CN114580162A - Equipment-oriented digital twin dynamic credibility calculation method and system - Google Patents

Equipment-oriented digital twin dynamic credibility calculation method and system Download PDF

Info

Publication number
CN114580162A
CN114580162A CN202210177451.7A CN202210177451A CN114580162A CN 114580162 A CN114580162 A CN 114580162A CN 202210177451 A CN202210177451 A CN 202210177451A CN 114580162 A CN114580162 A CN 114580162A
Authority
CN
China
Prior art keywords
model
credibility
reliability
value
digital twin
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
Application number
CN202210177451.7A
Other languages
Chinese (zh)
Inventor
张霖
陆涵
崔晋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN202210177451.7A priority Critical patent/CN114580162A/en
Publication of CN114580162A publication Critical patent/CN114580162A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Economics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Computational Mathematics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a device-oriented digital twin dynamic credibility calculation method and a device-oriented digital twin dynamic credibility calculation system, which are applied to the field of modeling simulation and comprise the following steps: determining the life cycle stage characteristics of the model; analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes; carrying out reliability calculation on the plurality of evaluation indexes to obtain a plurality of reliability index values; and aggregating the plurality of reliability index values and outputting a reliability calculation value. The method and the system can fully reflect the complexity, the dynamic property, the interactivity and the uncertainty of the digital twin through the analysis and the data of the user requirement, the evaluation of a basic unit model, an evolution unit model, an integration model and a support; factors which can change the credibility of the digital twin are mainly considered through multi-layer dynamic judgment and time binding of a data body and the credibility, wherein the factors comprise demand change, time sequence drive, evolution event drive and the like, so that the credibility of the digital twin emphasizing dynamic interaction has practical effectiveness.

Description

Equipment-oriented digital twin dynamic credibility calculation method and system
Technical Field
The invention relates to the field of modeling simulation, in particular to a digital twin dynamic credibility calculation method and system for equipment.
Background
Digital Twin (DT), which refers to a simulation model that is highly consistent with a physical object, has been a focus of research in recent years. The digital twin assimilates the collected real-time data and thus remains consistent with the physical object in real-time after the model is built. Therefore, the digital twin system can perform simulation test on the twin object which is more fit with the real state in the virtual environment, and a simulation result with more practical guiding significance is obtained.
The digital twin belongs to the field of Modeling and Simulation (M & S). In the M & S field, if the model is not subjected to credibility evaluation, the model cannot play a value in practice. Digital twinning systems also require a set of trusted evaluation methods. Only if the confidence level of the system is ensured to be higher than the use threshold value, the digital twin can really exert the application value of the system.
The core of the digital twinning system is a digital twinning model, which is highly complex and dynamic. On one hand, the physical equipment is a multi-granularity object fused in multidisciplinary fields, and the corresponding digital twin model has the characteristics of multiple physical domains, multiple scales and multiple resolutions. As the degree of integration of the model further increases, the complexity of the digital twin system also increases dramatically. The high complexity of the digital twin model means that the credible evaluation of the digital twin model is a system project, and a great variety of models in different life cycle stages are evaluated from multiple dimensions on the basis of grasping the emphasis of the demand. On the other hand, the digital twin model can keep consistent with the physical object in the whole life cycle according to the dynamic interaction data collected in real time, and an optimization control signal is fed back to the digital twin model according to the operation requirement of the physical system. Therefore, the digital twin model has high dynamics and strong interactivity, which also results in high uncertainty. The dynamics, the interactivity and the uncertainty put higher requirements on the credible evaluation method of the digital twin system. The dynamics also make the plausibility of the digital twin system variable. As the system evolves, its confidence value evolves in real time accordingly.
Therefore, how to provide an equipment-oriented digital twin dynamic credibility calculation method and system capable of effectively integrating complexity, dynamics, interactivity and uncertainty of a digital twin model in a digital twin system is a problem to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides an apparatus-oriented digital twin dynamic credibility calculation method and system. Through analysis of user requirements, data evaluation, basic unit model evaluation, evolution unit model evaluation, integrated model evaluation and support evaluation, credibility related elements of the digital twin can be comprehensively and systematically investigated, finally calculated credibility can fully reflect complexity, dynamics, interactivity and uncertainty of the digital twin, and higher reference value is given to users. Meanwhile, dynamic value calculation is carried out through multilayer dynamic judgment and time binding of a data body and the reliability, factors which can change the digital twin reliability are intensively studied, including requirement change, time sequence drive, evolution event drive and the like, so that the reliability of the digital twin which emphasizes dynamic interaction also has effectiveness, and the actual physical significance of the reliability is enhanced.
