CN105316613B - A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology - Google Patents
A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology Download PDFInfo
- Publication number
- CN105316613B CN105316613B CN201510664300.4A CN201510664300A CN105316613B CN 105316613 B CN105316613 B CN 105316613B CN 201510664300 A CN201510664300 A CN 201510664300A CN 105316613 B CN105316613 B CN 105316613B
- Authority
- CN
- China
- Prior art keywords
- coating thickness
- zinc coating
- value
- mrow
- lab
- 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.)
- Active
Links
Abstract
The invention provides a kind of zinc coating thickness control method and system based on time_varying delay offset correction technology.It is modeled prediction producing line speed, air pressure, knife to thickness of coating using Data Modeling Method first and feedback control is carried out to eliminate influence of the external disturbance to thickness of coating to controlled quentity controlled variable according to model predication value away from influence of the factor to thickness of coating such as, height；Simultaneously, according to the characteristics of zinc-plated production process, the accurate lag time for calculating calibrator measurement, based on the lag time, using the thickness measuring apparatus measuring value that matches each other and the current model predication value of model predication value Information revision, the feedback problem of system time-varying large time delay metrical information is effectively overcome；In addition, introducing discrete inspection laboratory analysis of data and the difference of thickness measurement online instrument thickness measurements is modified to the setting value of zinc coat thickness control target, it is poor more than the stable state of closed-loop control to reduce.Realizing reduces thickness of coating fluctuation, improves control accuracy and qualification rate, the target for reducing excess zinc consumption.
Description
Technical field
The present invention relates to cold-rolling galvanization automatic process control system field, in particular it relates to which a kind of be based on time_varying delay deviation school
The zinc coating thickness control method and system of positive technology.
Background technology
Galvanizing production is Chinese society because its good corrosion resistance is widely used in building, household electrical appliances, automobile and other industries
Production and the widely used steel products of consumer field.And an important technical indicator for weighing galvanizing production quality is exactly to plate
Thickness degree, product thickness of coating have to comply with product specification, to ensure the etch resistant properties of product；Simultaneously as galvanizing process is deposited
Time-varying large time delay, model accuracy is not high, environmental disturbances are big the features such as, the thickness of coating of galvanizing production fluctuates larger, steel
Enterprise often ensures that galvanizing production is up-to-standard using the method for improving control allowance, causes the luxus consumption of zinc liquid, increases
Production cost is added, and uneven coating is even that product surface flatness can be caused to be deteriorated, and influences the performances such as the spot weldability of product.
Therefore, there is an urgent need to the control technology and method for proposing to be adapted to galvanizing process production, reduce the fluctuation of product thickness of coating, reduce
The luxus consumption of zinc liquid, improve product quality level.
In zinc coat thickness control system, line speed, air knife knife are influence thickness of coating three away from, air pressure
Principal element.In actual production, realize that thickness of coating automatically controls and has very big difficulty, be primarily due to zinc-plated production system
System has following three features：
1) there is sizable measurement variable delay in zinc-plated production system.The output variable of this control system is plating thickness
Degree is, it is necessary to which calibrator measurement obtains.Because technological design and producing line such as plan at the reason, calibrator installation site apart from air knife compared with
Far, the application controlled quentity controlled variable at air knife is caused to detect that thickness of coating response change shows in the presence of very big measurement delay to calibrator
As.Meanwhile in the zinc-plated production of reality, the speed of production line changes at any time, when causing the time of the measurement delay to be
Become.The time-varying large time delay characteristic of zinc-plated production adds the difficulty of zinc-plated production thickness control, control system can be caused to produce
Larger overshoot and longer regulating time.
2) zinc-plated production process model is difficult the input/output relation for real-time and accurately reacting real system.First, it is zinc-plated
It is related to the complicated physical processes such as aerodynamics, laminar flow theory in production process mechanism, is mutually coupled between input variable, it is defeated
Enter and very strong non-linear relation between output variable be present, at present either using modelling by mechanism or black-box modeling method all very
Hardly possible is established and accurately produces model completely.Meanwhile a variety of random perturbation factors be present in zinc-plated production process, as line speed,
The change of zinc liquid composition and temperature, annealing furnace temperature change, plate shape change, fluctuation of air knife air pressure etc. all can be to the thickness of coating
Having an impact, the fluctuation of these operating modes can cause the change of production system characteristic, original model is produced mismatch with real system,
Influence the effect of control.
3) there is the deviation of change between thickness measuring apparatus measuring value and inspection laboratory values.Calibrator belongs to on-line measurement instrument,
Because environment and the difference of Theoretical Design condition are safeguarded in installation, cause to deposit between calibrator measurement data and inspection laboratory analysis of data
In larger deviation, the qualification rate of product is cannot be guaranteed using calibrator data progress closed loop feedback control, in actual production
Enterprise often uses the higher laboratory analysis method of confidence level to carry out sampling analysis to product to ensure that product can meet rule
Lattice index, but because laboratory analyze data is discrete and hysteresis is larger, it is impossible to directly as the feedback information of closed-loop system.
In three principal elements for influenceing thickness of coating, line speed is influenceed by upstream and downstream factor, can not independently be adjusted
Section, therefore most of zinc-plated control systems are handled line speed as feedforward disturbance, meanwhile, for simplify control
The difficulty of system design, keep constant as far as possible using air pressure as major control variable, air knife distance.At present, by above-mentioned
The restriction of three difficult points, cause the closed-loop control effect of Zinc Coating Thickness system not ideal enough, domestic most enterprises are still with artificial
Based on control device, due to the limitation that operative employee is controlled based on experience, thickness of coating fluctuation is big, and control effect is poor.For upper
Problem is stated, the country also has some researchers and mechanism to propose solution, such as the patent that publication No. is CN103205665A《One
Kind continuous hot galvanizing line zinc coating thickness autocontrol method》, the invention is using BP neural network structure adaptive controller and prolongs
Estimating strategy realizes coating closed-loop control late, wherein, to overcome the large time delay problem of system, the invention uses Smith predictor,
And control object is replaced using one order inertia model, solve the problems, such as time-varying large time delay to a certain extent.However, this method is adopted
Model output is compensated with fixed time lag, is only applicable to the situation of producing line constant airspeed, in the production process of reality, by
Influenceed in by upstream and downstream process production capacity, producing line speed is often changing and uncontrollable；Meanwhile using Smith predictor
Required precision to model is very high, and preferable control effect can be obtained when model is more accurate, and works as model error
When larger, the quality of Predictive Compensation Control can degenerate, or even loss of stability, and therefore, the system is for production system operating mode
It is poor to change adaptive faculty.And in document《The closed-loop control brief introduction of hot galvanizing line zinc coating thickness》(bang dragon, Sichuan is metallurgical, and 2006:
28 (6)) in, zinc coat thickness control system that introduction of authors VAI companies provide, controlling increment is by a delay in the system
Exported after block, solve the problems, such as time-varying large time delay to a certain extent.However, the system uses Sensitivity Method (local linear
Change model) controlled quentity controlled variable is calculated, there is certain deviation in model and real system, and can not equally carry out the online school of model in itself
Just, therefore, change adaptive faculty of the system for production system operating mode is also poor.Further, since installation safeguards that environment is set with theory
Difference between meter condition, cause calibrator measurement data not accurate enough, the above method lacks effective solution.To sum up,
At present although domestic existing thickness of coating closed-loop control system introduces the large time delay characteristic that certain mechanism overcomes system, but
By model accuracy, random perturbation, working conditions change, calibrator measured deviation be larger etc., factor is influenceed these methods, control effect
It is unsatisfactory.
