CN108647481A - A kind of rotary kiln burning zone temperature flexible measurement method - Google Patents
A kind of rotary kiln burning zone temperature flexible measurement method Download PDFInfo
- Publication number
- CN108647481A CN108647481A CN201810923361.1A CN201810923361A CN108647481A CN 108647481 A CN108647481 A CN 108647481A CN 201810923361 A CN201810923361 A CN 201810923361A CN 108647481 A CN108647481 A CN 108647481A
- Authority
- CN
- China
- Prior art keywords
- model
- temperature
- kiln
- burning zone
- zone temperature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a kind of cement rotary kiln burning zone temperature flexible measurement methods.The method includes：It is basic model with rotary kiln end temperature model and kiln end temperature model, migrates to obtain rotary kiln burning zone temperature model to carry out burning zone temperature prediction by model.
Description
Technical field
The present invention relates to automatic measurement technical field more particularly to the flexible measurement methods of rotary kiln burning zone temperature.
Background technology
Cement is one of important foundation material of architectural engineering.New dry process for cement production flow includes mainly：It is raw
Expect preparation, clinker burning, cement production systD and manufacture.Wherein, clinker burning link is the important link of manufacture of cement, clinker
The key of quality is the control of rotary kiln burning zone temperature.Coal dust is sprayed by rotary kiln end to be mainly used for heating cement slurry,
It is allowed to temperature and reaches 1350 to 1450 degrees Celsius, rotary kiln burning zone temperature is particularly important to clinker quality, and temperature is excessively high, generates
Amount of liquid phase it is excessive, viscosity is smaller, brings difficulty to sintering reaction, while excessive temperature is larger to equipment loss；Temperature mistake
It is low, at amount of liquid phase it is less, sintering reaction is almost stagnated, and the clinker quality of generation is poor, not up to standard.Due to rotary kiln clinkering zone temperature
Degree is in close relations with product quality and energy consumption, therefore, how burning zone temperature is effectively detected, to the refinement of manufacture of cement
It is of great significance.
It is of great significance for the detection of rotary kiln burning zone temperature.First, real-time burning zone temperature value is provided, is convenient for
Judgement of the operator to rotary kiln operating mode, is better achieved rotary kiln Precise control；It secondly can be by controlling rational temperature
Degree is conducive to that kiln lining, stable thermal regulation is protected to extend the service life of equipment.Meanwhile greenhouse gases row can be reduced
It puts, environmental protection, reduces the generation of harmful NOx and CO gases, reach the environmental protection index of requirement.
Existing rotary kiln burning zone temperature mainstream is hardware detection and software detection.Hardware detection mainly has colour temperature
The methods of meter and kiln cylinder body scanning.Wherein, two-color thermometer belongs to direct detecting method, and adverse circumstances can influence colorimetric in kiln
The accuracy of detection of thermometer, while sensor life-time is not grown, detectable effective range is smaller；Kiln cylinder body scanning belongs to indirect
Detection, it is easy to be affected by environment, and operation conditions difference is affected to testing result in kiln.Software detection includes mainly base
In the method for mechanism model and soft sensor modeling and image recognition based on data, matter is mainly passed through based on mechanism model method
Amount conservation and the conservation of energy extrapolate burning zone temperature, explanatory strong, but complicated mechanism, and there are undetectable links；Base
Mainly there are the methods of BP neural network, support vector machines in data-driven soft sensor modeling, is built using inputoutput data
Mould, it is simple and practicable, there is stronger robustness.But the affinity information that existing method is underused between submodel is exported
Estimation.
Invention content
The present invention is intended to provide a kind of rotary kiln burning zone temperature flexible measurement method based on model migration.
The present invention provides a kind of flexible measurement method of rotary kiln burning zone temperature, the method includes：With temperature of kiln head mould
Type Ybase1With kiln end temperature model Ybase2For basic model, migrate to obtain burning zone temperature model Y by modelnew, to carry out
Burning zone temperature is predicted；
Ybase1=f (Xbase1), Ybase2=f (Xbase2)
Ynew=f (Ybase1,Ybase2)。
In another preferred example, the temperature of kiln head model and kiln end temperature model are used according to field data respectively is based on
The modeling method of least squares support of instant learning obtains.
