CN1605958A - Combined modeling method and system for complex industrial process - Google Patents
Combined modeling method and system for complex industrial process Download PDFInfo
- Publication number
- CN1605958A CN1605958A CN 200410009801 CN200410009801A CN1605958A CN 1605958 A CN1605958 A CN 1605958A CN 200410009801 CN200410009801 CN 200410009801 CN 200410009801 A CN200410009801 A CN 200410009801A CN 1605958 A CN1605958 A CN 1605958A
- Authority
- CN
- China
- Prior art keywords
- model
- modeling
- data
- module
- built
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention provides a combined modeling method and system for complex industrial process, wherein the modeling system comprises a model building data extraction module, a data processing module, a unit model extraction module, a unit model storage module, a combination model output module, a unit model building module, a complex model combination modeling module, a model estimation module, a model operation module and a modeling algorism library, by disintegrating the complex model into a plurality of integrated unit models, and employing algebraical combination of multiple-unit models, the purpose of complex process modeling can be achieved.
Description
Technical field
The invention belongs to the control field of industrial processes, a kind of compositional modeling method and system thereof of complex industrial process is provided especially, the coupled relation between effective modeling subprocess or the part process, the adaptability and the accuracy of raising model.
Background technology
The control of complex industrial process is one of field of enterprise and scientific research institution's primary study always, and it is the essential link that modern industry is produced.And modern industry control that is to say the mathematical model of process and the core that algorithm is process control thereof all based on model.The modeling of general industrial production run often relates to multi-field knowledge, modeling as metallurgical process often relates to Fluid Mechanics Computation, calculates thermal conduction study, calculates Combustion, metallurgical reaction engineering, statistics, artificial intelligence etc., therefore, the modeling of complex industrial process is the difficult point in process control field always, and Chinese scholars has been carried out a large amount of research in the industrial process modeling field.Document " adaptive fuzzy system and the application in industrial process modeling and control thereof " has been studied the application of fuzzy self-adaption technology at industrial process modeling, " research and the realization of Several Methods in industrial process modeling and the optimization " studied the RBF neural network in the catalytic cracking process Application in Modeling, " the application metallurgical automation 1990 of several Recursive Least Squares in industrial process modeling " studied the application of Recursive Least Squares in the control of boiler drum level adjustment process, " a kind of wavelet-neural net Multivariate Mixed process model and use the 11 process control scientific report can 2000 " have been studied wavelet-neural net and done application in the estimation at the Atmospheric Tower raw gasoline.These research work all are to use one or more modeling methods certain production run is carried out modeling, these models are all done as a whole appearance, this modeling pattern between complicated, that comprise a plurality of subprocess, the subprocess mutually the big industrial process of coupling often can not get good effect.
The general industrial process modeling approach mainly is to regard a process for the treatment of modeling as an integral body at present, selects for use mechanism model, statistical model, model of mind or triplicity that this process is carried out modeling then.But, because the industrial processes of a complexity are coupled by a plurality of subprocess or a plurality of part often or combine, though it is done as a whole modeling, can save the complicacy that is coupled between the model, but single model can't accurate description subprocess model or department pattern between coupled relation, the model that obtains often precision is not high, perhaps precision is much of that, less stable, along with the variation model of process can't be used, when especially the coupled relation between subprocess or part process changed, block mold can't these variations of effecting reaction.
Summary of the invention
The object of the present invention is to provide a kind of compositional modeling method and system thereof of complex industrial process, the solution block mold is ignored the coupling between subprocess or the part process, the problem that causes model accuracy and stability to reduce, propose a kind ofly a plurality of model of elements to be carried out algebraic combination or import an output being combined to form the method for built-up pattern, and to have proposed a cover be the built-up pattern development system of core with this method, the effective coupled relation between modeling subprocess or the part process improves the adaptability and the accuracy of model.
The present invention at first is decomposed into complex industrial process 2～10 unit, and each unit is set up model of element respectively, by the model combination technique 2～10 model of elements is made up, and forms the whole final mask of complex industrial production run at last.