The content of the invention
For in the prior art the defects of, realize reduce thickness of coating fluctuation, improve control accuracy and qualification rate, reduced
The target of zinc consumption is measured, it is an object of the invention to provide a kind of zinc coating thickness control method based on time_varying delay offset correction technology
And system.
According to a kind of zinc coating thickness control method based on time_varying delay offset correction technology provided by the invention, in control week
In phase t, including：Workflow management, which is calculated, by laboratory offset correction amount obtains current laboratory offset correction amount Δ CWLab(t),
Corrected zinc coating thickness predictor calculation workflow management corrected after zinc coating thickness predicted value CWp ' (t), according to institute
State current laboratory offset correction amount Δ CWLab(t) zinc coating thickness predicted value CWp ' (t) and zinc coating thickness setting value after, correcting
Rcw (t), generate current air pressure setting value Pfbc(t) as the feedback control amount in current control period；
The laboratory offset correction amount calculation process includes：
A1, judge whether there is new laboratory detected value CW in databaseLab(t),
Have, then into step A2,
Nothing, then current laboratory offset correction amount Δ CWLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) to be previous
The laboratory offset correction amount of individual controlling cycle；
A2, by the new laboratory detected value CWLab(t) and synchronization zinc coating thickness thickness measuring apparatus measuring value CWm
(t) it is saved into together in database as a record, the K bars note preserved in the first preset time is taken out from the database
Record, calculate the current laboratory offset correction amount Δ CWLab(t)；
The zinc coating thickness predictor calculation flow of the correction includes：
B1, according to default production process model and current air pressure measured value P (t), air knife knife away from measured value D
(t), line speed measured value S (t) calculates the first model prediction zinc coating thickness value CWp (t),
According in the second preset time preserved in current manufacturing lines velocity measurement S (t) and the database several
The line speed value of controlling cycle, calculates lag time τ (t) corresponding to current zinc coating thickness thickness measuring apparatus measuring value CWm (t),
Second model zinc coating thickness predicted value CWd (t) is calculated according to the process variable at t- τ (t) moment corresponding with the lag time τ (t),
The process variable at t- τ (t) moment corresponding with the lag time τ (t) includes air pressure history value P (t- τ (t)), production
Linear velocity history value S (t- τ (t)), air knife knife are away from history value D (t- τ (t))；
B2, according to the first model prediction zinc coating thickness value CWp (t), several zinc layers in the 3rd preset time are thick
Spend thickness measuring apparatus measuring value CWm (t), CWm (t-1) ..., and the second model zinc coating thickness predicted value CWd (t) corresponding thereto,
CWd (t-1) ..., offset correction calculating is carried out, obtains the zinc coating thickness predicted value CWp ' (t) after the correction.
As a kind of prioritization scheme, in the step A2, it is inclined that current laboratory is calculated using the method for sliding-window filtering
Poor correction amount delta CWLab(t), it is specially：
The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2,
CWm2) ..., (CWLab K, CWmK),
The K is window size, then, and current laboratory offset correction amount
Wherein αk, k=1,2 ..., K is one group of default weighted filter coefficients.
As a kind of prioritization scheme, it is characterised in that current zinc coating thickness thickness measuring apparatus measuring value CWm described in step B1
(t) computational methods of lag time τ (t) corresponding to are specially：
L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：
Wherein, Δ T is the controlling cycle of control system, S (t), S (t-1) ... for the current manufacturing lines velocity amplitude and
The line speed value of several controlling cycles in the second preset time preserved in database.
As a kind of prioritization scheme, it is characterised in that the offset correction is carried out in step B2 the correction is calculated
The process of zinc coating thickness predicted value CWp ' (t) afterwards is specially：
Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
As a kind of prioritization scheme, it is characterised in that according to the current laboratory offset correction amount Δ CWLab(t), school
Zinc coating thickness predicted value CWp ' (t) and zinc coating thickness setting value Rcw (t) after just, generate current air pressure setting value Pfbc
(t) process includes：
C1, current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition is corrected
The zinc coating thickness setting value R ' cw (t) crossed, i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；
C2, using corrected zinc coating thickness setting value R ' cw (t), correction after zinc coating thickness predicted value CWp ' (t) as
Input quantity, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
Based on same inventive concept, the present invention also provides a kind of zinc coating thickness control based on time_varying delay offset correction technology
System, including feedback controller module, production process model module, lag time computing module, model bias correction module, change
Test room detection error correction module, database；
The feedback controller module is used for：Current laboratory deviation is obtained from the laboratory detection error correction module
Correction amount delta CWLab(t), from the model bias correction module corrected after zinc coating thickness predicted value CWp ' (t), according to
Zinc coating thickness setting value Rcw (t), the current laboratory offset correction amount Δ CWLab(t) and the correction after zinc coating thickness
Predicted value CWp ' (t), generate current air pressure setting value Pfbc(t) as the feedback control amount in current control period；
The laboratory detection error correction module is used for：Judge whether there is new laboratory detected value CW in databaseLab
(t),
Nothing, then current laboratory offset correction amount Δ CWLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) to be previous
The laboratory offset correction amount of individual controlling cycle,
Have, then by the new laboratory detected value CWLab(t) and synchronization zinc coating thickness thickness measuring apparatus measuring value CWm
(t) it is saved into together in database as a record, the K bars note preserved in the first preset time is taken out from the database
Record, calculate the current laboratory offset correction amount Δ CWLab(t)；
The lag time computing module is used for：Protected according in current manufacturing lines velocity measurement S (t) and the database
The line speed value of several controlling cycles, calculates current zinc coating thickness thickness measuring apparatus measuring value in the second preset time deposited
Lag time τ (t) corresponding to CWm (t), the process at t- τ (t) moment corresponding with the lag time τ (t) is obtained from database
Amount, the process variable at t- τ (t) moment corresponding with the lag time τ (t) include air pressure history value P (t- τ (t)), life
Producing line speed history value S (t- τ (t)), air knife knife are away from history value D (t- τ (t))；
The production process model module is used for：
According to default production process model and current air pressure measured value P (t), air knife knife away from measured value D
(t), line speed measured value S (t) calculates the first model prediction zinc coating thickness value CWp (t),
And the second model zinc layers thickness is calculated according to the process variable at t- τ (t) moment corresponding with the lag time τ (t)
Spend predicted value CWd (t)；
The model bias correction module is used for：According to the first model prediction zinc coating thickness value CWp (t), the 3rd is pre-
If several zinc coating thickness thickness measuring apparatus measuring value CWm (t), CWm (t-1) in the time ..., and the second model corresponding thereto
Zinc coating thickness predicted value CWd (t), CWd (t-1) ..., carry out offset correction calculating, the zinc coating thickness obtained after the correction is pre-
Measured value CWp ' (t).
