CN109492059A - A kind of multi-source heterogeneous data fusion and Modifying model process management and control method - Google Patents
A kind of multi-source heterogeneous data fusion and Modifying model process management and control method Download PDFInfo
- Publication number
- CN109492059A CN109492059A CN201910003356.3A CN201910003356A CN109492059A CN 109492059 A CN109492059 A CN 109492059A CN 201910003356 A CN201910003356 A CN 201910003356A CN 109492059 A CN109492059 A CN 109492059A
- Authority
- CN
- China
- Prior art keywords
- template
- model
- feature
- fusion
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 440
- 230000008569 process Effects 0.000 title claims abstract description 182
- 230000004927 fusion Effects 0.000 title claims abstract description 82
- 238000007499 fusion processing Methods 0.000 claims abstract description 32
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000005516 engineering process Methods 0.000 claims abstract description 16
- 238000013461 design Methods 0.000 claims abstract description 12
- 238000004088 simulation Methods 0.000 claims description 13
- 238000005457 optimization Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 9
- 230000008901 benefit Effects 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 3
- 238000003745 diagnosis Methods 0.000 claims description 3
- 230000036541 health Effects 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000003339 best practice Methods 0.000 claims 1
- 230000000903 blocking effect Effects 0.000 claims 1
- 241001269238 Data Species 0.000 abstract description 2
- 238000007500 overflow downdraw method Methods 0.000 abstract description 2
- 239000000725 suspension Substances 0.000 description 8
- 238000012938 design process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002715 modification method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of multi-source heterogeneous data fusion disclosed by the invention and Modifying model process management and control method, belong to field of automation technology.The present invention is directed to multi-layer complex engineering system, according to data, the characteristics of model and process, process is further subdivided into subprocess, define the process template of each subprocess, including data template, feature templates, method template and process template, pass through a variety of flexible configurations of each template, it establishes and adapts to a variety of isomeric datas, the link of data and information handled by the fusion process of model and fusion method, it matches to obtain optimal method according to inference machine and is merged and verified, keep fusion and the Modifying model link of test data more efficient, and it can be improved the precision of model after multi-source heterogeneous data fusion amendment, improve multi-layer complex engineering System Engineering Design efficiency, the precision of prediction that multi-layer complex engineering system is applied to corresponding field of engineering technology is improved using model after the amendment of multi-source heterogeneous data fusion, solve correlation engineering Technical problem.
Description
Technical field
The present invention relates to a kind of multi-source heterogeneous data fusions and Modifying model process management and control method, belong to automatic technology neck
Domain.
Background technique
The systems such as continental rise plateform system, aerospace system, complex mechanical system are normal in being designed into scheme determination process
Determining to designed system progress parameter by the mode of theoretical modeling, Computer Simulation modeling and test sampling analysis,
Performance optimization, soundness verification etc..And it usually can be to some links for the ease of solution when the modeling of theoretical model, simulation model
Simplified, causes degree of fitting between solving result and real response poor.Experimental test process is being carried out to model machine at the same time
In often due to cause multiple groups test data inconsistent in the presence of the error for being difficult to eliminate, and exist and cause since test number (TN) increases
The excessive situation of sample size.Theoretical, simulation model and test data can often be merged, be corrected in design process, with
Precision is improved, however since test data merges, multi-source heterogeneous data and model, make it difficult to involved in Modifying model process
A whole set of data fusion and Modifying model process are carried out to a design process with conventional method, therefore designed a set of towards multi-source
The fusion management-control method of isomeric data and model is very necessary.
Towards multi-source heterogeneous data, the method for Model Fusion, there are following technical problems in the prior art: current be directed to sets
Variety classes during meter, the data of different storage forms, are difficult to be managed collectively between model different modeling patterns, and there are orphans
Island phenomenon.To more independent between theoretical, simulation model amendment and the link handled test data, and usually lean on
Experience chooses the more method for adapting to different scenes existing for a certain fusion link, and therefore, it is difficult to efficient, flexible, quasi-
Really multi-source heterogeneous data, model are modified and are applied, leads to continental rise plateform system, aerospace system, complicated machinery
Modeling amendment Link Efficiency is lower in the multi-layers complex engineering system such as system design R&D process, and selecting modification method is not most
Excellent, model accuracy is to be improved.
Summary of the invention
It is asked for above-mentioned technology in the background technology is deposited towards multi-source heterogeneous data, the method for Model Fusion in the prior art
Topic, a kind of multi-source heterogeneous data fusion disclosed by the invention are with Modifying model process management and control method technical problems to be solved:
For multi-layer complex engineering system, the characteristics of according to data, model and process, process is further subdivided into subprocess,
The process template of each subprocess, including data template, feature templates, method template and process template are defined, each template is passed through
A variety of flexible configurations, foundation can adapt to data handled by the fusion process of a variety of isomeric datas, model and fusion method and letter
The link of breath, and match to obtain optimal method according to inference machine and merged and verified, thus make the fusion of test data with
And Modifying model link is more efficient, and can be improved the precision of model after multi-source heterogeneous data fusion amendment, and then improve multilayer
It is complicated to improve multi-layer using model after the multi-source heterogeneous data fusion amendment of foundation for grade complex engineering System Engineering Design efficiency
Engineering system is applied to the precision of prediction of corresponding field of engineering technology, solves correlation engineering technical problem.
The multi-layer complex engineering system includes continental rise plateform system, aerospace system, complex mechanical system.It is described
Multi-layer complex engineering system common feature: including multiple subsystem models, there are the inferior spies of non-linear or high-order for model itself
Property.
When the present invention is applied to continental rise plateform system, can be improved domain suspended model, road surface model, body construction has
Meta-model and the simulation accuracy of other continental rise platform models are limited, emulation cycle is reduced.Model after being improved using precision can be with
Preferably instruct the design of Structural Parameters of continental rise platform, continental rise Platform Vibration characteristic optimizing, handling index optimization and seating
Comfort index optimization, is with a wide range of applications and benefit.
When the present invention is applied to aerospace system, aerospace platform Each part finite element mould can be improved
Type, the simulation accuracy of aerodynamics model and other models reduce emulation cycle.Model after being improved using precision can be more
The optimization of each subsystem structure and aerodynamic properties optimization for instructing aerospace system well, are with a wide range of applications
With benefit.
