CN106960065A - The robust appraisal procedure and system of a kind of Complex simulation systems confidence level - Google Patents
The robust appraisal procedure and system of a kind of Complex simulation systems confidence level Download PDFInfo
- Publication number
- CN106960065A CN106960065A CN201610012759.0A CN201610012759A CN106960065A CN 106960065 A CN106960065 A CN 106960065A CN 201610012759 A CN201610012759 A CN 201610012759A CN 106960065 A CN106960065 A CN 106960065A
- Authority
- CN
- China
- Prior art keywords
- assessment
- index
- data
- confidence level
- evaluation index
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
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
The present invention discloses the robust appraisal procedure and system of a kind of Complex simulation systems confidence level, described this method include set up can graphically, the index system of credibility evaluation interface of stratification, by reliability assessment index of the index decomposition of Complex simulation systems reliability assessment for each emulation element model, and calculate each index weights;The data collected in each l-G simulation test and reliability assessment result are organized and managed in the form of data set;By multiple assessment data sources, various forms of qualitative, quantitative information merged, bottom-up successively synthesis obtains the reliability assessment result of analogue system.Technical scheme of the present invention can carry out Unify legislation to multiple assessment information, new assessment knowledge is constantly incorporated during assessment result consistency analysis and feedback iteration, the influence of the uncertain factor in evaluation process is corrected, the assessment result of stable convergence is obtained and using Multi-level comprehensive evaluation model realization to the effective utilizations for assessing source information more.
Description
Technical field
The present invention relates to a kind of credibility evaluation method, more particularly to a kind of Complex simulation systems confidence level
Robust appraisal procedure and system.
Background technology
The confidence level of analogue system refers to relative to specific Simulation Application purpose, its process, phenomenon and knot
The degree of the correct reflection real system of fruit.It is that analogue system can be applied to truly with higher confidence level
System simulation and the premise of analysis.By complication system is had the special feature that in itself, its analogue system can
Reliability is assessed and generally involves multidisciplinary, multi-level simulation model, multiple l-G simulation test processes, Yi Jiduo
The assessment experts in field, are the evaluation processes of a system.Reliability assessment result is easily assessed work
The effect of various factors in work, major influence factors include:Assess object and assess main body.Assess
Object, i.e. Complex simulation systems, the features such as composition having due to itself is complicated, internal correlation is complicated,
The system general performance of often leading to goes out the behavioral trait of complexity, is that the accurate measurement of confidence level and assessment are brought
Uncertainty.Assessment main body reflects the method and evidence for metric evaluation object, due to assessing main body
There is preference selection to evaluation index and evaluation index weight distribution, lead to as " data " of assessing evidence
There is Dependability Problem often from multiple sources such as l-G simulation test, actual experimental and expertise.This
It frequently can lead to be difficult to obtain consistent assessment result a bit.Traditional Simulation Credibility Evaluation process includes fixed
Justice assess target, analysis system key element, set up evaluation index system, set up evaluation structure and assessment level,
Expert is estimated, and obtains assessment result, and this evaluation process is typically uni-directional, single.Assess master
Body is often difficult to the factor that impact evaluation credible result degree is found from the angle of macroscopic view, and their effect
Degree.Even if repeatedly being assessed, it is also possible to simply obtain different assessment results, lacking one kind will comment
The evaluation process for estimating results-driven stable and consistent with comprehensive utilization different assessed available information in data source and commented
Estimate method.For unidirectional, the single feature of traditional Complex simulation systems reliability assessment process, it is simultaneously
It is effective to solve in traditional Complex simulation systems reliability assessment, due to assessing the subjective preferences of main body and many
The assessment result that the isomerism that data are assessed in source is caused is uncertain.
Accordingly, it is desirable to provide the robust appraisal procedure and system of a kind of Complex simulation systems confidence level, build
The robust evaluation process model of Complex simulation systems confidence level and the confidence level that information fusion is assessed towards multi-source
Comprehensive estimation method, by setting up the feedback optimized mechanism of analysis result, forms the assessment iterated
Journey, realizes the assessment knowledge for being continuously added merged multi-source heterogeneous assessment information in an iterative process, amendment
The influence of uncertain factor in evaluation process, obtains the assessment result of stable convergence.
The content of the invention
It is an object of the invention to provide a kind of robust appraisal procedure of Complex simulation systems confidence level and it is
System, solves unidirectional, the single feature of traditional Complex simulation systems reliability assessment process, and due to not
The problem of assessment result is unstable caused by certainty factor.
In order to solve the above technical problems, the present invention uses following technical proposals:
A kind of robust appraisal procedure of Complex simulation systems confidence level, described this method includes
Set up can graphically, the index system of credibility evaluation interface of stratification, can by Complex simulation systems
The index decomposition that reliability is assessed emulates the reliability assessment index of element model for each, and calculates each index
Weight;
The data collected in each l-G simulation test and reliability assessment result are carried out in the form of data set
Organization and management;
The information of multiple assessment data sources is merged, it is bottom-up successively comprehensively to obtain analogue system
Reliability assessment result;
Set up the consistency analysis of reliability assessment result and the iterative process of feedback modifiers.