System of the present invention is made up of modeling data extraction module, data processing module, model of element extraction module, model of element memory module, built-up pattern output module, model of element MBM, complex model compositional modeling module, model evaluation module, model running module and modeling algorithm storehouse.Concrete scheme comprises:
1, from relational database Oracle, SQL Server or text, EXCEL file, extract the required data of modeling by the modeling data extraction module,, be sent to data processing module as data such as temperature, pressure:
2, data processing module is handled the raw data that the modeling data extraction module is sent to, the processing that comprises processing that the null value value of averaging is substituted or rejects, abnormal data is rejected, to data carry out that average one standard is looked into or greatly-minimum standardized processing, when the variable number is very big, (generally surpass more than 50) to variable carry out the dimension yojan processing, data are carried out clustering processing, data are carried out at random or stratified sampling is handled, data are carried out the processing of various nonlinear transformations, the data after the processing are sent to the model of element MBM;
3, the data after the utilization of model of element MBM is handled, the algorithm that calls in the modeling algorithm storehouse carries out the modeling of model of element, modeling algorithm can adopt models of mind such as linear regression, non-linear regression, neural network, expertise, supporting vector machine, also can adopt the mechanism model of describing by partial differential equation, algebraic equation, after MBM calculates model, with the square error and the R of computation model
2Coefficient, so that the user carries out the assessment of model, if the user is dissatisfied to model result, can directly in the model of element MBM, change the parameter setting of model, the type of change model is set, again modeling need not the work of repeating data extraction and data processing module, and model of element need be assessed by the model evaluation module after setting up;
4, the model of element of model evaluation module to setting up assessed on new data set, calculates evaluate parameter: square error MSE, R
2, Mean Square Error ASE, the user can determine according to these parameters whether model of element meets the demands, if do not satisfy then return the first step and extract data again and carry out the unit modeling,, then model is kept in the hard disk by the model memory module if model meets the demands;
5, after model of element meets the demands, just can carry out the compositional modeling work of complex model.The first step, the composition of at first definite built-up pattern, complex model compositional modeling module determines according to the requirement of model physical significance complex model by which model of element is made up; Second step, determine the array mode of model, the output of model of element can be carried out the combination of any-mode by the relation of algebraically, and a kind of feasible mode is as built-up pattern=α
1* model of element 1 * model of element 2+ α
2* model of element 3 * model of element 4 * model of element 5+ α
3* model of element 6+ α
4, α wherein
1, α
2, α
3, α
4Be constant, need given; The 3rd step, the selected cell model, determined array mode after, select the model of element file stored for the model of element in each group item; In the 4th step, the built-up pattern input variable determines that the input variable of built-up pattern might repeat, as the x that is input as of built-up pattern in each model of element
1, x
2, x
3, x
4, the input variable of model of element 1 is x
1, x
2, model of element 2 input variable be x
2, x
3, x
4, model of element 3 input variable be x
1, x
3, model of element 4 input variable be x
4, therefore after the input variable of built-up pattern is determined, the input variable of each input variable and each model of element need be mated.Obtain final built-up pattern after coupling work is finished, assess by the model evaluation module;
6, the model evaluation module is assessed final built-up pattern, calculates evaluate parameter: square error MSE, R on test data set
2, Mean Square Error ASE, judge by the user whether built-up pattern reaches requirement, if model result do not meet the demands, then returned for the 5th step and carry out compositional modeling again; If meet the demands, be to leave in the hard disk with information package such as the parameter of built-up pattern, operation computing formula, input/output variable number types then by the dynamic link libraries of model running module invokes by the model output module, confession model running module invokes;
7, the model running module will be responsible for the practical application of built-up pattern, store the real-time data base or the text of real-time field data by connection, the real-time running data of the input variable of built-up pattern is read in the system, and the built-up pattern dynamic link libraries file that is stored in the hard disk by link calls corresponding built-up pattern, the real-time estimate output quantity.
Described model input variable is meant the operating state signal of various production equipments in the production run, as temperature value, force value, displacement, velocity amplitude etc., this tittle deposits in the storer after ripple, amplification, the mould/number conversion after obtaining through sensor measurement after filtration.
