
(1),e(2),…,e(k)}≤es, k≤l or max e (k-l+1), e (k-l+2) ..., and e (k) }≤es, when k > l, on the one hand
Achieved the purpose that adjust, on the other hand the filling error that finally obtains in this case often from system accidental error,
Cannot be using the error as the foundation for calculating controlling increment, therefore the control amount of PLC output is consistent with last round of filling control amount,
Using control mode four:
U (k)=u (k-1)
4. carrying out on-line parameter identification optimal control codes output
Proceed to certain number when filling, as n > m, considers the mission nonlinear of controlling increment and actual filling quantity output
Relationship carries out local linearization to system using least square method, the control to mode one, mode two and mode three in step 3
Amount output optimizes.
Wherein, the identification expression formula based on least square method are as follows:
Control amount output expression formula on the basis of the filling sensitivity of the system of identification, after obtaining optimization are as follows:
Then replace mode one, mode two and the calculated control amount of mode three defeated with the control amount output after the optimization
Out.
5. real-time control and feedback result
PLC calculates the filling control amount u (k) in current filling period via step 3 and step 4, by control amount servo row
Journey passes to servo-driver, and servo is driven device driving servo motor and rotated according to certain angle and angular speed, by encoder reality
When measurement servo motor speed and feed back to servo-driver complete closed-loop control guarantee servomechanism stroke accurately servo is followed to drive
Dynamic device given value.Last hydraulic pump completes to take out under the action of servo-driver material with it is filling, after this is filling, by this
Control amount and filling error save and refresh data in history array are recorded.
Claims (3)
1. a kind of beverage filling control method based on on-line parameter identification, which comprises the following steps:
1) the filling array of data of history of Beverage filling machine system, including history filling period filling servo stroke are obtained first,
The i.e. filling period filling control amount array u of history [1,2 ..., n-1] and actual filling quantity deviation array e [1,2 ..., n]=
Set-y [1,2 ..., n], and filtering is averaging to it, obtain the filling period filling control amount of filtered history and filling amount
Deviation is respectively as follows:With
Wherein, set is setting filling amount, and h is to be averaging array length, and n is the history array length saved, and k is present sample
Moment；
2) by above-mentioned steps 1) the obtained filling period filling amount deviation of filtered historyDetermine that this filling period fills
The increment Delta u (k) of control amount is filled, then by the filling period filling control amount of filtered historyIt is filled with this filling period
Increment Delta u (k) summation of dress control amount obtains the filling control amount u (k) in this filling period；
3) the filling control amount u (k) in this filling period found out according to step 2) exports to adjust servomechanism stroke, carries out real
When control, the filling control amount in this period and filling amount deviation are saved after this is filling, more new historical fill
Fill array of data；
In step 2),
Wherein,To control penalty coefficient, λ0It indicates filling amount and servo stroke scale factor, is determined by physical relation；
Control penalty coefficientDetermination method are as follows: whenWhen,WhenWhen,
WhenWhen,
Wherein,It is respectively the error upper and lower bound set with e；pmaxAnd pminRespectively set minimum and maximum compensating proportion；
As n > m, on-line parameter identification is carried out by principle of least square method first, obtains Beverage filling machine system sensitivity s
(k):
Wherein, m is the identification array length of setting, and V (k-1), V (k-2) ..., V (k-m) is respectively (k-1) secondary to (k-m)
Filling actual filling quantity；
Then, the s (k) real-time identification gone out is substituted into the calculation formula of Δ u (k), replaces the λ in calculation formula0, and tune is added
Coefficient gamma is saved, the increment Delta u (k) of this filling period filling control amount is obtained are as follows:
2. the beverage filling control method according to claim 1 based on on-line parameter identification, which is characterized in that the step
It is rapid 1) in, in order to meet each stage control demand, the given method of the array length h of averaging are as follows:
The averaging array length N of established standards firstr；With the increase of filling number, by the practical filling number of Beverage filling machine