When the present invention is applied to complex mechanical system, cutter, machine tool body structure model, thermodynamics mould can be improved
Type, fault analysis model, the simulation accuracy of status monitoring model reduce emulation cycle.Model after being improved using precision can be with
Preferably instruct the Structural Design of complex mechanical system, moreover it is possible to be health monitoring, fault diagnosis and the knot of Time variable structure
Structure vibration control etc. application provides strong support, is with a wide range of applications.
Object of the present invention is to what is be achieved through the following technical solutions.
A kind of multi-source heterogeneous data fusion disclosed by the invention and Modifying model process management and control method, include the following steps:
Step 1: the subprocess list that the data fusion of multi-layer complex engineering system, Modifying model main-process stream are covered
Memberization, modularization, building are suitable for combining the process template of configuration, and forming process template library, realize from user demand, are corrected
The feature of object is that the fusion process of guiding configures on demand, flexible combination, to realize the fusion process control of flexibility.
Correspond to technical field according to multi-layer complex engineering system, to by the data fusion of multi-layer complex engineering system,
Modifying model main-process stream is split, and forms each process unit, and by process unit and corresponding technical method and
The information that process application scenarios are included is integrated, building process template, and forms fusion process by abundant process template
Template library, realize from user demand, be corrected object feature be oriented to fusion process configure on demand, flexible combination, with reality
The fusion process control of existing flexibility.
The fusion process template mentioned in step 1 is by procedural information dictionary, process feature explanation, independent method template with
Four part of additional characteristic feature template is constituted.Procedural information dictionary includes the essential information of the process, and the essential information includes process
Template name, template number, date created;Process feature illustrates to include the head node of the process template and the number of tail node
Number, process function explanation carry out template selection for user and refer to sequence, and the head node and tail node are number
Number and association mode be it is one-to-many, the function declaration be explanatory note.Independent method template include in method template library with
The relevant all method templates of the process;Additional characteristic feature template includes feature templates relevant to the process in feature templates library.
Step 2: according to each template type of fusion process template and technical field, the field is found for difference
Typical technology method under problem scenario, by modifying to corresponding typical technology method, encapsulating generation method solver,
And corresponding typical technology method relevant information construction method template is extracted, the relevant information includes input/output information.It is logical
The matching of inference machine implementation method template and process template constructed by step 5 is crossed, realizes the flexible tune of method in fusion process
With, on demand configuration with rationally application.
Contained content is by method dictionary of information, method characteristic explanation, method solver three in method template in step 2
Part is constituted.Method dictionary of information essential information includes the method template title of the process, template number, date created;Method
Feature templates include the application message and application scenarios feature of this method, and the application message includes the input of method and defeated
Out, the application scenarios feature includes data dimension, sample average, data variance domain, data capacity, data type, and model is imitative
The coefficient of determination of true result and test value, model parameter number, the model solution time, whether model is non-linear, model order,
Types of models, model field;Method solver is the computing module of method template, is responsible for converting output for input.
Step 3: according in fusion process when choosing method template during required data characteristics and Modifying model
The model information building data characteristics template that method template needs when choosing.By the initial data of importing and model according to character modules
Plate carries out the calculating and extraction of characteristic value, forms multiple characteristic sets.
Feature templates described in step 3 are mainly by feature templates dictionary of information, associated process template information and described
Characteristic value three parts are constituted.Feature templates dictionary of information includes the number title of the feature templates, template number, template class
Type, date created;Association process template embodies the one-to-one incidence relation between feature templates and a certain process template;It is described
Characteristic value is the characteristic value needed when a certain process template is matched with method template, needs matched feature in data characteristics template
Value includes data type, data capacity, data dimension, sample average, data variance, the characteristic value packet that aspect of model template needs
Include types of models, model field, the input/output format of model, model parameter number, model solution speed, model emulation result
Multiple correlation coefficient.The data and model of importing are passed through by extracting feature with the selected associated feature templates of n process template
The n group feature set selects n method template associated with process template.
Step 4: feature templates and method template are carried out matched inference machine by building, choose optimal side for step 7
Method template.
Step 4 concrete methods of realizing is as follows:
Step 4.1: all methods in the MPi in method template set belonging to process template i are calculated and choose probability,
The i is process template number calculation formula such as following formula (1)-(3).
MPi=MPi1, MPi2 ..., MPin }, (i=1,2 ..., n) (1)
Wherein, P (MPin) is the selection probability of method template n, P (Cviz) it is the spy extracted according to feature templates Ctemk
It collects in Cvi under the value condition of feature Cviz, chooses the probability of this method template, ωzIt is P (Cviz) weight, reflect the spy
Sign probability chooses significance level to method and meets formula constraint.Characteristic value collection in the feature templates be Cvi=Cvi1,
Cvi2 ..., Cvin }, (z=1,2 ..., n), i is characterized template number.
Step 4.2: calculating the corresponding this method template point of each feature in feature set Cvi and choose probability P (Cviz), it calculates
Such as formula (4).
P(Cviz)=Ppri (Cviz)*Sim(Mviz,Cviz), z=1,2 ..., n (4)
Mviz is the z in the recommended characteristics collection Mvi={ Mvi1, Mvi2 ..., Mvin } of a certain method template in above formula
A feature, α are normalization coefficient, Sim (Mviz,Cviz) it is right therewith in the recommended characteristics and feature templates of a certain method template
The degree of fitting of feature is answered, and meets formula constraint.
Step 4.3: by step 4.1,4.2 building inference machines in parameter and degree of fitting judgement schematics in each item
Weight, realize and carry out relating value between the feature extraction of data/model, method template and feature templates according to feature templates
It solves, to realize the combination configuration between each template, i.e. feature templates and method template are carried out matched inference machine by building.
With the growth of fusion number, fusion process form assembly, the corresponding method template of each process template and method template and special
The combination case sample size of sign template is gradually increased, and the order of accuarcy of prior probability in inference machine is improved, to improve inference machine
Inferential capability.
Step 5: according to the data fusion and Modifying model main-process stream of user demand and multi-layer complex engineering system
It needs, n template is chosen from the fusion process template library constructed in step 1, configuration group synthetic population merges template, is formed
The main-process stream mixing operation is from beginning last complete procedure stream.