In another preferred example, after selecting sample by instant learning (Just In Time, JIT) strategy, using minimum
Two, which multiply support vector machines (Least Square SVM, LSSVM), is modeled, and kiln end temperature and temperature of kiln head model are obtained.
In another preferred example, the instant learning strategy includes step：
(a) similar samples selection；
(b) locally fine point is carried out using similar sample；With
(c) library is updated the data.
The present invention also provides a kind of flexible measurement method of rotary kiln burning zone temperature, the method includes：
(1) it chooses auxiliary variable and robust multivariable pretreatment is carried out to the process data of acquisition；
(2) after pretreated data being selected modeling sample by instant learning (Just In Time, JIT) strategy,
It is modeled using least square method supporting vector machine (Least Square SVM, LSSVM), obtains temperature of kiln head model and kiln tail
The model of temperature model；With
(3) it is basic model Y with temperature of kiln head model and kiln end temperature modelbase1And Ybase2, migrate to obtain by model
Burning zone temperature model YnewCarry out burning zone temperature prediction；
Ybase1=f (Xbase1), Ybase2=f (Xbase2)
Ynew=f (Ybase1,Ybase2)。
In another preferred example, for kiln end temperature model, the auxiliary variable is dore furnace tremie pipe temperature, kiln tail cigarette
Chamber pressure and kiln owner's electromechanics stream；For temperature of kiln head model, the auxiliary variable be kiln owner's electromechanics stream, kiln hood three times air temperature and
One chamber pressure of grate-cooler.
In another preferred example, the pretreatment of the data includes carrying out abnormity point elimination to the data of acquisition；Using horse
Family name describes each variable apart from (Mahalanobis Distance),
Covariance matrixes of the wherein C between input data,For the mean value of input data；
Chi square distribution is used to determine whether for outlier, ifThen it is considered outlier；IfThen it is considered normal point.
In another preferred example, the least square method supporting vector machine expression-form is
Wherein K (xi,xj) it is kernel function；Kernel function is Radial basis kernel function, K (xi,xj)=exp (- (x-y)2/2δ2)。
In another preferred example, the method further includes using deviation correction method according to model predictive error to clinkering zone temperature
Degree model is corrected.
Accordingly, method provided by the invention makes full use of the affinity information between submodel to carry out output estimation.
Description of the drawings
Fig. 1 is cement firing system whole process figure.
Fig. 2 is kiln end temperature soft-sensing model figure.
Fig. 3 is temperature of kiln head soft-sensing model figure.
Fig. 4 is the model migration strategy figure of Kernel-based methods similitude.
Fig. 5 is shown using robust multivariable abnormality processing to kiln end temperature abnormity point elimination effect.
Fig. 6 is shown using robust multivariable abnormality processing to temperature of kiln head abnormity point elimination effect.
Fig. 7 is kiln end temperature prediction model design sketch.
Fig. 8 is temperature of kiln head prediction model design sketch.
Fig. 9 is burning zone temperature prediction model design sketch.
Specific implementation mode
The present invention is based on " process are similar ", by the information for extracting basic model kiln end temperature and temperature of kiln head model；If
Meter experiment, obtains the input/output relation of basic model；Model migration is carried out using the strategy of output planning, obtains clinkering zone temperature
Spend model.On this basis, the present invention is completed.
The present invention provides a kind of rotary kiln burning zone temperature flexible measurement method migrated based on model, and the method includes such as
Lower step：
Step 1, auxiliary variable are chosen：For kiln end temperature soft-sensing model, the auxiliary variable used is dore furnace blanking
Tube temperature degree, kiln tail smoke-box pressure and kiln owner's electromechanics stream；For temperature of kiln head soft-sensing model, the auxiliary variable used is kiln owner's machine
Electric current, kiln hood one chamber pressure of air temperature and grate-cooler three times.
Step 2, data prediction：Abnormity point elimination, this method are carried out using robust multivariable technology to the data of acquisition
The correlation between variable has been fully considered when rejecting data, and traditional Pauta criterion does not account for the correlation between variable
Property.
Step 3, data modeling：Modeling sample selection is carried out using instant learning strategy to pretreated data, is used
Modeling method be least square method supporting vector machine, obtain kiln end temperature soft-sensing model and temperature of kiln head soft-sensing model.