Store multiple Model Calculation formula in the described modeling algorithm storehouse, these algorithms all are that the form with various functions provides computing function to caller, mainly comprise the modeling algorithm that linear regression, non-linear regression, partial least squares regression, pivot recurrence, supporting vector recurrence, BP neural network, expertise reasoning etc. are commonly used in the algorithms library.
Described storer is for being equipped with 120G hard disk and 512M internal memory.
Described relational database is SQL Server2000 or Oracle9i.
Described text is for being the ASCII character file of suffix with txt or csv.
Described model of element is that complex model decomposes the least part that can carry out accurate modeling that the back forms, and can be decomposed into reference temperature T as temperature T
sWith transformation temperature Δ T, can set up T respectively
sModel of element and the model of element of Δ T.
Described built-up pattern is the block mold of complex industrial process, can make up the model that obtains by a plurality of model of elements.
The invention has the advantages that: in the modeling to complex industrial process, use single technology and model to be difficult to a complex process is described accurately, the model that obtains is deleterious after operating mode changes often, the modeling of having to carry out again this moment, the stability of model and accuracy all can not get effective assurance.The present invention is decomposed into a plurality of model of elements with the complex industrial process model, and respectively to each model of element modeling, carry out then arbitrarily " with ", the combination of " taking advantage of " mode, can obtain complicated arbitrarily built-up pattern.Because it is few that each model of element after decomposing relates to factor with respect to original complex model, model is simpler, and modeling is more prone to, and the model that obtains is also more accurate.After operating mode changes, only need adjust changing related model of element, the classic method efficient that contrasts whole model adjustment is higher, and the stability of model is also better.
Description of drawings
Fig. 1 is system's composition diagram of complex industrial process compositional modeling of the present invention.System mainly is made up of ten parts: data extraction module, data processing module, model of element extraction module, model of element memory module, built-up pattern output module, model of element MBM, complex model compositional modeling module, model evaluation module, model running module and modeling algorithm storehouse.
Fig. 2 is the system construction drawing of complex industrial process compositional modeling of the present invention.Data collecting card can installation system place computing machine on, also can be installed on the independent data gather computer, be delivered on the data collecting card by the process status signal of sensor production equipment, capture card is a digital signal with conversion of signals, deposits in the real-time data base or text in the storer.
Storer is the SCSI high speed hard-disk of 120G.
Database is a relational database SQL Server 2000 or Oracle 9i.
Data extraction module is extracted the modeling desired data from relational database or text, and sends data processing module to.
Data processing module with original data processing after, use for unit or built-up pattern MBM.Adopt to reject for assigning null data that place record, this column average value substitute, default value substitutes three kinds of disposal routes, for irrational data, as the age be negative value, numerical value defined by the user substitutes.Simultaneously, can carry out clustering processing before the modeling, to the modeling respectively of different classes; The process bigger to data volume can adopt the mode of stochastic sampling or stratified sampling to reduce the modeling workload.
Model of element modeling and built-up pattern modeling all need treated data that model is assessed, and call the various algorithms in the modeling algorithm storehouse simultaneously in modeling process.
Fig. 3 is the menu map of complex industrial process compositional modeling of the present invention system.The modeling process of whole built-up pattern is to adopt the man-machine conversation interactive approach to finish.The user can carry out various selections according to the information of oneself grasping, and has improved the efficient of modeling.
The working procedure figure of Fig. 4 complex industrial process compositional modeling system.
1) at first carry out the decomposition of complex model, this step is very big for the accuracy relation of final mask, model decompose rationally, final built-up pattern will have higher accuracy and stability.
2) data extraction module is extracted the needed raw data of modeling from database or text.
3) raw data need just can be used for modeling through data processing, and the BP neural network needs data through standardization, and linear regression, nonlinear regression model (NLRM) need between the data degree of correlation little, and the cluster of data is helped to improve modeling precision.