Step 5.1: according to user demand and the data fusion and Modifying model main-process stream of multi-layer complex engineering system
Needs, the data fusion class or Modifying model class being located in process subtemplate library;
Step 5.2: processing data are selected from process subtemplate library, the process unit template that model needs;
Step 3: the process subtemplate selected according to step 5.2 sorts according to the fusion process in relevant art field,
It is combined into main-process stream, it is that data or model import subtemplate that main-process stream, which starts node, and end node is export subtemplate.Process mould
Plate set expression is PTEM={ Ptem1, Ptem2 ..., Ptemn }, that is, forms the main-process stream mixing operation from beginning the complete of end
It has suffered journey stream.
Step 6: feature templates corresponding to the process template chosen using step 5 are to by data to be processed, model
The characteristic value described in step 5 extracts, and obtains corresponding to source data or the source model of epicycle data fusion or Modifying model
N group feature, for the method matching in guiding step seven.
Step 6.1: being corresponded with step 5 process template and feature templates, therefore determine process template in step 5
In the case where, n feature templates also determine therewith.Feature templates set expression be CTEM=Ctemi1, Ctemi2 ...,
Ctemin }, in is n feature templates number.
Step 6.2: characteristics extraction is carried out to data or model.In the n feature templates that step 6.1 obtains, Ge Gemo
The characteristic value number and type that plate is included difference.The characteristic value collection that feature templates i is included is expressed as according to each
The type for the characteristic value for including in a template calculates the data of importing, model, obtains data, the model pair of this fusion
The n group feature answered.
Step 7: using the inference machine in step 4, the n group feature set extracted according to step 6 is true from step 5
The concentration of method template corresponding to fixed process template is screened, and matching obtains method template corresponding with each process template,
And then the information link that multi-layer complex engineering system corresponds to the data fusion of technical field, Modifying model main-process stream is established, from
And keep the fusion of test data and Modifying model link more efficient, and can be improved model after multi-source heterogeneous data fusion amendment
Precision.
Step 7.1: recalled from method template library first each process template corresponding method template collection MPi=MPi1,
MPi2 ..., MPit }, (i=1,2 ..., n), wherein n is the number of process template, and t is the corresponding method of i-th of process template
The number of template centralized way.
Step 7.2: and then inference machine is used, by the feature set obtained with the associated feature templates of process template and the process
Each method that method template corresponding to template is concentrated matches one by one, obtains the selection probability of each method.I-th of process
Method corresponding to template concentrates the selection probability of each method to be expressed as: P (MPj), (j=1,2 ..., n).
Step 7.3: the method template for choosing maximum probability is concentrated from the corresponding method of process template, to selection maximum probability
Method template judged, it is when meeting default use demand, the method template of the maximum probability of selection is corresponding
Process template is associated, and the method template of the maximum probability with selection is carried out to the characteristic value for including in matched feature templates
It is stored in historical data base, and updates prior probability Ppri (MPj), (j=1,2 ..., t);When the method template of the maximum probability
When not meeting default use demand, the method for the maximum probability of selection is concentrated from the method for step 7.1 and is rejected, and return step
7.3。
Prior probability Ppri described in step 7.3 is that successful match is crossed in history fusion process by certain method template
The feature set statistics that feature templates are included obtains, embodies the feature that this method template is most suitable for the data of processing.Priori is general
The factor that rate will be chosen in inference machine as method template.After feature templates and method template successful match, feature templates
The feature set for including is stored into historical data base, and prior probability is calculated by formula (7).
Ppri=n/N (7)
In formula, n is the number that the value of this feature occurred in history, and N is total history number.
Step 7.4: each template in process template set PTEM={ Ptem1, Ptem2 ..., Ptemn } is walked
Rapid 7.1~7.3, matching obtains method template corresponding with each process template and then establishes the corresponding skill of multi-layer complex engineering system
Data fusion, the information link of Modifying model main-process stream in art field, to make fusion and the Modifying model ring of test data
Save precision that is more efficient, and can be improved model after multi-source heterogeneous data fusion amendment.
Step 8: it matches to obtain method template corresponding with each process template using step 7 and then establishes multi-layer complexity
Engineering system corresponds to the information link of the data fusion of technical field, Modifying model main-process stream, and then improves multi-layer complexity work
Project system engineering design efficiency improves multi-layer complex engineering system using model after the multi-source heterogeneous data fusion amendment of foundation
Applied to the precision of prediction of corresponding field of engineering technology, correlation engineering technical problem is solved.
The multi-layer complex engineering system includes continental rise plateform system, aerospace system, complex mechanical system.It is described
Multi-layer complex engineering system common feature: including multiple subsystem models, there are the inferior spies of non-linear or high-order for model itself
Property.
When the present invention is applied to continental rise plateform system, can be improved domain suspended model, road surface model, body construction has
Meta-model and the simulation accuracy of other continental rise platform models are limited, emulation cycle is reduced.Model after being improved using precision can be with
Preferably instruct the design of Structural Parameters of continental rise platform, continental rise Platform Vibration characteristic optimizing, handling index optimization and seating
Comfort index optimization, is with a wide range of applications and benefit.
When the present invention is applied to aerospace system, aerospace platform Each part finite element mould can be improved
Type, the simulation accuracy of aerodynamics model and other models reduce emulation cycle.Model after being improved using precision can be more
The optimization of each subsystem structure and aerodynamic properties optimization for instructing aerospace system well, are with a wide range of applications
With benefit.
When the present invention is applied to complex mechanical system, cutter, machine tool body structure model, thermodynamics mould can be improved
Type, fault analysis model, the simulation accuracy of status monitoring model reduce emulation cycle.Model after being improved using precision can be with
Preferably instruct the Structural Design of complex mechanical system, moreover it is possible to be health monitoring, fault diagnosis and the knot of Time variable structure
Structure vibration control etc. application provides strong support, is with a wide range of applications.
The utility model has the advantages that
1, a kind of multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method, to what is established in design process
The resources such as data obtained in model, test are integrated, and continental rise plateform system, aerospace system, complicated machine are effectively solved
Data silo problem of the multi-layers complex engineering system such as tool system during modeling and simulation.