In order to achieve the purpose, the invention adopts the following technical scheme:
an equipment-oriented digital twin dynamic credibility calculation method comprises the following steps:
and determining the life cycle stage characteristics of the model.
And analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes.
And carrying out reliability calculation on the plurality of evaluation indexes to obtain a plurality of reliability index values.
And aggregating the plurality of reliability index values and outputting a reliability calculation value.
Optionally, in step 3), the multiple evaluation indexes are data, a basic unit model, an evolution unit model, an integration model, and support elements.
The data and the basic unit model are used for evaluating the basic model and the data which are not accessed to the actual system; the evolution unit model, the integration model and the support element are used for evaluating a digital twin which operates in a closed loop, and a data sample at the time T needs to be input.
Obtaining the data credibility index value CdAnd a reliability index value C of the basic unit modelbuAnd a reliability index value C of the evolution unit modelruAnd an integrated model credibility index value CimAnd a support element reliability index value CS。
Optionally, in step 3), the reliability index value C of the datadThe calculation method comprises the following steps:
all data involved in the digital twin are evaluated in a credible manner to ensure that the benchmark of the subsequent model evolution is correct; the calculation method is divided into a data source and a data body; the data source refers to a data source, including acquisition equipment such as a sensor, organizations related to people and software application, and an important index for evaluating the credibility of the data source is authority Ad(ii) a A data body refers to a data body arriving at the system, mainly from its fidelity F, regardless of the data source and data formdAnd the value of use VdStarting to investigate credibility; dividing the data into a data source and a data body, respectively carrying out credibility evaluation on authority of the data source and fidelity and practical value of the data body, and synthesizing credibility evaluation results to obtain credibility index value C of the datadThe formula is as follows:
Cd＝Ad*(αd1Fd+αd2Vd)；
wherein C isdAs a reliability index value of the data, αd1、αd2The weight coefficient is determined according to the relative contribution of the corresponding index to the user demand, the sum of the two weights is 1, Ad、Fd、VdCredibility index values corresponding to authority, fidelity and use value indexes respectively, and the value range of the credibility index values is [0,1 ]]。
Optionally, in step 3), a reliability index value C of the basic unit modelbuThe calculation method comprises the following steps:
a multi-dimensional analysis method is generally adopted, different mutually orthogonal views are looked at, and key elements influencing the credibility of the model are deeply analyzed; the basic unit model is credible, and is a model constructed based on modeling data (without considering real-time data updating), and reflects the capability degree of the real operation condition of an actual physical object according to development requirements; starting from two specific dimensions of the subject field and the model characteristics, selecting a specific credibility assessment method aiming at different characteristic models according to key contents needing attention in model credibility assessment.
Aiming at the subject field view, the basic unit model is divided into different types of sets C according to the subject field to which the basic unit model belongs, wherein the sets C are composed of part or all of the subject fields of acoustic Sc electric Ec gas Ac liquid Hc dynamic Mc thermal Tc and are represented as follows:
Figure BDA0003519467610000031
after the division is finished, designing a multi-index hierarchical evaluation method of historical use data of the comprehensive model, field expert experience and actual operation data of the model in the field aiming at each unit model of the specific category to obtain a credibility value C under the view of the subject fieldbud(ii) a The principle of quantitative and qualitative combination and direct and indirect combination is followed, the historical use data of the model is the basis of quantitative and indirect evaluation, the field expert experience is the basis of qualitative and indirect evaluation, and the actual operation data of the model is the basis of quantitative and direct evaluation.