Preferably, described this method further comprises
Selective shows certain reliability assessment for once assessing each evaluation index at all levels afterwards
As a result and analogue system confidence level, or on each level after repeatedly assessment result is average each index it is credible
The confidence level of degree and analogue system.
Preferably, the foundation can graphically, the index system of credibility evaluation of stratification and calculate it is each refer to
Mark weight further comprises
Confidence of simulation system evaluation index is decomposed into the confidence level of each subsystem of composition analogue system
Evaluation index;
By the confidence level that the reliability assessment index decomposition of each subsystem is the subsystem more to next stage
Evaluation index, untill decomposing the reliability assessment index of emulation element model of minimum;
To each emulation element model according to modeling requirement, conceptual model, Mathematical Modeling, simulation model
And the several aspects of simulation result are estimated index decomposition, the confidence level of each emulation element model is obtained
Evaluation index;
The evaluation index composing indexes system of each comprehensive element model, then to the importance of evaluation index
Compared two-by-two and set up Judgement Matrix, calculated by characteristic value and obtain evaluation index weight.
Preferably, the assessment of the data and different reliability assessment tasks that are collected in the l-G simulation test
Result data is organized according to data set;
The description information of data set further comprises dataset name, l-G simulation test task or assesses task name
Number of folders and data file number in title, data name, creation time, storing path, data set
Amount.
Preferably, methods described is merged to qualitative, the quantitative assessment information of multiple assessment data sources;
The assessment data source further comprises l-G simulation test data, real system test data and expert's warp
Test knowledge;
The multiple data source of assessing is for the fusion method formula of some evaluation index:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
A kind of robust assessment system of Complex simulation systems confidence level, the system includes reliability assessment index
System module, assessment data management module, confidence level comprehensive assessment module and assessment result analysis show mould
Block;
The index system of credibility evaluation module, for set up can graphically, the confidence level of stratification comments
Assessment system interface, is each emulation element mould by the index decomposition of Complex simulation systems reliability assessment
The reliability assessment index of type, and calculate each index weights;
The assessment data management module, for the data and confidence level that will be collected in each l-G simulation test
Assessment result is organized and managed in the form of data set;
The confidence level comprehensive assessment module, for the information of multiple assessment data sources to be merged, from
The lower reliability assessment result for successively comprehensively obtaining analogue system upwards;
The assessment result analyzes display module, each layer after once being assessed for selective display
The reliability assessment result and the confidence level of analogue system of each evaluation index on secondary, or repeatedly assess knot
The confidence level and the confidence level of analogue system of each index on each level after fruit is average.
Preferably, to set up evaluation index hierarchical relationship in the index system of credibility evaluation module further
Including
Confidence of simulation system evaluation index is decomposed into the confidence level of each subsystem of composition analogue system
Evaluation index;
By the confidence level that the reliability assessment index decomposition of each subsystem is the subsystem more to next stage
Evaluation index, untill decomposing the reliability assessment index of emulation element model of minimum;
To each emulation element model according to modeling requirement, conceptual model, Mathematical Modeling, simulation model
And the several aspects of simulation result are estimated index decomposition, the confidence level of each emulation element model is obtained
Evaluation index;
The evaluation index composing indexes system of each comprehensive element model, then to the importance of evaluation index
Compared two-by-two and set up Judgement Matrix, calculated by characteristic value and obtain evaluation index weight.
Preferably, the assessment of the data and different reliability assessment tasks that are collected in the l-G simulation test
Result data is organized according to data set;
The description information of data set further comprises dataset name, l-G simulation test task or assesses task name
Number of folders DBMS number of files in title, data name, creation time, storing path, data set
Amount.
Preferably, the system is merged to qualitative, the quantitative assessment information of multiple data sources;
In the assessment data management module
Multiple assessment data sources further comprise that l-G simulation test data, actual experimental data and expertise are known
Know;
The multiple data source of assessing is for the fusion method formula of some evaluation index:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
Beneficial effects of the present invention are as follows:
Technical scheme of the present invention has the advantage that:
1st, the robust for realizing the Complex simulation systems confidence level with feedback optimized mechanism is assessed, and is passed through
Confidence level verification, multi-data source fusion based on multigroup emulation data are assessed, assessment result consistency analysis
And feedback, the evaluation process iterated is formed, constantly incorporates know from expertise in an iterative process
In the assessment information for assessing data sources such as knowledge, test data and emulation data, amendment evaluation process not more
The influence of certainty factor, makes the comments of expert tend to objective, assessment result tended to stable, closed
Reason with it is consistent;
2nd, the multi-source heterogeneous confidence level comprehensive estimation method for assessing information is merged, to coming from experiment number
Unify legislation is carried out according to the multiple assessment information of, emulation data and expertise knowledge, is utilized multistage comprehensive
Assessment models are realized to the effective utilizations for assessing source information more;
3rd, there is qualitative, qualitative assessment index suitable for evaluation index system, and assess data
Source has the reliability assessment of the Complex simulation systems of diversified feature, it is adaptable to improve science and techniques of defence field
Complex equipment and civil use industry complex product etc. develop the reliability assessment level of emulation, it is contemplated that this skill
Art achievement has good industrialization prospect.