2, a kind of multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method, data fusion, model are repaired
Positive whole process is split, and definition standard is packaged into process template, realize multi-layer complex engineering system data fusion and
Modifying model process configures combination on demand, improves resource centralization and reuse rate.
3, a kind of multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method, to test data, model into
Row feature extraction provides effectively reference and credible evaluation for data and model analysis.
4, a kind of multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method, repair data fusion, model
The method of each subprocess is automatically determined from method base during just, and method and each subprocess is associated, realization side
Method efficiently, flexible utilization, eliminate on data fusion and the method choice during Modifying model of multi-layer complex engineering system
Empirical.
5, a kind of multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method, the inference machine of building pass through spy
Derivation method is levied, and priori data is introduced wherein, improves the intelligent level during data fusion and Modifying model.
Detailed description of the invention
A kind of multi-source heterogeneous data fusion of Fig. 1 present invention and Modifying model process management and control method general flow chart;
The building process template of a kind of multi-source heterogeneous data fusion of Fig. 2 present invention and Modifying model process management and control method is illustrated
Figure;
The construction method template of a kind of multi-source heterogeneous data fusion of Fig. 3 present invention and Modifying model process management and control method is illustrated
Figure;
The construction feature template of a kind of multi-source heterogeneous data fusion of Fig. 4 present invention and Modifying model process management and control method is illustrated
Figure;
A kind of inference machine Computing Principle of multi-source heterogeneous data fusion and Modifying model process management and control method of Fig. 5 present invention
Figure;
A kind of Fig. 6 multi-source heterogeneous data fusion of the present invention and Modifying model process management and control method needed according to user and
Fusion demand configures forming process flow diagram;
The feature set of a kind of multi-source heterogeneous data fusion of Fig. 7 present invention and Modifying model process management and control method, which is extracted, to be illustrated
Figure;
The method template and process mould of a kind of multi-source heterogeneous data fusion of Fig. 8 present invention and Modifying model process management and control method
Plate is associated with to form information link schematic diagram.
Specific embodiment
Objects and advantages in order to better illustrate the present invention with reference to the accompanying drawing do further summary of the invention with example
Explanation.
Embodiment 1:
As shown in Figure 1, a kind of multi-source heterogeneous data fusion and Modifying model process management and control method disclosed in the present embodiment, tool
Body realizes that steps are as follows:
Step 1: the subprocess list that the data fusion of multi-layer complex mechanical system and Modifying model main-process stream are covered
Memberization, modularization, building are suitable for combining the process template of configuration, and forming process template library such as Fig. 2.
The fusion process template mentioned in step 1 is by procedural information dictionary, process feature explanation, independent method template with
Four part of additional characteristic feature template is constituted.By taking the process template of entitled parameter generating process as an example, listing the template is included
Information such as table 1.
1 parameter generating process template information table of table
Step 2: according to each template type of fusion process template and technical field, the field is found for difference
Typical technology method under problem scenario, construction method template library.
Contained content is by method dictionary of information, method characteristic explanation, method solver three in method template in step 2
Part is constituted.By taking the process template of entitled parameter generating process as an example, method template collection and each method which includes
The information of template such as Fig. 3.
Step 3: according in fusion process when choosing method template during required data characteristics and Modifying model
The model information building data characteristics template library that method template needs when choosing.
Feature templates described in step 3 are associated process template information and the characteristic value three by dictionary of information
Divide and constitutes.Information such as Fig. 4 that feature templates and feature templates corresponding to each process template are responsible for a task until it is completed.
Step 4: feature templates and method template are carried out matched inference machine by building, choose optimal side for step 7
Method template such as Fig. 5.
ω when the corresponding method collection probability calculation of the 6th process template in step 7zValue such as table 2.
The corresponding method collection each method template of 2 process template Ptem0026 of table chooses probability tables
Reflection this feature probability is calculated to choose significance level to method and meet formula constraint.Feature in the feature templates
Value set is Cvk={ Cc1, Cc2 ..., Ccn }, (k=1,2 ..., n).Calculate the corresponding party of each feature in feature set Cvk
Method template point chooses probability P (Ccz).
Step 5: according to user demand and the need of the Modifying model main-process stream of continental rise mobile platform chassis Suspension Model
It wants, 8 process templates is chosen from the fusion process template library constructed in step 1, configure and be combined into main-process stream, described in formation
Main-process stream mixing operation is from beginning last complete procedure stream.
Step 5.1: according to user demand and the need of the Modifying model main-process stream of continental rise mobile platform chassis Suspension Model
It wants, the Modifying model class being located in process subtemplate library;
Step 5.2: 8 process templates are selected from process template library;
Step 5.3: the 8 process subtemplates selected according to step 5.2 sort according to the process of Modifying model, combination
At main-process stream, it is that model imports subtemplate that main-process stream, which starts node, and end node is export subtemplate.Process template set expression
For PTEM=Ptem0021, Ptem0022, Ptem0023, Ptem0024, Ptem0025, Ptem0026, Ptem0027,
Ptem0028 }, that is, the main-process stream mixing operation is formed from the complete procedure stream such as Fig. 6 to begin to end.
Step 6: feature templates corresponding to 8 process templates chosen using step 5 are to chassis suspension vibration model
Characteristics extraction is carried out, obtains 8 feature sets, for the method matching in guiding step seven.
Step 6.1: being corresponded with step 5 process template and feature templates, therefore determine process template in step 5
In the case where, 8 feature templates also determine therewith.Feature templates set expression be CTEM=Ctem1021, Ctem1022,
Ctem1023,Ctem1024,Ctem1025,Ctem1026,Ctem1027,Ctem1028}。
Step 6.2: characteristics extraction is carried out to chassis suspension vibration model.In 8 feature templates that step 6.1 obtains,
It by taking feature templates Ctem1026 as an example, carries out feature and is calculated, show that the 6th characteristic set of the model is Ctem1026=
{ 0.082,2,2s, 1,2, model of vibration, continental rise platform chassis suspension }, successively obtain remaining 7 characteristic set, such as Fig. 7.
Step 7: the inference machine in invocation step four, the 8 groups of feature sets extracted according to step 6 are true from step 5
The concentration of method template corresponding to fixed process template is screened, and matching obtains method template corresponding with each process template such as
Fig. 8.