Aiming at a model characteristic view, according to the difference of the characteristics of a digital twin unit model, the digital twin unit model is divided into a first model, a second model, a third model and three model categories, wherein the first model is a model represented by a common data driving model such as a neural network and the like, has an unknown mechanism and only takes input and output as guidance; the second model is a model represented by a common descriptive model such as CAD (computer-aided design) and the like, a mathematical equation model with known principles and the like, and has a clear and controllable mechanism; the third model is a model represented by a common industrial mechanism and industrial big data model, and has the characteristics of the first model and the second model. Designing and developing different credible evaluation algorithms according to the three types of models; for example, by means of a mathematical statistics method, a time-frequency domain analysis method, a deep learning algorithm and the like, a black box model credibility evaluation method with input and output as a guide is researched; researching a white box model credible evaluation method taking a mechanism as a core by means of model key parameter proofreading, mathematical theorem demonstration and the like; by means of methods such as invisible feature mining and incidence relation reasoning, the credible evaluation method for researching the gray box model has the following formulas:
Cbum＝αm1Bm+αm2Wm+αm3Gm；
wherein, CbumThe reliability index value under the model characteristic view is obtained; alpha is alpham1、αm2、αm3Respectively providing contribution weights of the first model, the second model and the third model to the use requirement under the user specified scene, wherein the sum of the three terms is 1; b ism、Wm、GmThe reliability index values of the first model, the second model and the third model are respectively.
Reliability index value C of basic unit modelbuComprises the following steps:
Cbu＝(Cbud+Cbum)/2。
optionally, in step 3), the reliability index C of the evolution unit modelruThe calculation method comprises the following steps:
analyzing a dynamic data assimilation core technical means, and selecting a credible key evaluation index and method for data assimilation; the evolution model is credible, which means the capability degree that the constructed model can be accurately self-evolved in real time under the drive of real-time dynamic data from an actual physical object and keep consistency with the actual physical object. The digital twin unit model is required to evolve by itself according to the state change of the actual physical system, and is embodied to perform reconfiguration of model parameters and even reconstruction of the model according to data acquired by a physical system sensor, various software instruction data and the like.The core technical approach of model evolution evaluation can be divided into three aspects, one aspect is 'fast', namely the model performs data interaction in real time or quasi-real time according to the state change of an actual physical system, the second aspect is 'quasi', namely the digital twin unit model can accurately calculate the change range, amplitude and the like of model parameters through collected actual system data, and the third aspect is 'stable', namely the model operation should be kept stable all the time, and the most representative aspects comprise model sensitivity analysis, robustness analysis and the like, so that the model is ensured not to have a destructive result which is difficult to use due to small disturbance; dividing the core technical approach of evaluation of the evolution unit model into three aspects of 'fast', 'accurate' and 'stable', respectively evaluating the credibility of the evolution unit model, and synthesizing the credibility evaluation result to obtain the credibility index value C of the evolution unit modelruThe calculation formula is as follows:
Cru＝αr1Fr+αr2Ar+αr3Sr；
wherein, CruThe reliability index value of the evolution unit model is obtained; alpha is alphar1、αr2、αr3The indexes are respectively the weighted values corresponding to the fast index, the quasi index and the stable index, and the sum is 1; fr、Ar、SrThe reliability index values corresponding to the rapidity index, the accuracy index and the stability index are respectively.
Optionally, in step 3), a reliability index value C of the integrated model is obtainedimThe calculation method comprises the following steps:
on the basis of a single model credible evaluation result of a digital twin system, a concept model, a mathematical model and a simulation model in the digital twin system with a full life cycle are provided with two types of integrated models, wherein one type of integrated model is an integrated model formed by fusing multidisciplinary models facing to the same entity, and the other type of integrated model is an integrated model with larger granularity formed by combining a plurality of models.