Step 4, model migration：Using the migration modeling method of Kernel-based methods similitude, kiln hood, kiln end temperature mould are defined
Type is basis model, and burning zone temperature model is object module, and correcting planing method using output carries out model migration, final to obtain
To burning zone temperature model.
Step 5, model correction：Using the data acquired in device operational process, deviation is used according to model predictive error
Correction method is corrected model.
Kiln end temperature soft-sensing model is used by the technique and process data analysis to cement firing system
Auxiliary variable be dore furnace tremie pipe temperature, kiln owner's electromechanics stream and kiln tail smoke-box pressure, dore furnace tremie pipe are exactly rotary kiln
Kiln tail, tremie pipe temperature directly directly influence kiln tail smoke-box temperature, and kiln owner's electromechanics stream reflects firing situation in kiln, host electric current
Big expression firing situation is good, kiln tail smoke-box stress reaction kiln ventilation situation, and to sum up above several auxiliary variables influence kiln tail temperature
Degree.In addition, for temperature of kiln head soft-sensing model, the auxiliary variable that uses is kiln hood air temperature, one chamber pressure of grate-cooler three times
With kiln owner's electromechanics stream, the intuitive reacting replacing heat situation of air temperature, the lower pressure of Room comb reflect thickness of feed layer, affect and change kiln hood three times
Thermal effect, to affect temperature of kiln head, kiln owner's electromechanics stream, which reflects, is burnt into situation in kiln, kiln owner's current of electric is bigger, firing
Situation is better.Therefore these auxiliary variables have been selected in above-mentioned steps one.
In order to improve the confidence level of model data, above-mentioned steps two are using mahalanobis distance (Mahalanobis distance)
To describe each variable, the covariance distance of Mahalanobis distance expression data.It is a kind of effective calculating two
The method of the similarity of unknown sample collection.It is in view of contacting between various characteristics unlike Euclidean distance.Covariance matrixes of the wherein C between input data,For the mean value of input data.
Chi square distribution is used to determine whether for outlier, chi square distribution is that examine the sample of sampling whether to meet specified
Probability distribution.The criterion that robust multivariable Data Preprocessing Technology uses in the present invention for：IfThen think
It is outlier, removes outlier；IfThen it is considered normal point, specificallyNumerical value is by searching for card side
Distribution table obtains.
Above-mentioned steps three need to complete two work, first, to pretreated data sample, using instant learning strategy into
Row modeling sample screens；Second is that sample after screening is obtained kiln end temperature model and kiln hood temperature using least square method supporting vector machine
Model is spent, that is, carries out two basic models of step 4.
Instant learning strategy is a kind of common Nonlinear Modeling strategy, and strategic thinking derives from database technology and part
Modeling technique.Single compared to world model and traditional partial model characterization operating mode, instant learning strategy is directed to wide variation
Process and the process of strong nonlinearity have good modeling effect.
Instant learning strategy basic thought is accumulated in data from history, is found out and is matched with current queries sample mode
Data sample is used for locally fine point, so as to obtain better modeling accuracy.Measuring similarity between sample is frequently with Europe
Family name's distance：
Wherein, xqIndicate query sample input to be predicted, the Euclidean distance between sample is bigger, shows between sample
Similarity is lower, thus the possibility for being selected as locally fine point sample is lower.Instant learning strategy step is as follows：
The first step, similar samples selection, new samples and data base querying sample carry out Euclidean distance measurement, the size of distance
The size of similitude is shown, Euclidean distance expression formula is such as
Second step is ranked up after calculating distance according to apart from size, cement process data characteristic is based on, in the present invention
A kind of embodiment in, before selected and sorted 70% sample carry out locally fine point；
Third walks, and after selecting similar sample, removes the data for most initially entering database, adds new point and enter data
Library selects 300 data points for the database volume of kiln tail, temperature of kiln head model in one embodiment of the invention,
Library can be effectively updated the data, ensures effect while reducing operation time.
Kiln tail, temperature of kiln head are modeled using modeling method of least squares support method.