As a kind of prioritization scheme, the laboratory detection error correction module is further used for：Filtered using sliding window
The method of ripple calculates current laboratory offset correction amount Δ CWLab(t), it is specially：
The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2,
CWm2) ..., (CWLab K, CWmK),
The K is window size, then, and current laboratory offset correction amount
Wherein αk, k=1,2 ..., K is one group of default weighted filter coefficients.
As a kind of prioritization scheme, the lag time computing module is used to calculate current zinc coating thickness thickness measuring apparatus measuring value
The process of lag time τ (t) corresponding to CWm (t) is specially：
L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：
Wherein, Δ T is the controlling cycle of control system, S (t), S (t-1) ... for the current manufacturing lines velocity amplitude and
The line speed value of several controlling cycles in the second preset time preserved in database.
As a kind of prioritization scheme, the model bias correction module be used to carrying out the offset correction be calculated it is described
The process of zinc coating thickness predicted value CWp ' (t) after correction is specially：
Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
As a kind of prioritization scheme, the feedback controller module is specifically used for：
Pass through current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition is corrected
The zinc coating thickness setting value R ' cw (t) crossed, i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；
Using the corrected zinc coating thickness setting value R ' cw (t), correction after zinc coating thickness predicted value CWp ' (t) as
Input quantity, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
Compared with prior art, the present invention has following beneficial effect：
(1) calculating of the present invention to system lag time is from its caused mechanism, with other using System Discrimination etc.
The method of method estimating system pure delay time parameter is compared, and the lag time that the present invention is calculated is more accurate, Neng Gouyou
Effect is avoided because of the fluctuation to model parameter estimation production process caused by inaccurate.
(2) present invention carries out online offset correction to production process model, is pre-established with use changeless black
The scheme of case or mechanism model is compared, and the present invention can be effectively overcome because model is inaccurate, production status changes, random disturbances
Etc. model caused by factor and real system mismatch problems.
(3) existing thickness of coating closed-loop control system is fed back using calibrator on-line measurement data as thickness measurements more
Information carry out closed-loop control, but calibrator due to metering system, installation safeguard etc. reason larger measured deviation be present, merely according to
The requirement of qualified product can not be fully met by relying calibrator data to carry out closed-loop control, and the present invention uses laboratory data
Correction module calculates the deviation between laboratory analysis and thickness measuring apparatus measuring value in real time, for the setting value progress to control system
Amendment, this method can effectively eliminate control more than it is poor, significantly improve zinc coat thickness control precision, ensure the qualification rate of product.
Brief description of the drawings
The detailed description made by reading with reference to the following drawings to non-limiting example, further feature of the invention,
Objects and advantages will become more apparent upon：
Fig. 1 is that a kind of zinc coating thickness control system framework based on time_varying delay offset correction technology in alternative embodiment is former
Reason figure；
Fig. 2 is a kind of zinc coating thickness control flow chart based on time_varying delay offset correction technology in alternative embodiment；
Fig. 3 is a kind of reduction thickness of coating ripple effect figure (thickness of coating 120g/m in alternative embodiment2Specification)；
Fig. 4 is a kind of reduction thickness of coating ripple effect figure (thickness of coating 80g/m in alternative embodiment2Specification).
Embodiment
The present invention is described in detail in a manner of specific embodiment below in conjunction with the accompanying drawings.Following examples will be helpful to
Those skilled in the art further understands the present invention, but the invention is not limited in any way.It should be pointed out that to ability
For the those of ordinary skill in domain, without departing from the inventive concept of the premise, various modifications and improvements can be made.These
Belong to protection scope of the present invention.
The defects of for above-mentioned existing thickness of coating automatic control technology and deficiency, the present invention propose that one kind is based on time_varying delay
The zinc coating thickness control method and system of offset correction technology.Thickness of coating is modeled using Data Modeling Method first pre-
Producing line speed, air pressure, knife are surveyed away from influence of the factor to thickness of coating such as, height, and according to model predication value to controlled quentity controlled variable
Feedback control is carried out to eliminate influence of the external disturbance to thickness of coating；Meanwhile according to the characteristics of zinc-plated production process, accurate meter
Calculate the lag time of calibrator measurement, it is pre- using the thickness measuring apparatus measuring value and the model that match each other based on the lag time
The current model predication value of measured value Information revision, effectively overcome the feedback problem of system time-varying large time delay metrical information；In addition, draw
Enter discrete inspection laboratory analysis of data and the difference of thickness measurement online instrument thickness measurements is entered to the setting value of zinc coat thickness control target
Row amendment, it is poor more than the stable state of closed-loop control to reduce.The present invention can effectively overcome zinc-plated production time-varying large time delay, external interference factor
Numerous, model mismatch and calibrator measure not accurate etc. enough controlling difficulties, and realizing reduces thickness of coating fluctuation, improves control essence
Degree and qualification rate, the target for reducing excess zinc consumption.
A kind of zinc coating thickness control method based on time_varying delay offset correction technology, in controlling cycle t, including：By
Offset correction amount in laboratory calculates workflow management and obtains current laboratory offset correction amount Δ CWLab(t), corrected zinc layers
Thickness prediction value calculation process calculates the zinc coating thickness predicted value CWp ' (t) after being corrected, inclined according to the current laboratory
Poor correction amount delta CWLab(t) zinc coating thickness predicted value CWp ' (t) and zinc coating thickness setting value Rcw (t) after, correcting, generation are worked as
Former air knife pressure set points Pfbc(t) as the feedback control amount in current control period；
The laboratory offset correction amount calculation process includes：
A1, judge whether there is new laboratory detected value CW in databaseLab(t),
Have, then into step A2,
Nothing, then current laboratory offset correction amount Δ CWLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) to be previous
The laboratory offset correction amount of individual controlling cycle；
A2, by the new laboratory detected value CWLab(t) and synchronization zinc coating thickness thickness measuring apparatus measuring value CWm
(t) it is saved into together in database as a record, the K bars note preserved in the first preset time is taken out from the database
Record, calculate the current laboratory offset correction amount Δ CWLab(t)；
The zinc coating thickness predictor calculation flow of the correction includes：
B1, according to default production process model and current air pressure measured value P (t), air knife knife away from measured value D
(t), line speed measured value S (t) calculates the first model prediction zinc coating thickness value CWp (t),
According in the second preset time preserved in current manufacturing lines velocity measurement S (t) and the database several
The line speed value of controlling cycle, calculates lag time τ (t) corresponding to current zinc coating thickness thickness measuring apparatus measuring value CWm (t),
Second model zinc coating thickness predicted value CWd (t) is calculated according to the process variable at t- τ (t) moment corresponding with the lag time τ (t),
The process variable at t- τ (t) moment corresponding with the lag time τ (t) includes air pressure history value P (t- τ (t)), production
Linear velocity history value S (t- τ (t)), air knife knife are away from history value D (t- τ (t))；
B2, according to the first model prediction zinc coating thickness value CWp (t), several zinc layers in the 3rd preset time are thick
Spend thickness measuring apparatus measuring value CWm (t), CWm (t-1) ..., and the second model zinc coating thickness predicted value CWd (t) corresponding thereto,
CWd (t-1) ..., offset correction calculating is carried out, obtains the zinc coating thickness predicted value CWp ' (t) after the correction.