4) the model of element MBM is used the data after handling, set up the units corresponding model by calling different modeling algorithms, model of element can be differential equation mechanism model, algebraic equation mechanism model, linear regression model (LRM), nonlinear regression model (NLRM), offset minimum binary PLS model, pivot regression model, supporting vector recurrence SVR model, neural network model.
5) model of element needs the accuracy of model is assessed after setting up, and the model evaluation module need calculate the evaluate parameter of model on the modeling data collection: square error MSE, Mean Square Error ASE, R
2
6) can whether meet the demands by the identifying unit model according to the evaluate parameter that calculates,, then returned for the 4th step and carry out modeling again, the parameter in the time of can changing modeling or select another model form if do not meet the demands; If meet the demands, then the unit is stored in the model file, use during for compositional modeling.
7) repeated for 1～6 step, all model of elements are all set up, enter the built-up pattern modelling phase after meeting the demands.
8) enter the built-up pattern modelling phase, at first extract the needed data set of built-up pattern.
9) set up built-up pattern, method that built-up pattern is set up and step are referring to the explanation of Fig. 5.
10) after built-up pattern is set up and finished, need assess it, the model evaluation module calculates corresponding model parameter according to the compositional modeling data set: Mean Square Error ASE, R
2, if model accuracy do not meet the demands, then at first need to determine whether to be that the mode of compositional modeling needs to adjust, if do not need to adjust, then again model of element is adjusted; Meet the demands as if the built-up pattern precision, then output model.
11) final built-up pattern output form is the dynamic link libraries of needed information when having packed all model runnings, so that in the module of incorporation model operation easily, can reduce the complicacy of model running module greatly, improves its versatility.
12) the model running module is equivalent to a platform, and the various built-up patterns of building up serve as to support to use in the different software environment with this platform.
Fig. 5 is a compositional modeling method flow diagram of the present invention.The modeling method of built-up pattern has dual mode: algebraic combination, input-output combination.In the algebraic combination method, a plurality of model of elements are combined into built-up pattern with the form of algebraic equation, as built-up pattern=α
1* model of element 1 * model of element 2+ α
2* model of element 3 * model of element 4 * model of element 5+ α
3* model of element 6+ α
4, α wherein
1, α
2, α
3, α
4It is constant; In the another kind of input-output combined method, a plurality of model of elements are coupled in the mode of input=output, be an input of model of element 2 as the output of model of element 1, and the output of model of element 2 and model of element 3 are two inputs of model of element 4.
Therefore, at first need to determine to use any array mode when compositional modeling, for the algebraic combination mode, modeling method is:
1) the at first input of combinations of definitions model, output variable.
2) definition algebraic term number, the part of separating with "+" or "-" is one, in each a plurality of model of elements can be arranged, a plurality of model of element acquiescences in " are taken advantage of " connection with phase.
3) define the number of each middle model of element.
4) set each model of element, promptly determine the model file of each model of element.
5) extraction unit model from model file.
6) input variable of built-up pattern and the input variable of each model of element are mated, the input variable of built-up pattern can be the input variable of a plurality of model of elements.
7) after coupling is finished, whether the matching condition that needs to check model satisfies, whether the integrality that is built-up pattern satisfies, whether whether all input variables of built-up pattern all mate with the input variable of built-up pattern with the input coupling of model of element, all input variables of each model of element, if there is certain condition not satisfy, then returning for the 6th step mates again; If two conditions all satisfy, then built-up pattern is set up and is finished, and can carry out model evaluation.
For the input-output array mode, modeling method is:
1) the at first input of combinations of definitions model, output variable.
2) model of element number in the combinations of definitions model.
3) set each model of element, promptly determine the model file of each model of element.
4) extraction unit model from model file.
5) input, the output variable of each model of element are mated, determine the output of which model of element equates that with which input of another model of element the output of a model of element can be the input of other a plurality of model of elements simultaneously.
6) input variable of each model of element of input of built-up pattern is mated, the output variable of the output variable of output variable and certain model of element or plurality of units model (when built-up pattern has a plurality of output) is mated.The input variable of built-up pattern can be the input variable of a plurality of model of elements.