Least square method supporting vector machine optimization problem is solved using Lagrangian method
Least square method supporting vector machine table
It is up to formWherein K (xi,xj) it is kernel function, the present invention uses kernel function for radial base core letter
Number, K (xi,xj)=exp (- (x-y)2/2δ2)。
Wherein it needs to be determined that parameter be γ and δ, the present invention be directed to kiln end temperature least square method supporting vector machine parameter：
γ=10, δ=10；For the least square method supporting vector machine parameter of temperature of kiln head：γ=20, δ=5.
Using the migration modeling method of Kernel-based methods similitude, kiln hood, kiln end temperature model are defined as basic model, is needed
Obtained burning zone temperature model is new model, and then corrects planing method using input and output and carry out model migration.It is basic
Step：The information extraction of basic model kiln end temperature and temperature of kiln head model；Contrived experiment obtains the input and output of basic model
Relationship；Difference assessment is carried out to basic model and object module；Model migration is carried out using the strategy of output planning；To model into
Row model is verified.
Basic model is temperature of kiln head model Ybase1With kiln end temperature model Ybase2, basic model information and new model are believed
Breath has similitude, such as when klining is in order, burning zone temperature is high, while kiln hood, kiln end temperature also can be higher；Work as kiln
Firing situation is poor, and burning zone temperature is low, while kiln hood, kiln end temperature also can be relatively low, therefore basic model information and new model
Information has similitude.Burning zone temperature new model is rotary kiln burning zone temperature model Ynew。
Ybase1=f (Xbase1), Ybase2=f (Xbase2)
Ynew=f (Ybase1,Ybase2)
Using actual field data, carried out using the least square method supporting vector machine algorithm based on instant learning strategy pre-
It surveys, instant learning database selection 30, least square method supporting vector machine parameter：γ=20, δ=5.
Above-mentioned steps five are modified model using new data during model running, are adopted according to model predictive error
Model is corrected with deviation correction method.
WhereinOutput valve after being corrected for current time,For the predicted value of current time model output, K is school
Positive coefficient, Y (t-1) andThe predicted value exported for previous moment actual value and model.Wherein, correction coefficient is especially
It is important, 6 hours are defined as a period, correction coefficient is the model error phase of present period model error and previous period
Except acquiring,
Wherein Y (ti) it is data in present period,For the average value of present period interior prediction value；Y(ti- T) be before
Data in one period,For the average value of previous period interior prediction value, K=mean (Ki), by KiIt is averaged and can be obtained
Correction factor.
In one embodiment of the invention, model is carried out to burning zone temperature new model using test set data to test
Card.
The feature that the features described above or embodiment that the present invention mentions are mentioned can be in any combination.Disclosed in this case specification
All features can be used in combination with any composition form, each feature disclosed in specification, any can provide it is identical,
The alternative characteristics of impartial or similar purpose replace.Therefore, it is only impartial or similar spy except having special instruction, revealed feature
The general example of sign.
Main advantages of the present invention are：
1, Kernel-based methods similitude carries out model migration, can utilize less number on the basis of the model having built up
According to the new model for establishing similar process.
2, using robust multivariable Data Preprocessing Technology, it is contemplated that the correlation between decision variable so that training mould
Type quality of data higher.
3, instant learning strategy and modeling method of least squares support algorithm are effectively combined, enables to model more preferable
Ground adapts to the frequent variation of operating mode.
4, method provided by the invention can effectively reduce hardware cost.
5, rotary kiln burning zone temperature flexible measurement method provided by the invention can be effectively predicted scene and be difficult to measure in real time
Burning zone temperature.
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In the following examples, the experimental methods for specific conditions are not specified, usually according to conventional strip
Part or according to the normal condition proposed by manufacturer.Unless otherwise stated, otherwise all percentage, ratio, ratio or number is pressed
Weight meter.The unit in percent weight in volume in the present invention is well-known to those skilled in the art, such as refers to 100
The weight of solute in the solution of milliliter.Unless otherwise defined, all professional and scientific terms used in text and this field are ripe
It is identical to practice meaning known to personnel.In addition, any method and material similar or impartial to described content all can be applied to
In the method for the present invention.The preferred methods and materials described herein are for illustrative purposes only.