Brief description of the drawings
The embodiment to the present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 is a kind of robust appraisal procedure flow chart of Complex simulation systems confidence level in the embodiment of the present invention;
Fig. 2 is a kind of robust assessment system schematic diagram of Complex simulation systems confidence level in the embodiment of the present invention.
Embodiment
In order to illustrate more clearly of the present invention, the present invention is done into one with reference to preferred embodiments and drawings
The explanation of step.Similar part is indicated with identical reference in accompanying drawing.Those skilled in the art
It should be appreciated that following specifically described content is illustrative and be not restrictive, it should not be limited with this
Protection scope of the present invention.
The invention provides a kind of robust appraisal procedure of Complex simulation systems confidence level, as shown in figure 1,
Described this method includes:
Set up can graphically, the index system of credibility evaluation interface of stratification, can by Complex simulation systems
The index decomposition that reliability is assessed emulates the reliability assessment index of element model for each, and calculates each index
Weight;
The data collected in each l-G simulation test and reliability assessment result are carried out in the form of data set
Organization and management;
The information of multiple assessment data sources is merged, it is bottom-up successively comprehensively to obtain analogue system
Reliability assessment result;
Set up the consistency analysis of reliability assessment result and the iterative process of feedback modifiers.
The robust that the inventive method realizes the Complex simulation systems confidence level with feedback optimized mechanism is commented
Estimate, verified by the confidence level based on multigroup emulation data, multi-data source fusion is assessed, assessment result one
The analysis of cause property and feedback, form the evaluation process iterated, constantly incorporate to act on one's own in an iterative process
The assessment information for assessing data source such as family's Heuristics, test data and emulation data, amendment was assessed more
The influence of uncertain factor in journey, makes the comments of expert tend to objective, tends to assessment result
It is stable, reasonable with it is consistent;
Also, the multi-source heterogeneous confidence level comprehensive estimation method for assessing information of the inventive method fusion, to coming
Unify legislation, profit are carried out from the multiple assessment information in test data, emulation data and expertise knowledge
With Multi-level comprehensive evaluation model realization to the effective utilizations for assessing source information more;
In addition, the inventive method has qualitative, qualitative assessment index suitable for evaluation index system, and
And assess the reliability assessment that data source has the Complex simulation systems of diversified feature, it is adaptable to improve
Science and techniques of defence field complex equipment and civil use industry complex product etc. develop the reliability assessment water of emulation
It is flat, it is contemplated that technique achievement has good industrialization prospect.
Preferably, described this method further comprises
Selective shows certain reliability assessment for once assessing each evaluation index at all levels afterwards
As a result and analogue system confidence level, or on each level after repeatedly assessment result is average each index it is credible
The confidence level of degree and analogue system.
Preferably, the foundation can graphically, the index system of credibility evaluation of stratification and calculate it is each refer to
Mark weight further comprises
Confidence of simulation system evaluation index is decomposed into the confidence level of each subsystem of composition analogue system
Evaluation index;
By the confidence level that the reliability assessment index decomposition of each subsystem is the subsystem more to next stage
Evaluation index, untill decomposing the reliability assessment index of emulation element model of minimum;
To each emulation element model according to modeling requirement, conceptual model, Mathematical Modeling, simulation model
And the several aspects of simulation result are estimated index decomposition, the confidence level of each emulation element model is obtained
Evaluation index;
The evaluation index composing indexes system of each comprehensive element model, then to the importance of evaluation index
Compared two-by-two and set up Judgement Matrix, calculated by characteristic value and obtain evaluation index weight.
Preferably, the assessment of the data and different reliability assessment tasks that are collected in the l-G simulation test
Result data is organized according to data set;
The description information of data set further comprises dataset name, l-G simulation test task or assesses task name
Number of folders and data file number in title, data name, creation time, storing path, data set
Amount.
Preferably, methods described is merged to qualitative, the quantitative assessment information of multiple assessment data sources;
The assessment data source further comprises l-G simulation test data, real system test data and expert's warp
Test knowledge;
The multiple data source of assessing is for the fusion method formula of some evaluation index:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
The invention also discloses a kind of robust assessment system of Complex simulation systems confidence level, as shown in Fig. 2
The system includes:Build the index system of credibility evaluation module 1 of complication system, assess data management module
2nd, confidence level comprehensive assessment module 3 and assessment result analysis display module 4.
Index system of credibility evaluation module 1, sets up the graphical, stratification credible of Complex simulation systems
Spend evaluation index system interface, and distribution weight calculated for each index, index system of credibility evaluation with
The organization chart of stratification shows that each icon represents display on an evaluation index, icon and each commented
Estimate the title and weight of index.
Assess data management module 2, to collected in each l-G simulation test l-G simulation test data, true system
Unite test data and assessment result data are in the form of data set and organized.
Confidence level comprehensive assessment module 3, will come from l-G simulation test data, real system test data and specially
The information of multiple assessment data sources of family's Heuristics is merged, bottom-up successively comprehensively to be emulated
The reliability assessment result of system.