Aiming at an integrated model formed by fusing multidisciplinary models of the same entity, because a digital twin system is arranged, the integrated model relates to a simulation model in multidisciplinary and multidisciplinary fields such as machinery, electronics, power, materials, control and the like, and each discipline adopts a simulation model which does not adopt the multidisciplinary modelWith the same modeling tool and data model, multidisciplinary model fusion requires a large amount of data conversion, which can affect the consistency and relevance of the model, and further affect the credibility of the multidisciplinary fusion model; aiming at the problems of interdigitation, difficult comprehensive coordination and the like of multidisciplinary fields, a disciplinary identification mechanism of a model is used, multidisciplinary connection relations such as disciplinary association, disciplinary mapping and the like among multidisciplinary characteristic research models are comprehensively considered, and a fusion mechanism of multidisciplinary models is mined based on data exchange, information sharing and interoperation among multidisciplinary and multidisciplinary heterogeneous models. Analyzing complex coupling and dynamic constraint among the multidisciplinary models, connecting modeling requirements and characteristics of each field, and using a relational network construction method among the multidisciplinary models according to a complex knowledge structure among the multiple fields. Unified quantitative representation is carried out on the multidisciplinary model fusion relation by using a multidisciplinary model fusion effect evaluation method based on knowledge, and the credibility index value C of the integrated model is solved based on the credibility of the unit model and the multidisciplinary fusion relation according to a multidisciplinary fusion model credibility quantitative calculation methodimd。
Aiming at a larger-granularity integrated model formed by aggregating a plurality of models, parts, components and functional units in the equipment are assembled to form complete equipment, and a corresponding equipment digital twin system is formed by combining the unit models into the integrated model to form a whole twin body; a series of uncertainties exist in complex interoperation between unit models, so that the unit models are difficult to trust and the combined models are difficult to trust. Therefore, on the basis of unit model credibility evaluation, the credibility of the integrated model is further evaluated according to the coupling and incidence relation among the models; the evaluation can be carried out based on the time sequence relation, the connection mechanism and the implicit association relation among the submodels in a transparent and non-transparent combined mode. In a transparent model combination mode, the connection and interoperation processes among unit models are known, the model dependency relationship and programs can be quantized according to the coupling relationship of sub models, the interoperation process and dynamic time sequence constraints, and the reliability of the combination model is deduced by adopting methods such as a Bayesian network and the like; under the model non-transparent combination mode, the connection and interaction between unit models are unknown, and the unit models need to be combined according to the combination modelThe reliability of the model is well evaluated by the integral input and output, if a machine learning method such as a generation countermeasure network is adopted, a implicit association relation is quantitatively extracted according to simulation historical data, and a function mapping relation between the implicit association relation and the combined model reliability index value is mined, so that the combined model reliability index value C is calculated by integrating the coupling mechanism and the implicit association relation between submodelsimg。
Credibility index value C of integrated modelimThe calculation formula is as follows:
Cim＝(Cimd+Cimg)/2。
optionally, in step 3), a reliability index value C of the support element is obtainedsThe calculation method comprises the following steps:
carrying out credibility evaluation on the support elements according to the uncertainty and the maturity respectively, and integrating the credibility evaluation result to obtain a credibility index value C of the support elementss。
Including uncertainty CuAnd maturity Cma(ii) a The digital twin system is a complex system, and uncertainty is inevitably introduced in the whole life cycle processes of construction, operation and the like, so that certain risk is brought to the operation of the system; therefore, a trusted digital twinning system must be tough against the risk of uncertainty; the principle of the credible evaluation of the uncertainty analysis is to ensure that the construction process of the digital twin unit model carries out perfect uncertainty management and minimize the failure probability of the model caused by uncertainty risks. The smaller the probability of failure, meaning the smaller the uncertainty, CuThe smaller the value of (A), CuHas a value range of [0,1 ]](ii) a Starting from the process analysis aiming at the uncertainty analysis, the adequacy and the completeness degree of uncertainty are considered in the construction and operation process of the digital twin system and are used as the key reference of the operation toughness of the digital twin system.
For the maturity, when the same digital twin system is evaluated and perfected for multiple times under various application scenes and different requirements, the system maturity is gradually improved, so that the system has higher reliability when meeting the requirements of scenes which are not practiced in the same category; confidence level isUnder a certain time slice, a single numerical value is evaluated according to a specific scene and specific requirements, and the maturity is the accumulated credible experience of the system from the perspective of the whole time axis of the system evolution; the higher the maturity of the system, the higher its confidence in the applicable field, CmaThe higher the value, the range of [0,1 ]]。
Reliability index value C of support elementSComprises the following steps:
Cs＝(Cma-Cu+1)/2。
optionally, in step 4), the calculation method for aggregating a plurality of reliability index values and outputting a calculated reliability value includes:
CT＝(4*(αT1Cd+αT2Cbu+αT3Cru+αT4Cim)+Cs/5)；
wherein, CTCalculating a value for the confidence of time T; alpha is alphaT1、αT2、αT3、αT4Respectively, data reliability index values CdAnd a reliability index value C of the basic unit modelbuAnd a reliability index value C of the evolution unit modelruAnd an integrated model reliability index value CimThe corresponding weights, the sum of which is 1; the support element is an indirect influence on the reliability, and the weight thereof is set to 20%.