Embodiment 1
Fig. 5, Fig. 6 are using robust multivariable abnormality processing to kiln end temperature, temperature of kiln head abnormity point elimination effect, parameter
Selection 0.95.IfThen it is considered outlier, removes outlier；IfThen it is considered just
Chang Dian.Pretreated normal data is modeled for kiln end temperature and temperature of kiln head.
Embodiment 2
Fig. 7 is kiln end temperature prediction model design sketch, instant learning database selection 300, least square method supporting vector machine
Parameter：γ=10, δ=10.
Embodiment 3
Fig. 8 is temperature of kiln head prediction model design sketch, instant learning database selection 300, least square method supporting vector machine
Parameter：γ=20, δ=5.
Embodiment 4
Fig. 9 is burning zone temperature prediction model design sketch, the kiln end temperature model that is respectively obtained according to embodiment 2 and 3 and
Temperature of kiln head model, the least square method supporting vector machine algorithm based on instant learning strategy predict burning zone temperature, i.e.,
When learning database selection 30, least square method supporting vector machine parameter：γ=20, δ=5.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not limited to the substantial technological content model of the present invention
It encloses, substantial technological content of the invention is broadly to be defined in the right of application, any technology that other people complete
Entity or method also or a kind of equivalent change, will if identical with defined in the right of application
It is considered as being covered by among the right.
Claims (9)
1. a kind of flexible measurement method of rotary kiln burning zone temperature, which is characterized in that the method includes：With temperature of kiln head model
Ybase1With kiln end temperature model Ybase2For basic model, migrate to obtain burning zone temperature model Y by modelnew, to be burnt
At band temperature prediction；
Ybase1=f (Xbase1), Ybase2=f (Xbase2)
Ynew=f (Ybase1,Ybase2)。
2. the flexible measurement method of rotary kiln burning zone temperature as described in claim 1, which is characterized in that the temperature of kiln head mould
Type and kiln end temperature model are obtained according to field data using the modeling method of least squares support based on instant learning respectively.
3. the flexible measurement method of rotary kiln burning zone temperature as claimed in claim 2, which is characterized in that pass through instant learning
After (Just In Time, JIT) strategy selects sample, using least square method supporting vector machine (Least Square SVM,
LSSVM it) is modeled, obtains kiln end temperature and temperature of kiln head model.
4. the flexible measurement method of rotary kiln burning zone temperature as claimed in claim 3, which is characterized in that the instant learning plan
It slightly include step：
(a) similar samples selection；
(b) locally fine point is carried out using similar sample；
(c) library is updated the data.
5. a kind of flexible measurement method of rotary kiln burning zone temperature, which is characterized in that the method includes：
(1) it chooses auxiliary variable and robust multivariable pretreatment is carried out to the process data of acquisition；
(2) it after pretreated data being selected modeling sample by instant learning (Just In Time, JIT) strategy, uses
Least square method supporting vector machine (Least Square SVM, LSSVM) is modeled, and temperature of kiln head model and kiln end temperature are obtained
The model of model；
(3) it is basic model Y with temperature of kiln head model and kiln end temperature modelbase1And Ybase2, migrated and be burnt by model
Band temperature model YnewCarry out burning zone temperature prediction；
Ybase1=f (Xbase1), Ybase2=f (Xbase2)
Ynew=f (Ybase1,Ybase2)。
6. the flexible measurement method of rotary kiln burning zone temperature as claimed in claim 5, which is characterized in that for kiln end temperature mould
Type, the auxiliary variable are dore furnace tremie pipe temperature, kiln tail smoke-box pressure and kiln owner's electromechanics stream；For temperature of kiln head model,
The auxiliary variable is kiln owner's electromechanics stream, kiln hood one chamber pressure of air temperature and grate-cooler three times.
7. the flexible measurement method of rotary kiln burning zone temperature as claimed in claim 5, which is characterized in that the pre- place of the data
Reason includes carrying out abnormity point elimination to the data of acquisition；Each change is described using mahalanobis distance (Mahalanobis Distance)
Amount,
Covariance matrixes of the wherein C between input data,For the mean value of input data；
Chi square distribution is used to determine whether for outlier, ifThen it is considered outlier；IfThen it is considered normal point.