Assessment result analyzes display module 4, shows each evaluation index at all levels in single evaluation activity
Reliability assessment result and whole analogue system confidence level, or repeatedly assessment result (after comprehensive analysis/
After average) each level every evaluation index reliability assessment result and whole analogue system can
Reliability.
Index system of credibility evaluation module 1, builds the index system of credibility evaluation of Complex simulation systems.
Because the reliability assessment of Complex simulation systems generally includes large number of factor, typically using layer
Secondaryization method divides the orderly level interknited, sets up stratification index system, index system should energy
Reflect the main aspect of analogue system, so index system of credibility evaluation module 1, main in building process
Step includes:(1) it is each of composition analogue system by Complex simulation systems reliability assessment index decomposition
The reliability assessment index of subsystem;(2) by the reliability assessment index decomposition of each subsystem for more
To the reliability assessment index of the subsystem of next stage, until decompose minimum emulation element model can
Untill reliability evaluation index;(3) to each emulation element model according to modeling requirement, conceptual model,
Mathematical Modeling, simulation model and the several aspects of simulation result are estimated index decomposition, obtain each and imitate
The reliability assessment index of true strength prime model.The evaluation index composing indexes system of comprehensive each element model,
Then the importance of evaluation index is compared two-by-two by expert and sets up Judgement Matrix, pass through characteristic value meter
Calculation obtains evaluation index weight.
Index system of credibility evaluation module 1, it is possible to provide graphic interface, user is newly-built on interface to be assessed
Index system root node, and successively extension addition lower floor evaluation index, so as to set up whole evaluation index body
System, each evaluation index is shown with icon, and the title of evaluation index is added in the description information of icon.
After evaluation index system is set up, index system of credibility evaluation module 1 provides the user same layer
The contrast interface two-by-two of secondary evaluation index importance, adds comparing result, and generate each level by user
The importance Judgement Matrix of evaluation index, the weight for obtaining each evaluation index is calculated by characteristic value, and
In the description information for being shown in each icon.
Data management module 2 is assessed, data set is built.
The reasonability for the Simulation result data that Complex simulation systems operation is produced can reflect that analogue system is credible
Property, l-G simulation test is less subject to the limitation of the conditions such as funds and place, the multigroup emulation of acquisition can be run multiple times
Result data, the underlying basis index collection emulation data in index system of credibility evaluation.Truly
System test data can import the robust assessment system of Complex simulation systems confidence level from external storage.
L-G simulation test data, real system test data and assessment result data are entered in the form of data set
Row organization and administration, the description information of data set includes the title of data set, l-G simulation test task or assessed to appoint
Title, data name, creation time, storing path, the number of folders in data set and the data of being engaged in are literary
Number of packages amount.Select after some file, the data file that can further select this document to be included in pressing from both sides,
User's utilization assesses data management module 2 and preselects l-G simulation test data set, real system test data
Collection or assessment result data set, phase is loaded when carrying out confidence level comprehensive assessment or assessment result analysis display
The data file answered.
Confidence level comprehensive assessment module 3, for carrying out multi-data source fusion assessment.
The robust of Complex simulation systems confidence level, which is assessed, needs the support of multiple assessment data source, both including fixed
The emulation data and test data of amount, also including qualitatively expertise knowledge.In order to qualitative and quantitative
Assessment information carry out Unify legislation, each assessment number is described using the two-dimensional representation of knowledge degree and reliability
According to the reliability information included in source.
For qualitative evaluation index, judged by expert and weighed, in the form of marking, directly inputted
Expertise depth and the assessment result to assessing object credibility, are used as knowledge degree and trusting degree number
According to;For qualitative assessment index, according to selected l-G simulation test and real system test data set, calculate
The departure degree between data and test data is emulated as knowledge degree, by l-G simulation test data and real system
Calculating confidence level check results are analyzed in test data, if including multi-group data in data set,
Result of calculation is assembled using the method for D-S evidence theory, the confidence level based on multi-group data is realized
Assess the fusion of information.
Based on unified description, obtain each and assess the assessment result that data source is directed to some evaluation index,
Using the multiple information for assessing data source of extension Bayes methods fusion, i.e., multiple assessment data sources are directed to certain
The fusion method formula of individual evaluation index is:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
Confidence level to underlying basis index comprehensively, can calculate to the credible of the composite index on upper strata
Degree, the final bottom-up synthetic reliability assessment result for obtaining whole analogue system.
Assessment result analyzes display module 4, for assessment result consistency analysis and feedback.
In the present embodiment, due to there is the uncertainty of expert personal experience preference and l-G simulation test, hold
It is easily caused the reliability assessment result based on emulation data, expert reliability evaluation result and evaluation index power
Reassign appearance inconsistent, it is necessary to eliminate the inconsistent of assessment result.Using the multigroup assessment result of calculating
The method of desired value and variance, if there is relatively large deviation in the reliability assessment result of multigroup emulation data,
Then deviate assessment result the larger emulation data of desired value to check, judge whether l-G simulation test is effective;
If on some evaluation index, there is relatively large deviation in the evaluation result of different experts, or is commented at some
Estimate when there is relatively large deviation in the weight distribution of index, pass through organizes expert investigation and eliminate personal assessment's preference
Influence.