7) after coupling is finished, whether the matching condition that needs to check model satisfies, whether the integrality that is built-up pattern satisfies, whether all inputs of all model of elements in the built-up pattern, output variable all mate, the input variable of model of element can not be mated with the output variable of a plurality of other model of elements simultaneously, if there is certain condition not satisfy, then returning for the 5th step mates again; If two conditions all satisfy, then built-up pattern is set up and is finished, and can carry out model evaluation.
Claims (6)
1, a kind of compositional modeling method of complex industrial process, it is characterized in that: complex industrial process is decomposed into 2～10 unit, and each unit set up model of element respectively, by the model combination technique 2～10 model of elements are made up, form the whole final mask of complex industrial production run at last.
2, a kind of system that realizes the compositional modeling method of claim 1 complex industrial process is characterized in that: be made up of modeling data extraction module, data processing module, model of element extraction module, model of element memory module, built-up pattern output module, model of element MBM, complex model compositional modeling module, model evaluation module, model running module and modeling algorithm storehouse; Concrete scheme comprises:
A, from relational database Oracle, SQL Server or text, EXCEL file, extract the required data of modeling,, be sent to data processing module as data such as temperature, pressure by the modeling data extraction module;
B, data processing module are handled the raw data that the modeling data extraction module is sent to, the processing that comprises processing that the null value value of averaging is substituted or rejects, abnormal data is rejected, to data carry out that average-standard is looked into or greatly-minimum standardized processing, when variable outnumber 50 with the time, to variable carry out the dimension yojan processing, data are carried out clustering processing, data are carried out at random or stratified sampling is handled, data are carried out the processing of various nonlinear transformations, the data after the processing are sent to the model of element MBM;
Data after c, the utilization of model of element MBM are handled, the algorithm that calls in the modeling algorithm storehouse carries out the modeling of model of element, modeling algorithm adopts models of mind such as linear regression, non-linear regression, neural network, expertise, supporting vector machine, or employing is by the mechanism model of partial differential equation, algebraic equation description, after MBM calculates model, with the square error and the R of computation model
2Coefficient, so that the user carries out the assessment of model, if the user is dissatisfied to model result, the then directly parameter setting of change model in the model of element MBM, the type of change model is set, again modeling need not the work of repeating data extraction and data processing module, and model of element need be assessed by the model evaluation module after setting up;
D, the model of element of model evaluation module to setting up are assessed on new data set, calculate evaluate parameter: square error MSE, R
2, Mean Square Error ASE, the user can determine according to these parameters whether model of element meets the demands, if do not satisfy then return the first step and extract data again and carry out the unit modeling,, then model is kept in the hard disk by the model memory module if model meets the demands;
After e, model of element meet the demands, carry out the compositional modeling work of complex model: the first step, determine the composition of built-up pattern, complex model compositional modeling module determines according to the requirement of model physical significance complex model by which model of element is made up; Second goes on foot, and determines the array mode of model, and the output of model of element can be carried out the combination of any-mode by the relation of algebraically, a kind of feasible mode, as: built-up pattern=α
1* model of element 1 * model of element 2+ α
2* model of element 3 * model of element 4 * model of element 5+ α
3* model of element 6+ α
4, α wherein
1, α
2, α
3, α
4Be constant, need given; The 3rd step, the selected cell model, determined array mode after, select the model of element file stored for the model of element in each group item; In the 4th step, the built-up pattern input variable determines that the input variable of built-up pattern might repeat, as the x that is input as of built-up pattern in each model of element
1, x
2, x
3, x
4, the input variable of model of element 1 is x
1, x
2, model of element 2 input variable be x
2, x
3, x
4, model of element 3 input variable be x
1, x
3, model of element 4 input variable be x
4, therefore after the input variable of built-up pattern is determined, the input variable of each input variable and each model of element need be mated; Obtain final built-up pattern after coupling work is finished, assess by the model evaluation module;
F, model evaluation module are assessed final built-up pattern, calculate evaluate parameter: square error MSE, R on test data set
2, Mean Square Error ASE, judge by the user whether built-up pattern reaches requirement, if model result do not meet the demands, then returned for the 5th step and carry out compositional modeling again; If meet the demands, be to leave in the hard disk with information package such as the parameter of built-up pattern, operation computing formula, input/output variable number types then by the dynamic link libraries of model running module invokes by the model output module, confession model running module invokes;
G, model running module will be responsible for the practical application of built-up pattern, store the real-time data base or the text of real-time field data by connection, the real-time running data of the input variable of built-up pattern is read in the system, and the built-up pattern dynamic link libraries file that is stored in the hard disk by link calls corresponding built-up pattern, the real-time estimate output quantity.