Optionally, in step 4), the reliability calculation value C at the output time T is calculatedTAnd then, the method further comprises the following steps:
judging whether the user requirement is changed, namely, the function expectation of the digital twin by the user in the same life cycle stage is changed, if so, re-evaluating from the user requirement analysis, at the moment, the digital twin is also possible to be reconfigured, reconstructed and the like, therefore, the evaluation of the basic model part is also performed again, returning to the step 2), and outputting the credibility calculation value C of the time T +1T+1。
If no change exists, whether the digital twin reaches the evolution condition or not needs to be judged; the evolution condition is composed of time drive and event drive; when the accumulated unexploited duration reaches a threshold value or some characteristics of the model are not availableWhen a new round of data is matched, the digital twin model evolves, and the reliability is updated at the moment; in this case, since the basic model has not been adjusted, only the evolution unit model, the integration model, and the support element may be evaluated for a new round, and the process returns to step 3), and only the digital twin operating in the closed loop may be evaluated, and the reliability calculation value C at the time T +1 may be outputT+1。
The invention also provides an equipment-oriented digital twin dynamic credibility calculation system, which comprises:
a determination module: for determining the life cycle stage characteristics of the model.
An analysis module: the method is used for analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes.
A calculation module: and the method is used for carrying out reliability calculation on the plurality of evaluation indexes to obtain a plurality of reliability index values.
A polymerization module: the reliability calculation value is used for aggregating a plurality of reliability index values and outputting the reliability calculation value.
A judging module: and the user demand judging module is used for judging whether the user demand changes or not after the credibility calculation value is output.
Compared with the prior art, the method and the system for calculating the dynamic credibility of the digital twin facing to the equipment can comprehensively and systematically examine the credibility related elements of the digital twin through analysis, data evaluation, basic unit model evaluation, evolution unit model evaluation, integrated model evaluation and support evaluation of user requirements, so that the finally calculated credibility can fully reflect the complexity, dynamics, interactivity and uncertainty of the digital twin and provide higher reference value for users. Meanwhile, dynamic value calculation is carried out through multilayer dynamic judgment and time binding of a data body and the reliability, factors which can change the digital twin reliability are intensively studied, including requirement change, time sequence drive, evolution event drive and the like, so that the reliability of the digital twin which emphasizes dynamic interaction also has effectiveness, and the actual physical significance of the reliability is enhanced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow chart of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
The embodiment of the invention discloses a device-oriented digital twin dynamic credibility calculation method, which comprises the following steps:
and determining the life cycle stage characteristics of the model.
And analyzing the user demand based on the life cycle stage characteristics, carrying out quantitative conversion on the user demand to obtain a demand index, and carrying out multi-dimensional refinement on the demand index according to an evaluation object to obtain a plurality of evaluation indexes.
The plurality of evaluation indexes are data, a basic unit model, an evolution unit model, an integration model and a support element.
The data and the basic unit model are used for evaluating the basic model and the data which are not accessed to the actual system; the evolution unit model, the integration model and the support element are used for evaluating a digital twin which operates in a closed loop, and a data sample at the time T needs to be input.
And carrying out reliability calculation on the plurality of evaluation indexes.