For single evaluation, assessment result analysis display module 4 obtains confidence level comprehensive assessment module 3
Result of calculation, and by the hierarchical structure of evaluation index system show each evaluation index at all levels can
The confidence level of reliability assessment result and whole analogue system.For repeatedly assessment, assessment result analysis is aobvious
Show that module 4 carries out consistency analysis after loading multiple assessment result data set, by the result after comprehensive analysis
It is shown in stratification evaluation index system structure.
In summary, technical scheme of the present invention, realizes the complex simulation with feedback optimized mechanism
The robust of system confidence level is assessed, by the way that the confidence level based on multigroup emulation data is verified, multi-data source melts
Assessment, assessment result consistency analysis and feedback are closed, the evaluation process iterated is formed, in iteration mistake
Constantly incorporate and commented from l-G simulation test data, real system test data and expertise knowledge etc. more in journey
Estimate the influence of the uncertain factor in the assessment information of data source, amendment evaluation process, make commenting for expert
Estimate opinion and tend to objective, make assessment result tend to it is stable, reasonable with it is consistent;Fusion is multi-source heterogeneous to assess letter
The confidence level comprehensive estimation method of breath, to coming from test data, emulation data and expertise knowledge
Multiple assessment information carries out Unify legislation, using Multi-level comprehensive evaluation model realization to many assessment source information
Effectively utilize;There is qualitative, qualitative assessment index suitable for evaluation index system, and assess data
The reliability assessment of Complex simulation systems of the source with diversified feature, it is adaptable to improve science and techniques of defence neck
Domain complex equipment and civil use industry complex product etc. develop the reliability assessment level of emulation, it is contemplated that this
Technological achievement has good industrialization prospect.
Obviously, the above embodiment of the present invention is only intended to clearly illustrate example of the present invention, and
It is not the restriction to embodiments of the present invention, for those of ordinary skill in the field,
It can also be made other changes in different forms on the basis of described above, here can not be to all
Embodiment be exhaustive, it is every to belong to the obvious change that technical scheme is extended out
Change or change the row still in protection scope of the present invention.
Claims (9)
1. a kind of robust appraisal procedure of Complex simulation systems confidence level, it is characterised in that described this method
Including
Set up can graphically, the index system of credibility evaluation interface of stratification, can by Complex simulation systems
The index decomposition that reliability is assessed emulates the reliability assessment index of element model for each, and calculates each index
Weight;
The data collected in each l-G simulation test and reliability assessment result are carried out in the form of data set
Organization and management;
The information of multiple assessment data sources is merged, it is bottom-up successively comprehensively to obtain analogue system
Reliability assessment result;
Set up the consistency analysis of reliability assessment result and the iterative process of feedback modifiers.
2. robust appraisal procedure according to claim 1, it is characterised in that described this method enters one
Step includes
Selective shows certain reliability assessment for once assessing each evaluation index at all levels afterwards
As a result and analogue system confidence level, or on each level after repeatedly assessment result is average each index it is credible
The confidence level of degree and analogue system.
3. robust appraisal procedure according to claim 1, it is characterised in that the foundation can figure
Change, the index system of credibility evaluation of stratification and each index weights of calculating further comprise
Confidence of simulation system evaluation index is decomposed into the confidence level of each subsystem of composition analogue system
Evaluation index;
By the confidence level that the reliability assessment index decomposition of each subsystem is the subsystem more to next stage
Evaluation index, untill decomposing the reliability assessment index of emulation element model of minimum;
To each emulation element model according to modeling requirement, conceptual model, Mathematical Modeling, simulation model
And the several aspects of simulation result are estimated index decomposition, the confidence level of each emulation element model is obtained
Evaluation index;
The evaluation index composing indexes system of each comprehensive element model, then to the importance of evaluation index
Compared two-by-two and set up Judgement Matrix, calculated by characteristic value and obtain evaluation index weight.
4. robust appraisal procedure according to claim 1, it is characterised in that in the l-G simulation test
The data and the assessment result data of different reliability assessment tasks collected carry out tissue according to data set
Management;
The description information of data set further comprises dataset name, l-G simulation test task or assesses task name
Number of folders and data file number in title, data name, creation time, storing path, data set
Amount.
5. robust appraisal procedure according to claim 1, it is characterised in that methods described is to multiple
Qualitative, the quantitative assessment information for assessing data source is merged;
The assessment data source further comprises l-G simulation test data, real system test data and expert's warp
Test knowledge;
The multiple data source of assessing is for the fusion method formula of some evaluation index:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
6. a kind of robust assessment system of Complex simulation systems confidence level, it is characterised in that the system includes
Index system of credibility evaluation module, assessment data management module, confidence level comprehensive assessment module and assessment
Interpretation of result display module;
The index system of credibility evaluation module, for set up can graphically, the confidence level of stratification comments
Assessment system interface, is each emulation element mould by the index decomposition of Complex simulation systems reliability assessment
The reliability assessment index of type, and calculate each index weights;
The assessment data management module, for the data and confidence level that will be collected in each l-G simulation test
Assessment result is organized and managed in the form of data set;
The confidence level comprehensive assessment module, for the information of multiple assessment data sources to be merged, from
The lower reliability assessment result for successively comprehensively obtaining analogue system upwards;
The assessment result analyzes display module, each layer after once being assessed for selective display
The reliability assessment result and the confidence level of analogue system of each evaluation index on secondary, or repeatedly assess knot
The confidence level and the confidence level of analogue system of each index on each level after fruit is average.