In the present embodiment, the laboratory offset correction amount calculation process and the zinc coating thickness predicted value meter of the correction
Calculate flow to carry out simultaneously, but the invention is not restricted to this.When first preset time, the second preset time, the 3rd preset
Between refer both in a period of time before current time, the present embodiment is by the way that multiple data of historical record and current data are carried out
Calculate, the error under individual data calculated case is advantageously reduced, so that predicted value is more accurate.First is pre- in the present embodiment
If time, the second preset time, the 3rd preset time can be identical period or different periods, the invention is not restricted to
This.The controlling cycle of heretofore described system is a moment.
As a kind of embodiment, in the step A2, current laboratory deviation is calculated using the method for sliding-window filtering
Correction amount delta CWLab(t), it is specially：
The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2,
CWm2) ..., (CWLab K, CWmK),
The K is window size, then, and current laboratory offset correction amount
Wherein αk, k=1,2 ..., K is one group of default weighted filter coefficients.This moment is calculated using laboratory Data correction module
Laboratory detection error correction value delta CWLab(t).Using calibrator and laboratory detection error correction value delta CWLab(t) to current
Thickness of coating setting amendment, obtain R ' cw (t).
As a kind of embodiment, when being lagged corresponding to current zinc coating thickness thickness measuring apparatus measuring value CWm (t) described in step B1
Between τ (t) computational methods be specially：
L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：
Wherein, Δ T is the controlling cycle of control system, S (t), S (t-1) ... for the current manufacturing lines velocity amplitude and
The line speed value of several controlling cycles in the second preset time preserved in database.The lag time τ (t), which is equal to, to be worked as
Coil of strip extension position is moved to calibrator elapsed time from air knife apparatus at preceding calibrator.
It is pre- that the zinc coating thickness that the offset correction is calculated after the correction is carried out as a kind of embodiment, in step B2
Measured value CWp ' (t) process is specially：
Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
As a kind of embodiment, according to the current laboratory offset correction amount Δ CWLab(t) zinc coating thickness after, correcting
Predicted value CWp ' (t) and zinc coating thickness setting value Rcw (t), generates current air pressure setting value Pfbc(t) process includes：
C1, current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition is corrected
The zinc coating thickness setting value R ' cw (t) crossed, i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；
C2, using corrected zinc coating thickness setting value R ' cw (t), correction after zinc coating thickness predicted value CWp ' (t) as
Input quantity, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
As shown in figure 1, the zinc layers after corrected thickness of coating setting value R ' cw (t) to be calculated and correction are thick
Predicted value CWp ' (t) is spent as input, is calculated using feedback controller and is obtained feedback control amount, is output to galvanization production line basis
Automate level.
It is analyzed as follows for the technique effect of above-described embodiment：
The present invention accurately calculates measurement delay time of production system using lag time computing module, and this module is from measurement
Mechanism is set out caused by hysteresis, and accurate calculate is applied to air knife detects to its output response by calibrator from controlled quentity controlled variable
The measurement delay time.With using System Discrimination the methods of estimating system pure delay time parameter method compared with, the present invention calculate
Obtained lag time is more accurate, can effectively avoid because to the ripple of model parameter estimation production process caused by inaccurate
It is dynamic.
The present invention uses model bias correction module, the model predication value and thickness measuring that the elapsed time matches before making full use of
Deviation information between apparatus measuring value, real-time online correction model predicted value, dynamically approaches thickness measuring apparatus measuring value.Using model
Offset correction module, the deviation between model prediction and thickness measuring apparatus measuring value can be reduced as far as possible, while overcome actual production
Because model is inaccurate in journey, the influence of the fluctuation of production status and the interference of enchancement factor to control effect.
The present invention uses laboratory Data correction module, is surveyed according to the calibrator at the same coil of strip same position at nearest moment
Value and inspection chemical examination measurement value information, calculate the thickness measuring apparatus measuring value and the deviation of inspection chemical examination measurement under current production status,
The setting value of zinc coat thickness control target is modified.Because the measurement data of calibrator has deviation, thickness measuring is completely dependent on
The qualification rate of product can not be completely secured in instrument data, and manufacturing enterprise samples often through laboratory method to thickness of coating
The foundation whether qualified as product is judged is analyzed, but laboratory method generally requires to take a long time, this discrete survey
Amount data can not be controlled directly as the measurement feedback information of closed loop to production process, the chemical examination number of chambers designed by the present invention
According to correction module, discrete laboratory analyze data can be incorporated into line closed loop control system and setting value is modified,
It is poor more than controlling to effectively reduce, and significantly improves the precision of coating control, and then improves the qualification rate of product, while reduces excess zinc
Waste.
To sum up, the thickness of coating closed-loop control system proposed by the present invention based on time_varying delay offset correction technology uses data
Modeling method is modeled the shadow of prediction producing line speed, air pressure, knife away from factors such as, height to thickness of coating to thickness of coating
Ring, and feedback control is used to suppress external disturbance according to predicted value；It is inclined based on accurate lag time computational methods and model
Poor computational methods, the model predication value after offset correction is calculated, substitutes tradition anti-using the model predication value after the correction
The calibrator detected value that feedback control system uses can effectively improve controller response speed, overcome change to survey as closed loop feedback amount
The harmful effect that amount hysteresis is brought to closed-loop control system；Meanwhile introduce laboratory Data correction mechanism and closed-loop system is set
Value is modified, advantageously reduce control stable state more than it is poor, reduce the consumption of excess zinc, improve the qualification rate of final products.
As a kind of embodiment, a kind of zinc coating thickness control method based on time_varying delay offset correction technology includes step
S101-S111。
S101. production process amount data preserve.By the extension position of the currently coil of strip at the coil of strip number of plating, calibrator,
Air pressure P (t), air knife knife are saved into process away from D (t), line speed S (t), calibrator thickness of coating measured value CWm (t)
Measure in historical data base；
S102. inspection assay system database is accessed, is inquired about compared with a upper controlling cycle, if has new laboratory detection
Data produce, if any S103 is then entered, otherwise into S104.New laboratory detection data, refer to that the Thickness sensitivity value belongs to
The coil of strip not detected in the past, it is necessary to while include coil of strip number, inspection chemical examination sampling point position and thickness of coating metrical information；
S103. this moment laboratory detection error correction value delta CW is calculated using laboratory Data correction moduleLab(t).First according to new
Laboratory detection data included coil of strip number, inspection chemical examination sampling point position information in process variable historical data library lookup coil of strip
It is number identical, the measurement position thickness measuring apparatus measuring value closest with inspection chemical examination sampled point, by the thickness measuring apparatus measuring value and laboratory
Measured value is saved into the database of laboratory together as a record.Deviation is calculated using the method for sliding-window filtering below
Correction value delta CWLab(t).Without loss of generality, it is assumed that window size K, and the K bars record preserved recently in the database of laboratory
It is worth for (CWLab 1, CWm1), (CWLab 2, CWm2) ..., (CWLab K, CWmK), then：
α in formula (1)k, k=1,2 ..., K is one group of weighted filter coefficients set in advance.