Confidence indicator of dataValue CdThe calculation method comprises the following steps:
all data involved in the digital twin are evaluated in a credible manner to ensure that the benchmark of the subsequent model evolution is correct; the calculation method is divided into a data source and a data body; the data source refers to a data source, and comprises acquisition equipment such as a sensor, an organization mechanism related to a person and software application, and the authority A is an important index for evaluating the credibility of the data sourced(ii) a A data body refers to a data body arriving at the system, mainly from its fidelity F, regardless of the data source and data formdAnd the value of use VdStarting to investigate credibility; dividing the data into a data source and a data body, respectively carrying out credibility evaluation on authority of the data source and fidelity and practical value of the data body, and synthesizing credibility evaluation results to obtain credibility index value C of the datadThe formula is as follows:
Cd＝Ad*(αd1Fd+αd2Vd)；
wherein C isdAs a reliability index value of the data, αd1、αd2The weight coefficient is determined according to the relative contribution of the corresponding index to the user demand, the sum of the two weights is 1, Ad、Fd、VdCredibility index values corresponding to authority, fidelity and use value indexes respectively, and the value range of the credibility index values is [0,1 ]]。
Credibility index value C of basic unit modelbuThe calculation method comprises the following steps:
a multi-dimensional analysis method is generally adopted, different mutually orthogonal views are looked at, and key elements influencing the credibility of the model are deeply analyzed; the basic unit model is credible, and is a model constructed based on modeling data (without considering real-time data updating), and reflects the capability degree of the real operation condition of an actual physical object according to development requirements; starting from two specific dimensions of the subject field and the model characteristics, selecting a specific credibility assessment method aiming at different characteristic models according to key contents needing attention in model credibility assessment.
Aiming at the subject field view, the basic unit model is divided into different types of sets C according to the subject field to which the basic unit model belongs, wherein the sets C are composed of part or all of the subject fields of acoustic Sc electric Ec gas Ac liquid Hc dynamic Mc thermal Tc and are represented as follows:
Figure BDA0003519467610000101
after the division is finished, designing a multi-index hierarchical evaluation method of historical use data of the comprehensive model, field expert experience and actual operation data of the model in the field aiming at each unit model of the specific category to obtain a credibility value C under the view of the subject fieldbud(ii) a The principle of quantitative and qualitative combination and direct and indirect combination is followed, the historical use data of the model is the basis of quantitative and indirect evaluation, the field expert experience is the basis of qualitative and indirect evaluation, and the actual operation data of the model is the basis of quantitative and direct evaluation.
Aiming at a model characteristic view, according to the difference of the characteristics of a digital twin unit model, the digital twin unit model is divided into a first model, a second model, a third model and three model categories, wherein the first model is a model represented by a common data driving model such as a neural network and the like, has an unknown mechanism and only takes input and output as guidance; the second model is a model represented by a common descriptive model such as CAD (computer-aided design) and the like, a mathematical equation model with known principles and the like, and has a clear and controllable mechanism; the third model is a model represented by a common industrial mechanism and industrial big data model, and has the characteristics of the first model and the second model. Designing and developing different credible evaluation algorithms according to the three types of models; for example, by means of a mathematical statistics method, a time-frequency domain analysis method, a deep learning algorithm and the like, a black box model credibility evaluation method with input and output as a guide is researched; researching a white box model credible evaluation method taking a mechanism as a core by means of model key parameter proofreading, mathematical theorem demonstration and the like; by means of methods such as invisible feature mining and incidence relation reasoning, formulas such as a credible evaluation method for researching a gray box model are as follows:
Cbum＝αm1Bm+αm2Wm+αm3Gm；
wherein, CbumThe reliability index value under the model characteristic view is obtained; alpha is alpham1、αm2、αm3Respectively providing contribution weights of the first model, the second model and the third model to the use requirement under the user specified scene, wherein the sum of the three terms is 1; b ism、Wm、GmThe reliability index values of the first model, the second model and the third model are respectively.