7. robust assessment system according to claim 6, it is characterised in that the reliability assessment
Evaluation index hierarchical relationship is set up in index system module to further comprise
Confidence of simulation system evaluation index is decomposed into the confidence level of each subsystem of composition analogue system
Evaluation index;
By the confidence level that the reliability assessment index decomposition of each subsystem is the subsystem more to next stage
Evaluation index, untill decomposing the reliability assessment index of emulation element model of minimum;
To each emulation element model according to modeling requirement, conceptual model, Mathematical Modeling, simulation model
And the several aspects of simulation result are estimated index decomposition, the confidence level of each emulation element model is obtained
Evaluation index;
The evaluation index composing indexes system of each comprehensive element model, then to the importance of evaluation index
Compared two-by-two and set up Judgement Matrix, calculated by characteristic value and obtain evaluation index weight.
8. robust assessment system according to claim 6, it is characterised in that in the l-G simulation test
The data and the assessment result data of different reliability assessment tasks collected carry out tissue according to data set
Management;
The description information of data set further comprises dataset name, l-G simulation test task or assesses task name
Number of folders DBMS number of files in title, data name, creation time, storing path, data set
Amount.
9. robust assessment system according to claim 6, it is characterised in that the system is to multiple
Qualitative, the quantitative assessment information of data source is merged;
In the assessment data management module
Multiple assessment data sources further comprise that l-G simulation test data, actual experimental data and expertise are known
Know;
The multiple data source of assessing is for the fusion method formula of some evaluation index:
In formula:P ∈ { 1,2 ..., p }, to assess data source;
For evaluation index SmnRelative to the probability that purpose of appraisals can receive (yes);
For p-th assess data source reflect for evaluation index SmnKnowledge degree;
For p-th assess data source reflect for evaluation index SmnTrusting degree;
The knowledge degree, for quantitative target, for representing the deviation journey between emulation data and test data
Degree;For qualitative index, for representing knowledge depth of the assessment experts in the field;
The degree of belief, for quantitative target, the confidence level check results for representing l-G simulation test data;
For qualitative index, the trusting degree given to assessing object for representing assessment experts.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610012759.0A CN106960065A (en) | 2016-01-11 | 2016-01-11 | The robust appraisal procedure and system of a kind of Complex simulation systems confidence level |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610012759.0A CN106960065A (en) | 2016-01-11 | 2016-01-11 | The robust appraisal procedure and system of a kind of Complex simulation systems confidence level |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106960065A true CN106960065A (en) | 2017-07-18 |
Family
ID=59480469
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610012759.0A Pending CN106960065A (en) | 2016-01-11 | 2016-01-11 | The robust appraisal procedure and system of a kind of Complex simulation systems confidence level |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106960065A (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108647878A (en) * | 2018-05-08 | 2018-10-12 | 哈尔滨工业大学 | A kind of credibility evaluation method of training simulator |
CN108683564A (en) * | 2018-05-21 | 2018-10-19 | 中国科学院信息工程研究所 | A kind of network (WSN) emulation system credibility evaluation method based on Multidimensional decision-making attribute |
CN109241617A (en) * | 2018-09-03 | 2019-01-18 | 哈尔滨工业大学 | A kind of Guidance and control semi-matter simulating system credibility evaluation method based on genealogical classification |
CN109359388A (en) * | 2018-10-18 | 2019-02-19 | 北京仿真中心 | A kind of Complex simulation systems credibility evaluation method |
CN109918833A (en) * | 2019-03-21 | 2019-06-21 | 中国空气动力研究与发展中心 | A kind of quantitative analysis method of numerical simulation confidence |
CN109960863A (en) * | 2019-03-15 | 2019-07-02 | 北京航空航天大学 | A kind of Complex simulation systems credibility evaluation method based on network topology path |
CN111308912A (en) * | 2020-03-15 | 2020-06-19 | 西安爱生技术集团公司 | Reliability evaluation method for anti-radiation unmanned aerial vehicle guidance semi-physical simulation system |