N and NrCompare, selects averaging array length of that the relatively small parameter as the control period, i.e. h=min { n, Nr}。
3. the beverage filling control method based on on-line parameter identification described according to claim 1~any one of 2, feature
Be, in step 2) when continuous l times or error history value are both less than or when error filling equal to the permission of setting, filling control
Amount processed remains unchanged, it may be assumed that
As max { e (k-l+1), e (k-l+2) ..., e (k) }≤es, k > l or max e (1), e (2) ..., and e (k) }≤es,k≤l
When, the filling control amount in this filling period remains unchanged, i.e. u (k)=u (k-1)；
Wherein, esTo allow filling error, l is that stabilization judges array length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611236149.5A CN106773698B (en) | 2016-12-28 | 2016-12-28 | A kind of beverage filling control method based on on-line parameter identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611236149.5A CN106773698B (en) | 2016-12-28 | 2016-12-28 | A kind of beverage filling control method based on on-line parameter identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106773698A CN106773698A (en) | 2017-05-31 |
| CN106773698B true CN106773698B (en) | 2019-05-10 |
Family
ID=58924775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611236149.5A Expired - Fee Related CN106773698B (en) | 2016-12-28 | 2016-12-28 | A kind of beverage filling control method based on on-line parameter identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106773698B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110980613A (en) * | 2019-11-22 | 2020-04-10 | 湖南工业大学 | High-speed canning self-adaptive control system and method |
| CN117766956A (en) * | 2024-02-22 | 2024-03-26 | 宁德时代新能源科技股份有限公司 | Liquid injection system and liquid injection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100391793C (en) * | 2006-12-14 | 2008-06-04 | 北京航空航天大学 | Servo control system of magnetically suspended control moment gyroscope frame with precise friction compensation |
| CN101161589A (en) * | 2007-01-18 | 2008-04-16 | 李奕 | A method dor detecting and controlling liquid loading degree |
-
2016
- 2016-12-28 CN CN201611236149.5A patent/CN106773698B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN106773698A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8875960B2 (en) | Tilting-type automatic molten metal pouring method, tilting control system, and storage medium having tilting control program stored therein | |
| CN106773698B (en) | A kind of beverage filling control method based on on-line parameter identification | |
| CN102556629B (en) | Intelligent automatic dosing control system | |
| CN103640194B (en) | A kind of Intelligent injection machine and injecting method thereof | |
| CN105539932B (en) | A kind of method of adaptive correction weighing system error | |
| JP2003526856A (en) | Process control system | |
| CN109592095A (en) | A kind of filling accuracy control method and equipment | |
| CN104552610B (en) | The ingredient control method and system of concrete mixing plant | |
| CN104166405B (en) | Liquid level system PI control method based on virtual unmodeled dynamics compensation | |
| CN108333943B (en) | Crude oil blending online optimization method based on incremental mode | |
| CN106773652A (en) | A kind of PID systems and its parameter automatic adjusting method | |
| CN105415091A (en) | High-precision control method for overlap amount of disc scissors | |
| CN117270483A (en) | Full-flow dynamic optimization control method and device for chemical production device and electronic equipment | |
| JP4744978B2 (en) | Heavy-weight filling device | |
| CN113485470A (en) | Variable spray control method, device and system | |
| CN106694863A (en) | Control method and device of molten steel pouring | |
| CN116009398A (en) | Method and related device for segmenting gasoline and diesel oil mixing section | |
| CN112666827B (en) | Method for controlling liquid distribution in grading manner based on PID | |
| CN109870986B (en) | On-line control method of stirred tank reactor based on neural network and data drive | |
| CN103234351B (en) | Control method for carbon content of sintered material | |
| EP3320939A1 (en) | Closed loop capacity control method for respirator | |
| CN109062059B (en) | Batch process prediction control method based on system augmentation model | |
| CN110980613A (en) | High-speed canning self-adaptive control system and method | |
| CN108453242B (en) | Self-learning-based low-pressure casting pressurization control method and device | |
| CN114653538B (en) | Self-learning closed-loop dispensing system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190510 Termination date: 20201228 |