When S104. detecting data in not new laboratory, current laboratory offset correction value is equal to last moment
Laboratory offset correction value Δ CWLab(t)=Δ CWLab(t-1)。
S105. the model predication value of current thickness of coating is calculated.The air pressure P (t) that will currently measure, air knife knife is away from D
(t), line speed S (t) inputs as production process model module, calculates the output of production process model module, i.e., current
First model prediction zinc coating thickness value CWp (t).
S106. time lag calculates.The present invention calculates the measurement delay time τ (t) of current calibrator using mechanism method, and this is stagnant
The time is moved to calibrator elapsed time equal to coil of strip extension position at current calibrator from air knife apparatus afterwards.Assuming that from gas
The distance of knife system to calibrator is L, and according to speed, time, the fundamental relation apart from three is as follows：
Wherein Δ T is control system controlling cycle, the line speed value S (t), S (t- of nearest several controlling cycles
1) ... can be obtained from process variable historical data base.Using the above-mentioned equation of following Algorithm for Solving (2)：
Initialization：Defined variable Lsum=0, k=0；
The first step：Calculate Lsum=Lsum+ΔT·S(t-k)；
Second step：If Lsum＜ L, make k=k+1, return to the first step, otherwise into the 3rd step；
3rd step：τ (t)=k is made, calculating terminates.
Calculated using algorithm above, time lag τ (t) can be obtained.
S107. process variable time domain translates, with the current calibrator measurement delay time τ (t) that is calculated in S106 for base
Plinth, to air pressure, air knife knife is away from these three process variables of line speed are translated in time domain.I.e. in process variable history number
According to variable P (t- τ (t)) is found in storehouse, S (t- τ (t)), D (t- τ (t)), these three production process amounts are in t- τ (t) times by gas
Acted at knife on coil of strip, after elapsed time length τ (t), just had influence on current thickness measuring apparatus measuring value.
S108. the model predication value of current thickness of coating is calculated.By air pressure P (t- τ (t)), air knife knife is away from S (t- τ
(t)), line speed D (t- τ (t)) as production process model module input, its output is calculated, i.e., with current thickness measuring
The second corresponding apparatus measuring value CWm (t) model prediction zinc coating thickness value CWd (t), using CWm (t) and CWd (t) value as one
Bar record preserves together to be entered in model bias correction database.
S109. model bias correction calculates.Specifically, the deviation in time window N is entered using slip window sampling
Row weighting is smooth, then data record (CWm (t), the CWd of N number of controlling cycle recently are read from model bias correction database
(t)), (CWm (t-1), CWd (t-1)) ..., (CWm (t-N+1), CWd (t-N+1)), the predicted value after being corrected as follows
CWp′(t)：
β in formula (3)k, k=0,1 ..., N-1 is one group of weight coefficient set in advance.
S110. thickness of coating set point correct.Using current calibrator and laboratory measured deviation amount Δ CWLab(t) repeatedly
In thickness of coating setting value before sovolin, obtain：
R ' cw (t)=Rcw (t)+Δ CWLab(t) (4)
S111. feedback control amount calculates.Its model predication value CWp ' (t) inputted after the correction being calculated for S109,
And the corrected thickness of coating setting value R ' cw (t) that S110 is calculated.Controlled quentity controlled variable is calculated using standard PID algorithm, it is defeated
Go out for air pressure setting value Pfbc(t)。
Listed in Fig. 3, Fig. 4 before being put into operation using the control system of the inventive method design and after putting into operation for reducing product
Thickness of coating fluctuation, the Contrast on effect for reducing excess zinc consumption.Fig. 3 is using thickness of coating as 120g/m2Exemplified by the product of specification, it is
Before system puts into operation, thickness of coating quality fluctuation is larger, in order to ensure that product quality meets the requirement of quality of coating lower specification limit, production
Enterprise has to zinc coat thickness control desired value being set in 125g/m2, otherwise will appear from substandard product；After system puts into operation,
Because zinc coat thickness control precision is greatly improved, the quality fluctuation of product greatly reduces, and is ensureing product qualification rate
Under the premise of, zinc coat thickness control desired value can be reduced to 122.5g/m2, it can be seen that the 2.5g/m reduced2As use
The excess zinc consumption of the unit product of reduction after the method for the invention.Fig. 4 is using thickness of coating as 80g/m2The product of specification is
Example, before system puts into operation, thickness of coating quality fluctuation is larger, in order to ensure that product quality meets wanting for quality of coating lower specification limit
Ask, manufacturing enterprise has to zinc coat thickness control desired value being set in 84.5g/m2, otherwise will appear from substandard product；System
After putting into operation, because zinc coat thickness control precision is greatly improved, the quality fluctuation of product greatly reduces, and is ensureing product conjunction
On the premise of lattice rate, zinc coat thickness control desired value can be reduced to 82.7g/m2, it can be seen that the 1.8g/m reduced2As
Consumed using the excess zinc of the unit product of reduction after the method for the invention.
Based on same inventive concept, the present invention also provides as shown in Figure 1 a kind of based on time_varying delay offset correction technology
Zinc coating thickness control system, including feedback controller module, production process model module, lag time computing module, model are inclined
Poor correction module, laboratory detection error correction module, database；
The feedback controller module is used for：Current laboratory deviation is obtained from the laboratory detection error correction module
Correction amount delta CWLab(t), from the model bias correction module corrected after zinc coating thickness predicted value CWp ' (t), according to
Zinc coating thickness setting value Rcw (t), the current laboratory offset correction amount Δ CWLab(t) and the correction after zinc coating thickness
Predicted value CWp ' (t), generate current air pressure setting value Pfbc(t) as the feedback control amount in current control period；
The laboratory detection error correction module is used for：Judge whether there is new laboratory detected value CW in databaseLab
(t),
Nothing, then current laboratory offset correction amount Δ CWLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) to be previous
The laboratory offset correction amount of individual controlling cycle,
Have, then by the new laboratory detected value CWLab(t) and synchronization zinc coating thickness thickness measuring apparatus measuring value CWm
(t) it is saved into together in database as a record, the K bars note preserved in the first preset time is taken out from the database
Record, calculate the current laboratory offset correction amount Δ CWLab(t)；
The lag time computing module is used for：Protected according in current manufacturing lines velocity measurement S (t) and the database
The line speed value of several controlling cycles, calculates current zinc coating thickness thickness measuring apparatus measuring value in the second preset time deposited
Lag time τ (t) corresponding to CWm (t), the process at t- τ (t) moment corresponding with the lag time τ (t) is obtained from database
Amount, the process variable at t- τ (t) moment corresponding with the lag time τ (t) include air pressure history value P (t- τ (t)), life
Producing line speed history value S (t- τ (t)), air knife knife are away from history value D (t- τ (t))；
The production process model module is used for：
According to default production process model and current air pressure measured value P (t), air knife knife away from measured value D
(t), line speed measured value S (t) calculates the first model prediction zinc coating thickness value CWp (t),
And the second model zinc layers thickness is calculated according to the process variable at t- τ (t) moment corresponding with the lag time τ (t)
Spend predicted value CWd (t)；
The model bias correction module is used for：According to the first model prediction zinc coating thickness value CWp (t), the 3rd is pre-
If several zinc coating thickness thickness measuring apparatus measuring value CWm (t), CWm (t-1) in the time ..., and the second model corresponding thereto
Zinc coating thickness predicted value CWd (t), CWd (t-1) ..., carry out offset correction calculating, the zinc coating thickness obtained after the correction is pre-
Measured value CWp ' (t).