3, according to the described system of claim 2, it is characterized in that: described model input variable is meant the operating state signal of various production equipments in the production run, comprise temperature value, force value, displacement, velocity amplitude etc., this tittle deposits in the storer after ripple, amplification, the mould/number conversion after obtaining through sensor measurement after filtration.
4, according to the described system of claim 2, it is characterized in that: store multiple Model Calculation formula in the described modeling algorithm storehouse, these algorithms all are that the form with various functions provides computing function to caller, mainly comprise the modeling algorithm that linear regression, non-linear regression, partial least squares regression, pivot recurrence, supporting vector recurrence, BP neural network, expertise reasoning etc. are commonly used in the algorithms library.
5, according to the described system of claim 2, it is characterized in that: described storer is for being equipped with 120G hard disk and 512M internal memory; Described relational database is SQL Server2000 or Oracle9i; Described text is for being the ASCII character file of suffix with txt or csv.
6, according to the described system of claim 2, it is characterized in that: described model of element is that complex model decomposes the least part that can carry out accurate modeling that the back forms, and can be decomposed into reference temperature T as temperature T
sWith transformation temperature Δ T, can set up T respectively
sModel of element and the model of element of Δ T.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100098010A CN1300649C (en) | 2004-11-16 | 2004-11-16 | Combined modeling method and system for complex industrial process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100098010A CN1300649C (en) | 2004-11-16 | 2004-11-16 | Combined modeling method and system for complex industrial process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1605958A true CN1605958A (en) | 2005-04-13 |
| CN1300649C CN1300649C (en) | 2007-02-14 |
Family
ID=34763097
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100098010A Expired - Fee Related CN1300649C (en) | 2004-11-16 | 2004-11-16 | Combined modeling method and system for complex industrial process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1300649C (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100371938C (en) * | 2005-06-03 | 2008-02-27 | 冶金自动化研究设计院 | Quality design method under minute new aluminium sample |
| CN102034005A (en) * | 2010-12-17 | 2011-04-27 | 中铁十四局集团有限公司 | Shield machine posture simulation detection system for shield tunnel construction |
| CN101587328B (en) * | 2008-05-23 | 2011-12-14 | 朱豫才 | Device for identifying non-linear process dynamic model |
| CN102479266A (en) * | 2010-11-26 | 2012-05-30 | 香港理工大学 | Method for generating structural surface by polishing |
| CN103279626A (en) * | 2013-06-17 | 2013-09-04 | 南车株洲电力机车有限公司 | Method and system for simulation modeling |
| CN105045764A (en) * | 2015-08-11 | 2015-11-11 | 精硕世纪科技（北京）有限公司 | Method and system for acquiring input parameters of model cluster |
| CN108491626A (en) * | 2013-05-09 | 2018-09-04 | 罗克韦尔自动化技术公司 | Carry out system, method and the storage medium of the simulation of industrial automation system |
| WO2019114421A1 (en) * | 2017-12-15 | 2019-06-20 | 阿里巴巴集团控股有限公司 | Model integration method and device |
| CN109966996A (en) * | 2019-02-26 | 2019-07-05 | 武汉恒力华振科技有限公司 | A kind of system using the melting state in big data analysis prediction hot melt adhesive production process |
| CN111948992A (en) * | 2020-08-05 | 2020-11-17 | 上海微亿智造科技有限公司 | Method and system for performing multistage progressive modeling on industrial batch type big data |
| CN112021626A (en) * | 2020-07-10 | 2020-12-04 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| US11243505B2 (en) | 2015-03-16 | 2022-02-08 | Rockwell Automation Technologies, Inc. | Cloud-based analytics for industrial automation |