Step 7.1: recalling the corresponding method template collection of a process template, from method template library first with process template
For the corresponding method template collection MP0126={ MP012601, MP012602, MP012603, MP012604 } of Ptem0026.
Step 7.2: inference machine is used, by the feature set obtained with the associated feature templates of process template 6 and the process mould
Each method that method template corresponding to plate is concentrated matches one by one, obtains the selection probability of each method.6th process mould
Method corresponding to plate concentrates the selection probability of each method to be expressed as: P (MP0126), side corresponding to the 6th process template
The various method probability of method such as table 3.
3 process template Ptem0026 each method template probabilities table of table
Step 7.3: at the beginning of the method template of the corresponding method concentration selection maximum probability of process template, 8 process templates
The maximum probability approach chosen and its probability such as following table are walked, the method template of 8 maximum probabilities of selection is judged, the 6th
The method chosen during a does not meet default use demand since building is complex, and this method template is picked from method concentration
It removes, method template maximum probability and meets use demand at this time.The corresponding process template of this method template is associated,
And the characteristic value in 8 feature sets is stored in historical data base, and update prior probability Ppri (MP0026).
Step 7.4: to process template set PTEM=Ptem0021, Ptem0022, Ptem0023, Ptem0024,
Ptem0025, Ptem0026, Ptem0027, Ptem0028 } in each template carry out step 7.1~7.3, matching obtain with respectively
The corresponding method template of process template such as table 3, and then the Modifying model for establishing continental rise mobile platform chassis suspension vibration model is total
The information link of process to keep the fusion of test data and Modifying model link more efficient, and can be improved multi-source heterogeneous
The precision of model after data fusion amendment.
The final associated method template table of each process template of table 4
Process template | Method template | Probability |
Ptem0021 | MP012102 | 0.820 |
Ptem0022 | MP012201 | 0.295 |
Ptem0023 | MP012301 | 0.362 |
Ptem0024 | MP012403 | 0.285 |
Ptem0025 | MP012503 | 0.420 |
Ptem0026 | MP012601 | 0.259 |
Ptem0027 | MP012702 | 0.360 |
Ptem0028 | MP012801 | 0.820 |
Step 8: it matches to obtain method template corresponding with each process template and then establish continental rise movement to put down using step 7
Information link such as Fig. 8 of the Modifying model main-process stream of platform chassis suspension vibration model, and then it is outstanding to improve continental rise mobile platform chassis
Frame model of vibration engineering design efficiency 40% shakes to continental rise mobile platform chassis suspension using the Modifying model information link of foundation
Movable model, which carries out parameters revision, can be improved the precision of prediction 30% of model.
Above-described specific descriptions have carried out further specifically the purpose of invention, technical scheme and beneficial effects
It is bright, it should be understood that the above is only a specific embodiment of the present invention, the protection model being not intended to limit the present invention
It encloses, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention
Protection scope within.
Claims (10)
1. a kind of multi-source heterogeneous data fusion and Modifying model process management and control method, it is characterised in that: include the following steps,
Step 1: subprocess blocking that the data fusion of multi-layer complex engineering system, Modifying model main-process stream are covered,
Modularization, building are suitable for combining the process template of configuration, and forming process template library, realize from user demand, are corrected object
Feature be guiding fusion process configure on demand, flexible combination, with realize flexibility fusion process control;
Technical field is corresponded to according to multi-layer complex engineering system, to by the data fusion of multi-layer complex engineering system, model
Amendment main-process stream is split, and forms each process unit, and pass through process unit and corresponding technical method and process
The information that application scenarios are included is integrated, building process template, and forms fusion process template by abundant process template
Library, realize from user demand, be corrected object feature be oriented to fusion process configure on demand, flexible combination is soft to realize
Property fusion process control;
Step 2: according to each template type of fusion process template and technical field, the field is found for different problems
Typical technology method under scene by modifying to corresponding typical technology method, encapsulating generation method solver, and mentions
Corresponding typical technology method relevant information construction method template is taken, the relevant information includes input/output information;Pass through step
The matching of inference machine implementation method template and process template constructed by rapid five, realize fusion process in method it is flexible call,
Configuration and rationally application on demand;
Step 3: according to method during required data characteristics and Modifying model when choosing method template in fusion process
The model information building data characteristics template that template needs when choosing;By the initial data of importing and model according to feature templates into
The calculating and extraction of row characteristic value, form multiple characteristic sets;
Step 4: feature templates and method template are carried out matched inference machine by building, choose best practice mould for step 7
Plate;
Step 5: according to the need of the data fusion and Modifying model main-process stream of user demand and multi-layer complex engineering system
It wants, n template is chosen from the fusion process template library constructed in step 1, configuration group synthetic population merges template, forms institute
Main-process stream mixing operation is stated from the complete procedure stream to begin to end;
Step 6: using step 5 choose process template corresponding to feature templates to by data to be processed, model to step
Rapid five characteristic value extracts, obtain epicycle data fusion or Modifying model source data or source model corresponding to n group
Feature, for the method matching in guiding step seven;
Step 7: using the inference machine in step 4, the n group feature set extracted according to step 6 is determined from step 5
The concentration of method template corresponding to process template is screened, and matching obtains method template corresponding with each process template, in turn
It establishes multi-layer complex engineering system and corresponds to the information link of the data fusion of technical field, Modifying model main-process stream, to make
The fusion of test data and Modifying model link are more efficient, and can be improved the essence of model after multi-source heterogeneous data fusion amendment
Degree.
2. a kind of multi-source heterogeneous data fusion as described in claim 1 and Modifying model process management and control method, it is characterised in that:
Further include step 8, matches to obtain method template corresponding with each process template using step 7 and then establish multi-layer complexity work
Journey system corresponds to the information link of the data fusion of technical field, Modifying model main-process stream, and then improves multi-layer complex engineering
System Engineering Design efficiency improves multi-layer complex engineering system using model after the multi-source heterogeneous data fusion amendment of foundation and answers
For the precision of prediction of corresponding field of engineering technology, correlation engineering technical problem is solved.