As a kind of prioritization scheme, the laboratory detection error correction module is further used for：Filtered using sliding window
The method of ripple calculates current laboratory offset correction amount Δ CWLab(t), it is specially：
The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2,
CWm2) ..., (CWLab K, CWmK),
The K is window size, then, and current laboratory offset correction amount
Wherein αk, k=1,2 ..., K is one group of default weighted filter coefficients.
As a kind of prioritization scheme, the lag time computing module is used to calculate current zinc coating thickness thickness measuring apparatus measuring value
The process of lag time τ (t) corresponding to CWm (t) is specially：
L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：
Wherein, Δ T is the controlling cycle of control system, S (t), S (t-1) ... for the current manufacturing lines velocity amplitude and
The line speed value of several controlling cycles in the second preset time preserved in database.
As a kind of prioritization scheme, the model bias correction module be used to carrying out the offset correction be calculated it is described
The process of zinc coating thickness predicted value CWp ' (t) after correction is specially：
Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
As a kind of prioritization scheme, the feedback controller module is specifically used for：
Pass through current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition is corrected
The zinc coating thickness setting value R ' cw (t) crossed, i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；
Using the corrected zinc coating thickness setting value R ' cw (t), correction after zinc coating thickness predicted value CWp ' (t) as
Input quantity, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
The key point of above example is：
(1) thickness of coating modeling and forecasting method；Due to being installed Processes and apparatus, measurement feedback exists galvanizing process
Larger change lag time, considerable influence is caused to closed-loop control quality, the present invention is based on production process model, to plating thickness
Degree is modeled prediction producing line speed, air pressure, knife away from influence of the factor to thickness of coating such as, height, and using feedback control
Make to eliminate influence of the above-mentioned disturbance to thickness of coating；
(2) measured value based on time_varying delay and predicted value matching process；The present invention is for zinc-plated production system measurement time-varying
Mechanismic design time lag computational methods caused by large time delay, when this method can accurately calculate the measurement delay of current calibrator
Between, so as to which time match will be carried out with model predication value with the thickness measuring apparatus measuring value for becoming hysteresis characteristic, using its difference as anti-
Feedback is corrected to forecast model, effectively overcomes the mould caused by the factors such as model inaccuracy, production status change, random disturbances
Type and real system mismatch problems, improve the prediction precision of model.
(3) the set point correct method based on discrete laboratory assay value Yu thickness measurement online instrument data, tradition are based on thickness measuring
The thickness of coating closed loop control method of instrument data is influenceed by calibrator measurement accuracy, it is impossible to it is qualified to fully meet thickness of coating
Requirement.The thickness measuring apparatus measuring value of this method same same inspection positions of coil of strip intraoral to time slip-window and laboratory assay value
Between deviation be weighted processing, the deviation of current calibrator and laboratory measurement is obtained, using the amendment of the deviation
Thickness of coating setting value, it can effectively prevent calibrator and detect the occurrence of qualified but product is finally unqualified, improve final
The qualification rate of product, while reduce the consumption of excess zinc.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make various deformations or amendments within the scope of the claims, this not shadow
Ring the substantive content of the present invention.
Claims (10)
- A kind of 1. zinc coating thickness control method based on time_varying delay offset correction technology, it is characterised in that in t, including：Through Cross laboratory offset correction amount and calculate the laboratory offset correction amount Δ CW that workflow management obtains current zinc coating thicknessLab(t), pass through The zinc coating thickness predictor calculation workflow management of overcorrect corrected after zinc coating thickness predicted value CWp ' (t), according to described The laboratory offset correction amount Δ CW of current zinc coating thicknessLab(t) the zinc coating thickness predicted value CWp ' (t) and zinc layers after, correcting are thick Setting value Rcw (t) is spent, generates current air pressure setting value Pfbc(t) as the feedback control amount in current control period；The laboratory offset correction amount calculation process includes：A1, judge the laboratory detected value CW for whether having new zinc coating thickness in databaseLab(t),Have, then into step A2,The laboratory offset correction amount Δ CW of nothing, then current zinc coating thicknessLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) For the laboratory offset correction amount of the zinc coating thickness of previous controlling cycle；A2, by the laboratory detected value CW of the new zinc coating thicknessLab(t) the zinc coating thickness calibrator with synchronization measures Value CWm (t) is saved into database together as a record, takes out what is preserved in the first preset time from the database K bars record, and calculate the laboratory offset correction amount Δ CW of the current zinc coating thicknessLab(t)；The zinc coating thickness predictor calculation flow of the correction includes：B1, according to default production process model and current air pressure measured value P (t), air knife knife away from measured value D (t), Line speed measured value S (t) calculates the first model prediction zinc coating thickness value CWp (t)；According to several controls in the second preset time preserved in current manufacturing lines velocity measurement S (t) and the database The line speed value in cycle, calculates lag time τ (t) corresponding to current zinc coating thickness thickness measuring apparatus measuring value CWm (t), according to The process variable at t- τ (t) moment corresponding with the lag time τ (t) calculates the second model zinc coating thickness predicted value CWp (t- π (t)), the second model zinc coating thickness predicted value CWp (t- π (t)) is designated as CWd (t), the described and lag time τ (t) again The process variable at corresponding t- τ (t) moment includes air pressure history value P (t- τ (t)), line speed history value S (t- τ (t)), air knife knife is away from history value D (t- τ (t))；B2, according to the first model prediction zinc coating thickness value CWp (t), several zinc coating thickness in the 3rd preset time are surveyed Thick apparatus measuring value CWm (t), CWm (t-1) ..., and the second model zinc coating thickness predicted value CWd (t), CWd corresponding thereto (t-1) offset correction calculating ..., is carried out, obtains the zinc coating thickness predicted value CWp ' (t) after the correction.