8. the flexible measurement method of rotary kiln burning zone temperature as claimed in claim 5, which is characterized in that the least square branch
Holding vector machine expression-form isWherein K (xi,xj) it is kernel function；Kernel function is radial base core letter
Number, K (xi,xj)=exp (- (x-y)2/2δ2)。
9. such as the flexible measurement method of claim 5-8 any one of them rotary kiln burning zone temperatures, which is characterized in that the side
Method further includes being corrected to burning zone temperature model using deviation correction method according to model predictive error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810923361.1A CN108647481A (en) | 2018-08-14 | 2018-08-14 | A kind of rotary kiln burning zone temperature flexible measurement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810923361.1A CN108647481A (en) | 2018-08-14 | 2018-08-14 | A kind of rotary kiln burning zone temperature flexible measurement method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647481A true CN108647481A (en) | 2018-10-12 |
Family
ID=63760828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810923361.1A Pending CN108647481A (en) | 2018-08-14 | 2018-08-14 | A kind of rotary kiln burning zone temperature flexible measurement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647481A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109282917A (en) * | 2018-11-01 | 2019-01-29 | 杭州质子科技有限公司 | A method of reducing human arm postural change influences oxter thermometric |
| CN110981240A (en) * | 2019-12-19 | 2020-04-10 | 华东理工大学 | Calcination process optimization method and system |
| CN111128312A (en) * | 2019-12-19 | 2020-05-08 | 湖南工业大学 | Zinc oxide volatilization kiln mixed modeling method based on mechanism and support vector machine |
| CN111950191A (en) * | 2020-07-07 | 2020-11-17 | 湖南大学 | Rotary kiln sintering temperature prediction method based on hybrid deep neural network |
| CN113177364A (en) * | 2021-05-21 | 2021-07-27 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere raceway |
| CN116523388A (en) * | 2023-04-17 | 2023-08-01 | 无锡雪浪数制科技有限公司 | Data-driven quality modeling method based on industrial Internet platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020010517A1 (en) * | 1998-12-17 | 2002-01-24 | Siemens Ag. | System of case-based reasoning for sensor prediction in a technical process, especially in a cement kiln, method and apparatus therefore |
| CN102629104A (en) * | 2011-12-01 | 2012-08-08 | 燕山大学 | Calcination predictive control system and method for rotary cement kiln |
| CN104634478A (en) * | 2015-03-06 | 2015-05-20 | 沈阳工业大学 | Soft measurement method for burning zone temperature of rotary kiln |
| CN104794263A (en) * | 2015-03-31 | 2015-07-22 | 济南大学 | Hybrid modeling method for rotary cement kiln burning zone temperature based on LS_SVM soft measurement |
| CN105159235A (en) * | 2015-01-08 | 2015-12-16 | 北方工业大学 | Comprehensive coordination control method and system for rotary kiln during calcining process |
-
2018
- 2018-08-14 CN CN201810923361.1A patent/CN108647481A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020010517A1 (en) * | 1998-12-17 | 2002-01-24 | Siemens Ag. | System of case-based reasoning for sensor prediction in a technical process, especially in a cement kiln, method and apparatus therefore |
| CN102629104A (en) * | 2011-12-01 | 2012-08-08 | 燕山大学 | Calcination predictive control system and method for rotary cement kiln |
| CN105159235A (en) * | 2015-01-08 | 2015-12-16 | 北方工业大学 | Comprehensive coordination control method and system for rotary kiln during calcining process |
| CN104634478A (en) * | 2015-03-06 | 2015-05-20 | 沈阳工业大学 | Soft measurement method for burning zone temperature of rotary kiln |
| CN104794263A (en) * | 2015-03-31 | 2015-07-22 | 济南大学 | Hybrid modeling method for rotary cement kiln burning zone temperature based on LS_SVM soft measurement |
Non-Patent Citations (4)
| Title |
|---|
| RONGXF123: "9煅烧过程异常情况与故障处理", 《HTTPS://WWW.DOC88.COM/P-8018615048186.HTML》 * |
| 张立 等: "基于模型迁移方法的回转窑煅烧带温度软测量", 《东北大学学报（自然科学版）》 * |
| 牛大鹏 等: "基于改进即时学习算法的湿法冶金浸出过程建模", 《化工学报》 * |
| 田甘露: "水泥烧成系统故障诊断技术研究", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅰ辑(月刊)》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109282917A (en) * | 2018-11-01 | 2019-01-29 | 杭州质子科技有限公司 | A method of reducing human arm postural change influences oxter thermometric |