CN112001067A (en) * | 2020-08-03 | 2020-11-27 | 北京机电工程研究所 | Uncertainty measurement-based reliability evaluation method for complex equipment simulation system |
CN112199842A (en) * | 2020-11-11 | 2021-01-08 | 中国电子科技集团公司第二十八研究所 | Task-oriented-based complex simulation system reliability evaluation method |
CN112558927A (en) * | 2020-12-09 | 2021-03-26 | 中国电子科技集团公司第十五研究所 | Software reliability index distribution method and device based on layer-by-layer decomposition method |
CN112672299A (en) * | 2020-12-09 | 2021-04-16 | 电子科技大学 | Sensor data reliability evaluation method based on multi-source heterogeneous information fusion |
CN112948974A (en) * | 2021-03-09 | 2021-06-11 | 北京机电工程研究所 | Aircraft performance evaluation method and system based on evidence theory |
CN113130017A (en) * | 2020-06-05 | 2021-07-16 | 北京世纪金光半导体有限公司 | Crystal growth analysis method and system based on artificial intelligence |
CN116432472A (en) * | 2023-04-27 | 2023-07-14 | 中国人民解放军国防科技大学 | Comprehensive analysis and evaluation system and method for target range joint simulation test data |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070101318A1 (en) * | 2005-10-31 | 2007-05-03 | Fujitsu Limited | Multi-core-model simulation method, multi-core model simulator, and computer product |
CN101620645A (en) * | 2009-08-17 | 2010-01-06 | 王钰 | Method and system of large-scale simulation electronic information system architecture |
CN102496069A (en) * | 2011-12-07 | 2012-06-13 | 山东电力集团公司青岛供电公司 | Cable multimode safe operation evaluation method based on fuzzy analytic hierarchy process (FAHP) |
CN103246762A (en) * | 2013-04-10 | 2013-08-14 | 哈尔滨工程大学 | Method of comprehensive evaluation for simulation credibility of electric propulsion system |
CN104077456A (en) * | 2014-07-06 | 2014-10-01 | 哈尔滨工业大学 | Method for performance evaluation of spacecraft attitude control ground simulation system |
CN105138733A (en) * | 2015-07-30 | 2015-12-09 | 河北工业大学 | Driving comfort based two-lane highway traffic safety evaluation method |
-
2016
- 2016-01-11 CN CN201610012759.0A patent/CN106960065A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070101318A1 (en) * | 2005-10-31 | 2007-05-03 | Fujitsu Limited | Multi-core-model simulation method, multi-core model simulator, and computer product |
CN101620645A (en) * | 2009-08-17 | 2010-01-06 | 王钰 | Method and system of large-scale simulation electronic information system architecture |
CN102496069A (en) * | 2011-12-07 | 2012-06-13 | 山东电力集团公司青岛供电公司 | Cable multimode safe operation evaluation method based on fuzzy analytic hierarchy process (FAHP) |
CN103246762A (en) * | 2013-04-10 | 2013-08-14 | 哈尔滨工程大学 | Method of comprehensive evaluation for simulation credibility of electric propulsion system |
CN104077456A (en) * | 2014-07-06 | 2014-10-01 | 哈尔滨工业大学 | Method for performance evaluation of spacecraft attitude control ground simulation system |
CN105138733A (en) * | 2015-07-30 | 2015-12-09 | 河北工业大学 | Driving comfort based two-lane highway traffic safety evaluation method |
Non-Patent Citations (5)
Title |
---|
刘博元: "仿真系统可信度的灰色层次分析法", 《信息与电脑》 * |
刘博元: "复杂产品仿真可信度鲁棒性评估方法", 《系统仿真技术》 * |
曲慧杨: "复杂装备虚拟采办平台技术研究", 《系统仿真技术及其应用》 * |
李莉红: "船舶电力推进系统仿真可信度评估研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
翟慎涛: "仿真模型可信度评估指标体系研究", 《系统仿真学报》 * |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108647878B (en) * | 2018-05-08 | 2021-10-01 | 哈尔滨工业大学 | Credibility evaluation method for training simulator |
CN108647878A (en) * | 2018-05-08 | 2018-10-12 | 哈尔滨工业大学 | A kind of credibility evaluation method of training simulator |
CN108683564B (en) * | 2018-05-21 | 2020-05-22 | 中国科学院信息工程研究所 | Network simulation system reliability evaluation method based on multidimensional decision attributes |
CN108683564A (en) * | 2018-05-21 | 2018-10-19 | 中国科学院信息工程研究所 | A kind of network (WSN) emulation system credibility evaluation method based on Multidimensional decision-making attribute |
CN109241617A (en) * | 2018-09-03 | 2019-01-18 | 哈尔滨工业大学 | A kind of Guidance and control semi-matter simulating system credibility evaluation method based on genealogical classification |
CN109359388A (en) * | 2018-10-18 | 2019-02-19 | 北京仿真中心 | A kind of Complex simulation systems credibility evaluation method |
CN109960863B (en) * | 2019-03-15 | 2021-04-20 | 北京航空航天大学 | Complex simulation system credibility evaluation method based on network topology path |
CN109960863A (en) * | 2019-03-15 | 2019-07-02 | 北京航空航天大学 | A kind of Complex simulation systems credibility evaluation method based on network topology path |
CN109918833B (en) * | 2019-03-21 | 2022-10-21 | 中国空气动力研究与发展中心 | Quantitative analysis method for numerical simulation reliability |
CN109918833A (en) * | 2019-03-21 | 2019-06-21 | 中国空气动力研究与发展中心 | A kind of quantitative analysis method of numerical simulation confidence |
CN111308912B (en) * | 2020-03-15 | 2023-06-27 | 西安爱生技术集团公司 | Credibility evaluation method of anti-radiation unmanned aerial vehicle guidance semi-physical simulation system |