- 2. according to the method for claim 1, it is characterised in that in the step A2, using the method for sliding-window filtering Calculate current laboratory offset correction amount Δ CWLab(t), it is specially：The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2, CWm2) ..., (CWLab K, CWmK),The K is window size, then, and current laboratory offset correction amountIts Middle αk, k=1,2 ..., K is one group of default weighted filter coefficients.
- 3. according to the method for claim 1, it is characterised in that current zinc coating thickness thickness measuring apparatus measuring value described in step B1 The computational methods of lag time τ (t) corresponding to CWm (t) are specially：L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：<mrow> <mi>L</mi> <mo>=</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>&tau;</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </munderover> <mi>&Delta;</mi> <mi>T</mi> <mo>&CenterDot;</mo> <mi>S</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow>Wherein, Δ T be control system controlling cycle, S (t), S (t-1) ... be the current manufacturing lines velocity amplitude and data The line speed value of several controlling cycles in the second preset time preserved in storehouse.
- 4. according to the method for claim 1, it is characterised in that carried out in step B2 the offset correction be calculated it is described The process of zinc coating thickness predicted value CWp ' (t) after correction is specially：<mrow> <msup> <mi>CWp</mi> <mo>&prime;</mo> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>C</mi> <mi>W</mi> <mi>p</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>+</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>N</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>&beta;</mi> <mi>k</mi> </msub> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <mi>C</mi> <mi>W</mi> <mi>d</mi> <mo>(</mo> <mrow> <mi>t</mi> <mo>-</mo> <mi>k</mi> </mrow> <mo>)</mo> <mo>-</mo> <mi>C</mi> <mi>W</mi> <mi>m</mi> <mo>(</mo> <mrow> <mi>t</mi> <mo>-</mo> <mi>k</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> </mrow>Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
- 5. according to the method for claim 1, it is characterised in that according to the current laboratory offset correction amount Δ CWLab (t) zinc coating thickness predicted value CWp ' (t) and zinc coating thickness setting value Rcw (t) after, correcting, current air pressure setting is generated Value Pfbc(t) process includes：C1, current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition obtains corrected Zinc coating thickness setting value R ' cw (t), i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；C2, input is used as using the zinc coating thickness predicted value CWp ' (t) after corrected zinc coating thickness setting value R ' cw (t), correction Amount, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
- 6. a kind of zinc coating thickness control system based on time_varying delay offset correction technology, it is characterised in that including feedback controller Module, production process model module, lag time computing module, model bias correction module, laboratory detection error straightening die Block, database；The feedback controller module is used for：The chemical examination of current zinc coating thickness is obtained from the laboratory detection error correction module Room offset correction amount Δ CWLab(t), from the model bias correction module corrected after zinc coating thickness predicted value CWp ' (t), according to zinc coating thickness setting value Rcw (t), the laboratory offset correction amount Δ CW of the current zinc coating thicknessLab(t) and institute The zinc coating thickness predicted value CWp ' (t) after correction is stated, generates current air pressure setting value Pfbc(t) it is used as current control period In feedback control amount；The laboratory detection error correction module is used for：Judge whether to have in database the laboratory of new zinc coating thickness to detect Value CWLab(t),The laboratory offset correction amount Δ CW of nothing, then current zinc coating thicknessLab(t)=Δ CWLab(t-1), the Δ CWLab(t-1) For the laboratory offset correction amount of the zinc coating thickness of previous controlling cycle,Have, then by the laboratory detected value CW of the new zinc coating thicknessLab(t) the zinc coating thickness calibrator with synchronization is surveyed Value CWm (t) is saved into database together as a record, is taken out in the first preset time and is preserved from the database K bars record, calculate the laboratory offset correction amount Δ CW of the current zinc coating thicknessLab(t)；The lag time computing module is used for：According to what is preserved in current manufacturing lines velocity measurement S (t) and the database The line speed value of several controlling cycles, calculates current zinc coating thickness thickness measuring apparatus measuring value CWm (t) in second preset time Corresponding lag time τ (t), the process variable at t- τ (t) moment corresponding with the lag time τ (t), institute are obtained from database Stating the process variable at t- τ (t) moment corresponding with the lag time τ (t) includes air pressure history value P (t- τ (t)), production line Speed history value S (t- τ (t)), air knife knife are away from history value D (t- τ (t))；The production process model module is used for：According to default production process model and current air pressure measured value P (t), air knife knife away from measured value D (t), life Producing line velocity measurement S (t) calculates the first model prediction zinc coating thickness value CWp (t),It is and pre- according to the process variable at t- τ (t) moment corresponding with the lag time τ (t) the second model zinc coating thickness of calculating Measured value CWp (t- π (t)), the second model zinc coating thickness predicted value CWp (t- π (t)) are designated as CWd (t) again；The model bias correction module is used for：When being preset according to the first model prediction zinc coating thickness value CWp (t), the 3rd In several zinc coating thickness thickness measuring apparatus measuring value CWm (t), CWm (t-1) ..., and the second model zinc layers corresponding thereto Thickness prediction value CWd (t), CWd (t-1) ..., carry out offset correction calculating, obtain the zinc coating thickness predicted value after the correction CWp′(t)。
- 7. system according to claim 6, it is characterised in that the laboratory detection error correction module is further used In：Current laboratory offset correction amount Δ CW is calculated using the method for sliding-window filteringLab(t), it is specially：The K bars preserved in the first preset time are taken out from the database and are recorded as (CWLab 1, CWm1), (CWLab 2, CWm2) ..., (CWLab K, CWmK),The K is window size, then, and current laboratory offset correction amountWherein αk, k=1,2 ..., K is one group of default weighted filter coefficients.
- 8. system according to claim 6, it is characterised in that the lag time computing module is used to calculate current zinc layers The process of lag time τ (t) corresponding to thickness thickness measuring apparatus measuring value CWm (t) is specially：L is the distance from air knife apparatus to calibrator, and according to speed, time, the fundamental relation apart from three is as follows：<mrow> <mi>L</mi> <mo>=</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>&tau;</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </munderover> <mi>&Delta;</mi> <mi>T</mi> <mo>&CenterDot;</mo> <mi>S</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow>Wherein, Δ T be control system controlling cycle, S (t), S (t-1) ... be the current manufacturing lines velocity amplitude and data The line speed value of several controlling cycles in the second preset time preserved in storehouse.
- 9. system according to claim 6, it is characterised in that the model bias correction module is used to carry out the deviation The process that the zinc coating thickness predicted value CWp ' (t) after the correction is calculated in correction is specially：<mrow> <msup> <mi>CWp</mi> <mo>&prime;</mo> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>C</mi> <mi>W</mi> <mi>p</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>+</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow> <mi>N</mi> <mo>-</mo> <mn>1</mn> </mrow> </munderover> <msub> <mi>&beta;</mi> <mi>k</mi> </msub> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <mi>C</mi> <mi>W</mi> <mi>d</mi> <mo>(</mo> <mrow> <mi>t</mi> <mo>-</mo> <mi>k</mi> </mrow> <mo>)</mo> <mo>-</mo> <mi>C</mi> <mi>W</mi> <mi>m</mi> <mo>(</mo> <mrow> <mi>t</mi> <mo>-</mo> <mi>k</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> </mrow>Wherein βk, k=0,1 ..., N-1 is one group of default weight coefficient.