3. a kind of multi-source heterogeneous data fusion as claimed in claim 1 or 2 and Modifying model process management and control method, feature exist
In: the fusion process template mentioned in step 1 is by procedural information dictionary, and process feature explanation, independent method template and subordinate are special
Four part of template is levied to constitute;Procedural information dictionary includes the essential information of the process, and the essential information includes process template name
Title, template number, date created;Process feature illustrate include the process template head node and tail node digital number,
The process function explanation carries out template selection for user and refers to sequence, and the head node and tail node are digital number
And association mode be it is one-to-many, the function declaration be explanatory note;Independent method template include in method template library with the mistake
The relevant all method templates of journey;Additional characteristic feature template includes feature templates relevant to the process in feature templates library.
4. a kind of multi-source heterogeneous data fusion as claimed in claim 3 and Modifying model process management and control method, it is characterised in that:
Feature templates described in step 3 are mainly by feature templates dictionary of information, associated process template information and the characteristic value three
Part is constituted;Feature templates dictionary of information includes the number title of the feature templates, template number, template type, creation day
Phase;Association process template embodies the one-to-one incidence relation between feature templates and a certain process template;The characteristic value is
The characteristic value that a certain process template needs when matching with method template, it includes number that matched characteristic value is needed in data characteristics template
According to type, data capacity, data dimension, sample average, data variance, the characteristic value that aspect of model template needs includes model class
Type, model field, the input/output format of model, model parameter number, model solution speed, model emulation result complex phase relationship
Number;The data and model of importing pass through the n group by extracting feature with the selected associated feature templates of n process template
Feature set is associated with process template to select n method template;
Contained content is by method dictionary of information, method characteristic explanation, method solver three parts in method template in step 2
It constitutes;Method dictionary of information essential information includes the method template title of the process, template number, date created;Method characteristic
Template includes the application message and application scenarios feature of this method, and the application message includes outputting and inputting for method, institute
Stating application scenarios feature includes data dimension, sample average, data variance domain, data capacity, data type, model emulation result
With the coefficient of determination of test value, model parameter number, the model solution time, whether model is non-linear, model order, model class
Type, model field;Method solver is the computing module of method template, is responsible for converting output for input.
5. a kind of multi-source heterogeneous data fusion as claimed in claim 4 and Modifying model process management and control method, it is characterised in that:
Step 4 concrete methods of realizing is,
Step 4.1: all methods in the MPi in method template set belonging to process template i are calculated and choose probability, it is described
I is process template number calculation formula such as following formula (1)-(3);
MPi=MPi1, MPi2 ..., MPin }, (i=1,2 ..., n) (1)
Wherein, P (MPin) is the selection probability of method template n, P (Cviz) it is the feature set extracted according to feature templates Ctemk
In Cvi under the value condition of feature Cviz, the probability of this method template, ω are chosenzIt is P (Cviz) weight, reflection this feature it is general
Rate chooses significance level to method and meets formula constraint;Characteristic value collection in the feature templates be Cvi=Cvi1,
Cvi2 ..., Cvin }, (z=1,2 ..., n), i is characterized template number;
Step 4.2: calculating the corresponding this method template point of each feature in feature set Cvi and choose probability P (Cviz), it calculates such as formula
(4);
P(Cviz)=Ppri (Cviz)*Sim(Mviz,Cviz), z=1,2 ..., n (4)
Mviz is special z-th in the recommended characteristics collection Mvi={ Mvi1, Mvi2 ..., Mvin } of a certain method template in above formula
Sign, α is normalization coefficient, Sim (Mviz,Cviz) be a certain method template recommended characteristics and feature templates in be corresponding to it spy
The degree of fitting of sign, and meet formula constraint;
Step 4.3: by step 4.1,4.2 buildings inference machines in parameter and degree of fitting judgement schematics in each power
Value is realized and carries out asking for relating value between the feature extraction of data/model, method template and feature templates according to feature templates
Solution, to realize the combination configuration between each template, i.e. feature templates and method template are carried out matched inference machine by building;With
The growth of fusion number, fusion process form assembly, the corresponding method template of each process template and method template and feature
The combination case sample size of template is gradually increased, and is improved the order of accuarcy of prior probability in inference machine, is pushed away to improve inference machine
Reason ability.
6. a kind of multi-source heterogeneous data fusion as claimed in claim 5 and Modifying model process management and control method, it is characterised in that:
Step 5 concrete methods of realizing is,
Step 5.1: according to the need of the data fusion and Modifying model main-process stream of user demand and multi-layer complex engineering system
It wants, the data fusion class or Modifying model class being located in process subtemplate library;
Step 5.2: processing data are selected from process subtemplate library, the process unit template that model needs;
Step 3: the process subtemplate selected according to step 5.2 sorts according to the fusion process in relevant art field, combination
At main-process stream, it is that data or model import subtemplate that main-process stream, which starts node, and end node is export subtemplate;Process template collection
Conjunction is expressed as PTEM={ Ptem1, Ptem2 ..., Ptemn }, that is, forms the main-process stream mixing operation from the complete mistake to begin to end
Cheng Liu.
7. a kind of multi-source heterogeneous data fusion as claimed in claim 6 and Modifying model process management and control method, it is characterised in that:
Step 6 concrete methods of realizing is,
Step 6.1: being corresponded with step 5 process template and feature templates, therefore determine the feelings of process template in step 5
Under condition, n feature templates also determine therewith;Feature templates set expression is CTEM={ Ctemi1, Ctemi2 ..., Ctemin },
In is n feature templates number;
Step 6.2: characteristics extraction is carried out to data or model;In the n feature templates that step 6.1 obtains, each template institute
The characteristic value number and type for including difference;The characteristic value collection that feature templates i is included is expressed as according to each mould
The type for the characteristic value for including in plate calculates the data of importing, model, show that the data of this fusion, model are corresponding
N group feature.