CN111308912A (en) * | 2020-03-15 | 2020-06-19 | 西安爱生技术集团公司 | Reliability evaluation method for anti-radiation unmanned aerial vehicle guidance semi-physical simulation system |
CN113130017B (en) * | 2020-06-05 | 2024-01-30 | 北京星云联众科技有限公司 | Crystal growth analysis method and system based on artificial intelligence |
CN113130017A (en) * | 2020-06-05 | 2021-07-16 | 北京世纪金光半导体有限公司 | Crystal growth analysis method and system based on artificial intelligence |
CN112001067A (en) * | 2020-08-03 | 2020-11-27 | 北京机电工程研究所 | Uncertainty measurement-based reliability evaluation method for complex equipment simulation system |
CN112199842B (en) * | 2020-11-11 | 2022-10-04 | 中国电子科技集团公司第二十八研究所 | Task-oriented-based complex simulation system reliability evaluation method |
CN112199842A (en) * | 2020-11-11 | 2021-01-08 | 中国电子科技集团公司第二十八研究所 | Task-oriented-based complex simulation system reliability evaluation method |
CN112672299B (en) * | 2020-12-09 | 2022-05-03 | 电子科技大学 | Sensor data reliability evaluation method based on multi-source heterogeneous information fusion |
CN112672299A (en) * | 2020-12-09 | 2021-04-16 | 电子科技大学 | Sensor data reliability evaluation method based on multi-source heterogeneous information fusion |
CN112558927A (en) * | 2020-12-09 | 2021-03-26 | 中国电子科技集团公司第十五研究所 | Software reliability index distribution method and device based on layer-by-layer decomposition method |
CN112558927B (en) * | 2020-12-09 | 2024-02-20 | 中国电子科技集团公司第十五研究所 | Software reliability index distribution method and device based on layer-by-layer decomposition method |
CN112948974A (en) * | 2021-03-09 | 2021-06-11 | 北京机电工程研究所 | Aircraft performance evaluation method and system based on evidence theory |
CN112948974B (en) * | 2021-03-09 | 2023-09-01 | 北京机电工程研究所 | Aircraft performance evaluation method and system based on evidence theory |
CN116432472A (en) * | 2023-04-27 | 2023-07-14 | 中国人民解放军国防科技大学 | Comprehensive analysis and evaluation system and method for target range joint simulation test data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106960065A (en) | The robust appraisal procedure and system of a kind of Complex simulation systems confidence level | |
Xu et al. | Scenario tree reduction in stochastic programming with recourse for hydropower operations | |
Hong et al. | Estimation of pavement performance deterioration using Bayesian approach | |
Haas et al. | Updating the decision aids for tunneling | |
Chen | A prioritized aggregation operator-based approach to multiple criteria decision making using interval-valued intuitionistic fuzzy sets: A comparative perspective | |
CA2992274C (en) | Ensemble based decision making | |
Ridler et al. | Data assimilation framework: Linking an open data assimilation library (OpenDA) to a widely adopted model interface (OpenMI) | |
Hamzeh et al. | Imprecise earned duration model for time evaluation of construction projects with risk considerations | |
Leontaris et al. | ANDURIL—A MATLAB toolbox for ANalysis and Decisions with UnceRtaInty: Learning from expert judgments | |
CN110322142A (en) | A kind of big data air control model and inline system configuration technology | |
Quilty et al. | A stochastic conceptual-data-driven approach for improved hydrological simulations | |
Kaka et al. | Development of a company‐level dynamic cash flow forecasting model (DYCAFF) | |
Okkan | Application of Levenberg-Marquardt optimization algorithm based multilayer neural networks for hydrological time series modeling | |
CN101561904A (en) | Process data-based method and system for determining cost of software project | |
Stehlik et al. | On ecosystems dynamics | |
Korde et al. | Probabilistic decline curve analysis in the permian basin using Bayesian and approximate Bayesian inference | |
CN117892945A (en) | Construction engineering cost measuring and calculating method based on neural network | |
CN106257506A (en) | Three layers of associating of big data quantity prediction dynamically select optimal models method | |
Mihon et al. | Grid based hydrologic model calibration and execution | |
Renault | PEGASOS refinement project: new findings and challenges from a PSHA for Swiss nuclear power plants | |
Ead | Using Monte Carlo simulation to evaluate performance of forecasting models in project control | |
Lynam | Scientific measurements and villagers' knowledge: an integrative multi-agent model from the semi-arid areas of Zimbabwe | |
Chlumsky | Rigorous validation of hydrologic models in support of decision-making | |
Nannini et al. | Improving the accuracy of project estimates at completion using the Gompertz function. International Research Network on Organizing by Projects (IRNOP) 2017, UTS ePRESS, Sydney: NSW | |
EP3323092B1 (en) | Ensemble based decision making |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170718 |