- 10. system according to claim 6, it is characterised in that the feedback controller module is specifically used for：Pass through current laboratory offset correction amount Δ CWLab(t) the current zinc coating thickness setting value Rcw (t) of superposition obtains corrected Zinc coating thickness setting value R ' cw (t), i.e. R ' cw (t)=Rcw (t)+Δ CWLab(t)；Input is used as using the zinc coating thickness predicted value CWp ' (t) after the corrected zinc coating thickness setting value R ' cw (t), correction Amount, calculated using standard PID algorithm and obtain current air pressure setting value Pfbc(t)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510664300.4A CN105316613B (en) | 2015-10-13 | 2015-10-13 | A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510664300.4A CN105316613B (en) | 2015-10-13 | 2015-10-13 | A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105316613A CN105316613A (en) | 2016-02-10 |
| CN105316613B true CN105316613B (en) | 2017-11-21 |
Family
ID=55244919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510664300.4A Active CN105316613B (en) | 2015-10-13 | 2015-10-13 | A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105316613B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018001296A (en) * | 2016-06-28 | 2018-01-11 | 株式会社荏原製作所 | Polishing device, polishing method, and polishing control program |
| CN107544455B (en) * | 2017-08-17 | 2019-12-31 | 浙江邦业科技股份有限公司 | Asymmetric online convergence correction method applied to quality control soft instrument |
| CN110184641B (en) * | 2018-07-27 | 2021-07-30 | 新阳硅密（上海）半导体技术有限公司 | Electroplating method of electroplating device |
| CN109885869B (en) * | 2019-01-09 | 2023-01-20 | 石化盈科信息技术有限责任公司 | Online user-defined process calculation method |
| CN115094364A (en) * | 2022-06-20 | 2022-09-23 | 首钢京唐钢铁联合有限责任公司 | Method and device for controlling thickness of zinc layer of strip steel of galvanizing production line |
| CN115689063A (en) * | 2022-12-30 | 2023-02-03 | 工业富联(杭州)数据科技有限公司 | Gold film thickness prediction method and device, electronic device and storage medium |
| CN117265449B (en) * | 2023-11-21 | 2024-03-19 | 江苏友发钢管有限公司 | Steel pipe galvanization wind ring control system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04263056A (en) * | 1991-02-18 | 1992-09-18 | Nippon Steel Corp | Method for controlling plating weight in hot dip metal plating steel strip |
| CN102912275A (en) * | 2012-10-23 | 2013-02-06 | 鞍钢股份有限公司 | Automatic control system for plating thickness of hot galvanizing line |
| CN103469137A (en) * | 2013-09-07 | 2013-12-25 | 鞍钢股份有限公司 | Preset control method for thickness flying gauge change of hot-dip zinc coating |
| CN104846306A (en) * | 2015-05-07 | 2015-08-19 | 浙江中控研究院有限公司 | Zincing thickness control system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102005013103A1 (en) * | 2005-03-18 | 2006-09-28 | Sms Demag Ag | Controlled thickness reduction in hot-dip coated hot rolled steel strip and equipment used in this case |
-
2015
- 2015-10-13 CN CN201510664300.4A patent/CN105316613B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04263056A (en) * | 1991-02-18 | 1992-09-18 | Nippon Steel Corp | Method for controlling plating weight in hot dip metal plating steel strip |
| CN102912275A (en) * | 2012-10-23 | 2013-02-06 | 鞍钢股份有限公司 | Automatic control system for plating thickness of hot galvanizing line |
| CN103469137A (en) * | 2013-09-07 | 2013-12-25 | 鞍钢股份有限公司 | Preset control method for thickness flying gauge change of hot-dip zinc coating |
| CN104846306A (en) * | 2015-05-07 | 2015-08-19 | 浙江中控研究院有限公司 | Zincing thickness control system and method |
Non-Patent Citations (2)
| Title |
|---|
| 热镀锌锌层厚度自适应控制模型的研究与应用;张岩等;《钢铁》;20120229;第47卷(第2期);62-66 * |
| 连续热镀锌线镀层厚度自动控制系统研究与应用;秦大伟等;《第九届中国钢铁年会论文集》;20131023;1-5 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105316613A (en) | 2016-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105316613B (en) | A kind of zinc coating thickness control method and system based on time_varying delay offset correction technology | |
| CN104846306B (en) | A kind of Zinc Coating Thickness control system and method | |
| CN103205665B (en) | An automatic control method for zinc layer thickness in a continuous hot galvanizing zinc line | |
| CN101863088B (en) | Method for forecasting Mooney viscosity in rubber mixing process | |
| CN107526927B (en) | Blast furnace molten iron quality online robust soft measurement method | |
| CN101957598A (en) | Gray model-free control method for large time lag system | |
| CN103336891A (en) | Pseudo-measurement generation method applied to estimating condition of distribution network | |
| CN109001979A (en) | Blast-melted quality adaptation Robust Predictive Control method based on Lazy learning | |
| CN110607435B (en) | Annealing furnace plate temperature control system and method | |
| CN111522229B (en) | Parameter self-tuning MIMO different factor offset format model-free control method | |
| CN107063424A (en) | The method of belt conveyer scale main error factor analysis based on multiple linear regression model | |
| CN115121626B (en) | Hot-rolled strip steel transient hot roll shape forecasting method based on error compensation | |
| CN106868440B (en) | Method for predicting and adjusting thickness of strip steel continuous hot-dip galvanized coating | |
| CN102277468B (en) | Real-time forecasting method of LF refining furnace molten steel temperature | |
| Yu et al. | Load forecasting based on smart meter data and gradient boosting decision tree | |
| Chen et al. | Port cargo throughput forecasting based on combination model | |
| CN103510032B (en) | The deviate control method of cold rolling hot dip galvanizing plating uniformity | |
| CN110307910A (en) | A kind of walking beam reheating furnace steel billet temperature acquisition methods and device | |
| Zhi-Kun et al. | Application of weighted multiple models adaptive controller in the plate cooling process | |
| Viumdal et al. | System identification of a non-uniformly sampled multi-rate system in aluminium electrolysis cells | |
| Verma et al. | Improving sigma level of galvanization process by zinc over-coating reduction using an integrated Six Sigma and design-of-experiments approach | |
| CN105921522B (en) | Section cooling temperature self-adaptation control method based on RBF neural | |
| Edwards et al. | Coating mass control system design for a continuous galvanizing line | |
| JPH0922402A (en) | Estimating method for daily maximum power demanded | |
| CN112700050A (en) | Method and system for predicting ultra-short-term 1 st point power of photovoltaic power station |
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 |