8. a kind of multi-source heterogeneous data fusion as claimed in claim 7 and Modifying model process management and control method, it is characterised in that:
Step 7 concrete methods of realizing is,
Step 7.1: recalled from method template library first each process template corresponding method template collection MPi=MPi1,
MPi2 ..., MPit }, (i=1,2 ..., n), wherein n is the number of process template, and t is the corresponding method of i-th of process template
The number of template centralized way;
Step 7.2: and then inference machine is used, by the feature set obtained with the associated feature templates of process template and the process template
Each method that corresponding method template is concentrated matches one by one, obtains the selection probability of each method;I-th of process template
Corresponding method concentrates the selection probability of each method to be expressed as: P (MPj), (j=1,2 ..., n);
Step 7.3: the method template for choosing maximum probability is concentrated from the corresponding method of process template, to the side for choosing maximum probability
Method template is judged, when meeting default use demand, by the corresponding process of the method template of the maximum probability of selection
Template is associated, and the method template of the maximum probability with selection is carried out the characteristic value for including in matched feature templates deposit
In historical data base, and prior probability Ppri (MPj) is updated, (j=1,2 ..., t);When the method template of the maximum probability is not inconsistent
When closing default use demand, the method for the maximum probability of selection is concentrated from the method for step 7.1 and is rejected, and return step 7.3;
Prior probability Ppri described in step 7.3 is by certain method template feature that successful match is crossed in history fusion process
The feature set statistics that template is included obtains, embodies the feature that this method template is most suitable for the data of processing;Prior probability will
A factor in inference machine is chosen as method template;After feature templates and method template successful match, feature templates include
Feature set be stored into historical data base, prior probability is calculated by formula (7);
Ppri=n/N (7)
In formula, n is the number that the value of this feature occurred in history, and N is total history number;
Step 7.4: step 7.1 is carried out to each template in process template set PTEM={ Ptem1, Ptem2 ..., Ptemn }
~7.3, matching obtains method template corresponding with each process template and then establishes the corresponding technology of multi-layer complex engineering system leading
Data fusion, the information link of Modifying model main-process stream in domain, so that the fusion for making test data and Modifying model link are more
Efficiently, and it can be improved the precision of model after the amendment of multi-source heterogeneous data fusion.
9. a kind of multi-source heterogeneous data fusion as claimed in claim 8 and Modifying model process management and control method, it is characterised in that:
The multi-layer complex engineering system includes continental rise plateform system, aerospace system, complex mechanical system;The multi-layer is multiple
Miscellaneous engineering system common feature: including multiple subsystem models, there are the inferior characteristics of non-linear or high-order for model itself.
10. a kind of multi-source heterogeneous data fusion as claimed in claim 9 and Modifying model process management and control method, feature exist
In: when be applied to continental rise plateform system when, can be improved domain suspended model, road surface model, body construction finite element model with
And the simulation accuracy of other continental rise platform models, reduce emulation cycle;Model after being improved using precision can be instructed preferably
The design of Structural Parameters of continental rise platform, continental rise Platform Vibration characteristic optimizing, handling index optimization and riding comfort index
Optimization, is with a wide range of applications and benefit;
When being applied to aerospace system, aerospace platform Each part finite element model, aerodynamics can be improved
The simulation accuracy of model and other models reduces emulation cycle;Model after being improved using precision can preferably instruct to navigate
Each subsystem structure of empty aerospace system optimizes and aerodynamic properties optimization, is with a wide range of applications and benefit;
When being applied to complex mechanical system, cutter, machine tool body structure model, thermodynamical model, accident analysis can be improved
Model, the simulation accuracy of status monitoring model reduce emulation cycle;Model after being improved using precision can preferably instruct multiple
The Structural Design of miscellaneous mechanical system, moreover it is possible to be health monitoring, fault diagnosis and the structural vibration control etc. of Time variable structure
Aspect application provides strong support, is with a wide range of applications.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910003356.3A CN109492059B (en) | 2019-01-03 | 2019-01-03 | Multi-source heterogeneous data fusion and model correction process control method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910003356.3A CN109492059B (en) | 2019-01-03 | 2019-01-03 | Multi-source heterogeneous data fusion and model correction process control method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109492059A true CN109492059A (en) | 2019-03-19 |
CN109492059B CN109492059B (en) | 2020-10-27 |
Family
ID=65713789
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910003356.3A Expired - Fee Related CN109492059B (en) | 2019-01-03 | 2019-01-03 | Multi-source heterogeneous data fusion and model correction process control method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109492059B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110388315A (en) * | 2019-07-05 | 2019-10-29 | 中国石油大学(北京) | Oil transfer pump fault recognition method, apparatus and system based on Multi-source Information Fusion |
CN111221018A (en) * | 2020-03-12 | 2020-06-02 | 南京航空航天大学 | GNSS multi-source information fusion navigation method for inhibiting marine multipath |
CN112256978A (en) * | 2020-12-22 | 2021-01-22 | 畅捷通信息技术股份有限公司 | Data processing method, device and medium based on data model |
CN112347537A (en) * | 2020-10-27 | 2021-02-09 | 青岛超源博创科技发展有限公司 | Calibration method and device for engineering structure numerical model, electronic equipment and medium |
CN112966825A (en) * | 2021-04-13 | 2021-06-15 | 杭州欣禾圣世科技有限公司 | Multi-model fusion parallel reasoning method, device and system based on python |
CN114661810A (en) * | 2022-05-24 | 2022-06-24 | 国网浙江省电力有限公司杭州供电公司 | Lightweight multi-source heterogeneous data fusion method and system |