| US11409251B2 (en) | 2015-03-16 | 2022-08-09 | Rockwell Automation Technologies, Inc. | Modeling of an industrial automation environment in the cloud |
| US11470157B2 (en) | 2012-02-09 | 2022-10-11 | Rockwell Automation Technologies, Inc. | Cloud gateway for industrial automation information and control systems |
| US11513477B2 (en) | 2015-03-16 | 2022-11-29 | Rockwell Automation Technologies, Inc. | Cloud-based industrial controller |
| US11676508B2 (en) | 2013-05-09 | 2023-06-13 | Rockwell Automation Technologies, Inc. | Using cloud-based data for industrial automation system training |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1995014281A1 (en) * | 1993-11-15 | 1995-05-26 | Siemens Aktiengesellschaft | Method of automatically modelling part of an overall process by means of a computer |
| JP2000081906A (en) * | 1998-09-04 | 2000-03-21 | Mitsubishi Electric Corp | Virtual plant simulation device and its method |
| US7117480B2 (en) * | 2001-11-27 | 2006-10-03 | 3M Innovative Properties Company | Reusable software components for invoking computational models |
-
2004
- 2004-11-16 CN CNB2004100098010A patent/CN1300649C/en not_active Expired - Fee Related
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100371938C (en) * | 2005-06-03 | 2008-02-27 | 冶金自动化研究设计院 | Quality design method under minute new aluminium sample |
| CN101587328B (en) * | 2008-05-23 | 2011-12-14 | 朱豫才 | Device for identifying non-linear process dynamic model |
| CN102479266B (en) * | 2010-11-26 | 2014-10-22 | 香港理工大学 | Method for generating structural surface by polishing |
| CN102479266A (en) * | 2010-11-26 | 2012-05-30 | 香港理工大学 | Method for generating structural surface by polishing |
| CN102034005B (en) * | 2010-12-17 | 2012-12-05 | 中铁十四局集团有限公司 | Shield machine posture simulation detection system for shield tunnel construction |
| CN102034005A (en) * | 2010-12-17 | 2011-04-27 | 中铁十四局集团有限公司 | Shield machine posture simulation detection system for shield tunnel construction |
| US11470157B2 (en) | 2012-02-09 | 2022-10-11 | Rockwell Automation Technologies, Inc. | Cloud gateway for industrial automation information and control systems |
| US11676508B2 (en) | 2013-05-09 | 2023-06-13 | Rockwell Automation Technologies, Inc. | Using cloud-based data for industrial automation system training |
| CN108491626A (en) * | 2013-05-09 | 2018-09-04 | 罗克韦尔自动化技术公司 | Carry out system, method and the storage medium of the simulation of industrial automation system |
| CN108491626B (en) * | 2013-05-09 | 2022-04-08 | 罗克韦尔自动化技术公司 | System, method and storage medium for performing simulations of industrial automation systems |
| US11295047B2 (en) | 2013-05-09 | 2022-04-05 | Rockwell Automation Technologies, Inc. | Using cloud-based data for industrial simulation |
| CN103279626A (en) * | 2013-06-17 | 2013-09-04 | 南车株洲电力机车有限公司 | Method and system for simulation modeling |
| CN103279626B (en) * | 2013-06-17 | 2017-03-15 | 南车株洲电力机车有限公司 | A kind of Building of Simulation Model method and system |
| US11243505B2 (en) | 2015-03-16 | 2022-02-08 | Rockwell Automation Technologies, Inc. | Cloud-based analytics for industrial automation |
| US11409251B2 (en) | 2015-03-16 | 2022-08-09 | Rockwell Automation Technologies, Inc. | Modeling of an industrial automation environment in the cloud |
| US11927929B2 (en) | 2015-03-16 | 2024-03-12 | Rockwell Automation Technologies, Inc. | Modeling of an industrial automation environment in the cloud |
| US11880179B2 (en) | 2015-03-16 | 2024-01-23 | Rockwell Automation Technologies, Inc. | Cloud-based analytics for industrial automation |
| US11513477B2 (en) | 2015-03-16 | 2022-11-29 | Rockwell Automation Technologies, Inc. | Cloud-based industrial controller |
| CN105045764A (en) * | 2015-08-11 | 2015-11-11 | 精硕世纪科技（北京）有限公司 | Method and system for acquiring input parameters of model cluster |