| CN110981240A (en) * | 2019-12-19 | 2020-04-10 | 华东理工大学 | Calcination process optimization method and system |
| CN111128312A (en) * | 2019-12-19 | 2020-05-08 | 湖南工业大学 | Zinc oxide volatilization kiln mixed modeling method based on mechanism and support vector machine |
| CN110981240B (en) * | 2019-12-19 | 2022-04-08 | 华东理工大学 | Calcination process optimization method and system |
| CN111128312B (en) * | 2019-12-19 | 2022-06-17 | 湖南工业大学 | Zinc oxide volatilization kiln mixed modeling method based on mechanism and support vector machine |
| CN111950191A (en) * | 2020-07-07 | 2020-11-17 | 湖南大学 | Rotary kiln sintering temperature prediction method based on hybrid deep neural network |
| CN113177364A (en) * | 2021-05-21 | 2021-07-27 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere raceway |
| CN113177364B (en) * | 2021-05-21 | 2023-07-14 | 东北大学 | Soft measurement modeling method for temperature of blast furnace tuyere convolution zone |
| CN116523388A (en) * | 2023-04-17 | 2023-08-01 | 无锡雪浪数制科技有限公司 | Data-driven quality modeling method based on industrial Internet platform |
| CN116523388B (en) * | 2023-04-17 | 2023-11-10 | 无锡雪浪数制科技有限公司 | Data-driven quality modeling method based on industrial Internet platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647481A (en) | A kind of rotary kiln burning zone temperature flexible measurement method | |
| Yuan et al. | Multi‐similarity measurement driven ensemble just‐in‐time learning for soft sensing of industrial processes | |
| CN112380738B (en) | Rotary cement kiln combustion field reconstruction error compensation and optimization method, storage medium and system | |
| CN109935280B (en) | Blast furnace molten iron quality prediction system and method based on ensemble learning | |
| CN110378064A (en) | A method of the prediction dead stock column temperature of blast furnace furnace core | |
| CN106191350A (en) | Bottom house air port working condition appraisal procedure based on fixed point radar | |
| CN111254243A (en) | Method and system for intelligently determining iron notch blocking time in blast furnace tapping process | |
| WO2018038279A1 (en) | Energy saving method and fems for applying same | |
| Shi et al. | Key issues and progress of industrial big data-based intelligent blast furnace ironmaking technology | |
| CN101509812A (en) | Soft measurement method for billet temperature distribution in smelting and heating-furnace | |
| Anand et al. | Design of neural network based expert system for automated lime kiln system | |
| CN109033511A (en) | A kind of quality coal in cement kiln systems heat consumption analysis method of combined data driving and data mining | |
| CN111915080A (en) | Raw fuel cost optimal proportioning method based on molten iron quality constraint | |
| Wu et al. | Integrated soft sensing of coke-oven temperature | |
| CN102952912A (en) | Blast furnace smelting method and system based on expert system and knowledge base | |
| CN106594794B (en) | A kind of hybrid intelligent boiler efficiency burning optimization model update method | |
| US20050137995A1 (en) | Method for regulating a thermodynamic process by means of neural networks | |
| CN116523075A (en) | Cement clinker strength prediction method, device, electronic equipment and storage medium | |
| CN112859815B (en) | Method for monitoring and diagnosing abnormal furnace conditions in roasting process | |
| CN107832880B (en) | Blast furnace state variable prediction method based on material distribution parameters | |
| Lu et al. | Online prediction method of cement clinker f-Cao based on K-ELM | |
| WO2022059753A1 (en) | Method for generating trained prediction model that predicts energy efficiency of melting furnace, method for predicting energy efficiency of melting furnace, and computer program | |
| TWI534574B (en) | Prediction Method of Hot Metal Temperature in Blast Furnace | |
| Trofimov et al. | Generalized structure of an intelligent control system for technological objects and experience of its use | |
| Zhaogang et al. | Temperature estimation of cement rotary kiln based on extended Kalman filter |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181012 |