CN114880385A (en) * | 2021-07-27 | 2022-08-09 | 云南省地质环境监测院(云南省环境地质研究院) | Method and device for accessing geological disaster data through automatic combined flow |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001051711A (en) * | 1999-08-04 | 2001-02-23 | Korea Electronics Telecommun | Supporting system constructing method for machine parts process and for machining condition based on classification code knowledge base and recording medium recording its procedure |
CN101980217A (en) * | 2010-10-18 | 2011-02-23 | 北京理工大学 | Template-based integrated design platform construction method |
CN101980218A (en) * | 2010-10-18 | 2011-02-23 | 北京理工大学 | Template-based integrated design platform |
CN104866598A (en) * | 2015-06-01 | 2015-08-26 | 北京理工大学 | Heterogeneous database integrating method based on configurable templates |
CN104951590A (en) * | 2015-04-20 | 2015-09-30 | 广东工业大学 | Knowledge service system and method for die designing |
CN107085758A (en) * | 2017-01-19 | 2017-08-22 | 北京恒华伟业科技股份有限公司 | A kind of project management method, apparatus and system |
CN107193858A (en) * | 2017-03-28 | 2017-09-22 | 福州金瑞迪软件技术有限公司 | Towards the intelligent Service application platform and method of multi-source heterogeneous data fusion |
CN108399154A (en) * | 2017-12-28 | 2018-08-14 | 北京金科汇智科技有限公司 | Engineering test data collecting system |
CN108846384A (en) * | 2018-07-09 | 2018-11-20 | 北京邮电大学 | Merge the multitask coordinated recognition methods and system of video-aware |
-
2019
- 2019-01-03 CN CN201910003356.3A patent/CN109492059B/en not_active Expired - Fee Related
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001051711A (en) * | 1999-08-04 | 2001-02-23 | Korea Electronics Telecommun | Supporting system constructing method for machine parts process and for machining condition based on classification code knowledge base and recording medium recording its procedure |
CN101980217A (en) * | 2010-10-18 | 2011-02-23 | 北京理工大学 | Template-based integrated design platform construction method |
CN101980218A (en) * | 2010-10-18 | 2011-02-23 | 北京理工大学 | Template-based integrated design platform |
CN104951590A (en) * | 2015-04-20 | 2015-09-30 | 广东工业大学 | Knowledge service system and method for die designing |
CN104866598A (en) * | 2015-06-01 | 2015-08-26 | 北京理工大学 | Heterogeneous database integrating method based on configurable templates |
CN107085758A (en) * | 2017-01-19 | 2017-08-22 | 北京恒华伟业科技股份有限公司 | A kind of project management method, apparatus and system |
CN107193858A (en) * | 2017-03-28 | 2017-09-22 | 福州金瑞迪软件技术有限公司 | Towards the intelligent Service application platform and method of multi-source heterogeneous data fusion |
CN108399154A (en) * | 2017-12-28 | 2018-08-14 | 北京金科汇智科技有限公司 | Engineering test data collecting system |
CN108846384A (en) * | 2018-07-09 | 2018-11-20 | 北京邮电大学 | Merge the multitask coordinated recognition methods and system of video-aware |
Non-Patent Citations (1)
Title |
---|
明振军、阎艳、王国新: "《基于可配置模板的产品设计知识多维度表达方法》", 《北京理工大学学报》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110388315A (en) * | 2019-07-05 | 2019-10-29 | 中国石油大学(北京) | Oil transfer pump fault recognition method, apparatus and system based on Multi-source Information Fusion |
CN111221018A (en) * | 2020-03-12 | 2020-06-02 | 南京航空航天大学 | GNSS multi-source information fusion navigation method for inhibiting marine multipath |
CN111221018B (en) * | 2020-03-12 | 2022-04-08 | 南京航空航天大学 | GNSS multi-source information fusion navigation method for inhibiting marine multipath |
CN112347537A (en) * | 2020-10-27 | 2021-02-09 | 青岛超源博创科技发展有限公司 | Calibration method and device for engineering structure numerical model, electronic equipment and medium |
CN112347537B (en) * | 2020-10-27 | 2021-09-14 | 青岛超源博创科技发展有限公司 | Calibration method and device for engineering structure numerical model, electronic equipment and medium |
CN112256978A (en) * | 2020-12-22 | 2021-01-22 | 畅捷通信息技术股份有限公司 | Data processing method, device and medium based on data model |
CN112256978B (en) * | 2020-12-22 | 2021-03-19 | 畅捷通信息技术股份有限公司 | Data processing method, device and medium based on data model |
CN112966825A (en) * | 2021-04-13 | 2021-06-15 | 杭州欣禾圣世科技有限公司 | Multi-model fusion parallel reasoning method, device and system based on python |
CN114880385A (en) * | 2021-07-27 | 2022-08-09 | 云南省地质环境监测院(云南省环境地质研究院) | Method and device for accessing geological disaster data through automatic combined flow |
CN114880385B (en) * | 2021-07-27 | 2022-11-22 | 云南省地质环境监测院(云南省环境地质研究院) | Method and device for accessing geological disaster data through automatic combination process |
CN114661810A (en) * | 2022-05-24 | 2022-06-24 | 国网浙江省电力有限公司杭州供电公司 | Lightweight multi-source heterogeneous data fusion method and system |
Also Published As
Publication number | Publication date |
---|---|
CN109492059B (en) | 2020-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109492059A (en) | A kind of multi-source heterogeneous data fusion and Modifying model process management and control method | |
CN107590319B (en) | Knowledge modeling method and system for auxiliary design of mechanical product scheme | |
CN105138802B (en) | A kind of gun barrel intelligent design system and design method | |
CN106776963B (en) | The online method for visualizing of light-weighted BIM big data and system | |
CN108932588B (en) | Hydropower station group optimal scheduling system with separated front end and rear end and method | |
CN111708531B (en) | Data processing method and device | |
CN109213754A (en) | A kind of data processing system and data processing method | |
CN108924198A (en) | A kind of data dispatching method based on edge calculations, apparatus and system | |
CN111177897A (en) | Production line production process data docking and simulation model rapid modeling method and system under cloud manufacturing mode | |
CN104820733B (en) | A kind of bullet train demand meta-model method for building up and device | |
CN110187647A (en) | Model training method and system | |
CN110221858A (en) | Analogue data generation method, device, equipment and computer readable storage medium | |
CN107632845A (en) | A kind of integrated knowledge cloud service method and system of the processing of runner bucket multi-axis milling | |
CN109189393A (en) | Method for processing business and device | |
CN115544775A (en) | Digital twin workshop multi-dimensional multi-level model construction and dynamic configuration method | |
CN111209634A (en) | Manufactured part design based on artificial intelligence | |
CN104020756A (en) | Logic network topology sorting and storing method for fault diagnosing system | |
CN109034403A (en) | A kind of urban rail transit equipment skeleton | |
CN101986282B (en) | Topology adaptation method and device | |
CN106096159B (en) | A kind of implementation method of distributed system behavior simulation analysis system under cloud platform | |
CN103514331B (en) | A kind of method from Simulink model conversion to uml model | |
CN113128041B (en) | Digital factory MES system development method based on digital twin virtual technology | |
CN109901830A (en) | A kind of symbol arranging method and system for scada system development | |
Ouyang et al. | Semantic enrichment of object associations across federated BIM semantic graphs in a common data environment | |
CN102594851A (en) | Method for dynamically building marine application service chain |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201027 |