| TWI745623B (en) * | 2017-12-15 | 2021-11-11 | 開曼群島商創新先進技術有限公司 | Model integration method and device |
| WO2019114421A1 (en) * | 2017-12-15 | 2019-06-20 | 阿里巴巴集团控股有限公司 | Model integration method and device |
| US11132624B2 (en) | 2017-12-15 | 2021-09-28 | Advanced New Technologies Co., Ltd. | Model integration method and device |
| CN109966996A (en) * | 2019-02-26 | 2019-07-05 | 武汉恒力华振科技有限公司 | A kind of system using the melting state in big data analysis prediction hot melt adhesive production process |
| CN112021626A (en) * | 2020-07-10 | 2020-12-04 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| CN112021626B (en) * | 2020-07-10 | 2021-08-17 | 张家口卷烟厂有限责任公司 | Intelligent control system and method for tobacco shred making link |
| CN111948992A (en) * | 2020-08-05 | 2020-11-17 | 上海微亿智造科技有限公司 | Method and system for performing multistage progressive modeling on industrial batch type big data |
| CN111948992B (en) * | 2020-08-05 | 2021-09-10 | 上海微亿智造科技有限公司 | Method and system for performing multistage progressive modeling on industrial batch type big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1300649C (en) | 2007-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1300649C (en) | Combined modeling method and system for complex industrial process | |
| CN103927346B (en) | Query connection method on basis of data volumes | |
| JP6784780B2 (en) | How to build a probabilistic model for large-scale renewable energy data | |
| CN103488537A (en) | Method and device for executing data ETL (Extraction, Transformation and Loading) | |
| CN1945482A (en) | Online energy source predicting system and method for integrated iron & steel enterprise | |
| CN112016828B (en) | Industrial equipment health management cloud platform architecture based on streaming big data | |
| CN105893669A (en) | Global simulation performance predication method based on data digging | |
| CN110619419B (en) | Passenger flow prediction method for urban rail transit | |
| CN114282725A (en) | Construction of transient oil reservoir agent model based on deep learning and oil reservoir prediction method | |
| CN112308298A (en) | Multi-scenario performance index prediction method and system for semiconductor production line | |
| CN115221793A (en) | Tunnel surrounding rock deformation prediction method and device | |
| CN111626497B (en) | People flow prediction method, device, equipment and storage medium | |
| CN116599857A (en) | Digital twin application system suitable for multiple scenes of Internet of things | |
| CN113434569B (en) | Ship propulsion device performance analysis system for monitoring health state of real ship equipment | |
| CN108717444A (en) | A kind of big data clustering method and device based on distributed frame | |
| CN113111588A (en) | NO of gas turbineXEmission concentration prediction method and device | |
| CN111680846A (en) | Simplified width learning system based on L1 and L2 norms | |
| CN112734107A (en) | Product collaborative design task data decomposition and distribution method based on error inverse propagation | |
| CN112380783B (en) | Industrial big data real-time acquisition and analysis method based on three-dimensional modeling simulation software | |
| CN117575542B (en) | Building engineering data control system and method based on modularized assembly | |
| CN1687935A (en) | Quality design method under minute new aluminium sample | |
| CN114297946B (en) | Industrial internet platform for realizing multidisciplinary simulation model order reduction | |
| CN114676522B (en) | Pneumatic shape optimization design method, system and equipment integrating GAN and migration learning | |
| CN114648014B (en) | Engineering data correlation method based on improved Gaussian mixture model | |
| CN115392543B (en) | Injection product quality prediction method combining L21 norm and residual cascade width learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20070214 Termination date: 20141116 |
|
| EXPY | Termination of patent right or utility model |