Reliability index value C of basic unit modelbuComprises the following steps:
Cbu＝(Cbud+Cbum)/2。
credibility index C of evolution unit modelruThe calculation method comprises the following steps:
analyzing a dynamic data assimilation core technical means, and selecting a credible key evaluation index and a method for data assimilation; the evolution model is credible, which means the capability degree that the constructed model can be accurately self-evolved in real time under the drive of real-time dynamic data from an actual physical object and keep consistency with the actual physical object. The digital twin unit model is required to evolve by itself according to the state change of the actual physical system, and is embodied to perform reconfiguration of model parameters and even reconstruction of the model according to data acquired by a physical system sensor, various software instruction data and the like. The core technical approach of model evolution evaluation can be divided into three aspects, one aspect is 'fast', namely the model performs data interaction in real time or quasi-real time according to the state change of an actual physical system, the second aspect is 'quasi', namely the digital twin unit model can accurately calculate the change range, amplitude and the like of model parameters through collected actual system data, and the third aspect is 'stable', namely the model operation should be kept stable all the time, and the most representative aspects comprise model sensitivity analysis, robustness analysis and the like, so that the model is ensured not to have a destructive result which is difficult to use due to small disturbance; dividing the core technical approach of evaluation of the evolution unit model into three aspects of 'fast', 'accurate' and 'stable', respectively evaluating the credibility of the evolution unit model, and comprehensively evaluating the credibilityAs a result, a reliability index value C of the evolution unit model is obtainedruThe calculation formula is as follows:
Cru＝αr1Fr+αr2Ar+αr3Sr；
wherein, CruThe reliability index value of the evolution unit model is obtained; alpha is alphar1、αr2、αr3The weighted values corresponding to the indexes of fast, quasi and stable are respectively, and the sum of the weighted values is 1; fr、Ar、SrThe reliability index values corresponding to the rapidity index, the accuracy index and the stability index are respectively.
Credibility index value C of integrated modelimThe calculation method comprises the following steps:
on the basis of a single model credible evaluation result of a digital twin system, a concept model, a mathematical model and a simulation model in the digital twin system with a full life cycle are provided with two types of integrated models, wherein one type of integrated model is an integrated model formed by fusing multidisciplinary models facing to the same entity, and the other type of integrated model is an integrated model with larger granularity formed by combining a plurality of models.
Aiming at an integrated model after multi-subject model fusion of the same entity, as a digital twin system is arranged, the integrated model relates to simulation models in multiple subjects and multiple fields such as machinery, electronics, power, materials, control and the like, and different modeling tools and data models are adopted by each subject, the multi-subject model fusion needs a large amount of data conversion, the consistency and the relevance of the model can be influenced, and the reliability of the multi-subject fusion model is further influenced; aiming at the problems of interdigitation, difficult comprehensive coordination and the like of multidisciplinary fields, a disciplinary identification mechanism of a model is used, multidisciplinary connection relations such as disciplinary association, disciplinary mapping and the like among multidisciplinary characteristic research models are comprehensively considered, and a fusion mechanism of multidisciplinary models is mined based on data exchange, information sharing and interoperation among multidisciplinary and multidisciplinary heterogeneous models. Analyzing complex coupling and dynamic constraint among the multidisciplinary models, connecting modeling requirements and characteristics of each field, and using a relational network construction method among the multidisciplinary models according to a complex knowledge structure among the multiple fields. Effect evaluation using knowledge-based multidisciplinary model fusionThe method carries out unified quantitative characterization on the multidisciplinary model fusion relationship, and further solves the credibility index value C of the integrated model based on the credibility of the unit model and the multidisciplinary fusion relationship according to a multidisciplinary fusion model credibility quantitative calculation methodimd。
Aiming at a larger-granularity integrated model formed by aggregating a plurality of models, parts, components and functional units in the equipment are assembled to form complete equipment, and a corresponding equipment digital twin system is formed by combining the unit models into the integrated model to form a whole twin body; a series of uncertainties exist in complex interoperation between unit models, so that the unit models are difficult to trust and the combined models are difficult to trust. Therefore, on the basis of unit model credibility evaluation, the credibility of the integrated model is further evaluated according to the coupling and incidence relation among the models; the evaluation can be carried out based on the time sequence relation, the connection mechanism and the implicit association relation among the submodels in the transparent and non-transparent combined mode. In a transparent model combination mode, the connection and interoperation processes among unit models are known, the model dependency relationship and programs can be quantized according to the coupling relationship of sub models, the interoperation process and dynamic time sequence constraints, and the reliability of the combination model is deduced by adopting methods such as a Bayesian network and the like; under the model non-transparent combination mode, the connection and interaction between unit models are unknown, the credibility of the unit models needs to be evaluated well according to the integral input and output of the combination models, if machine learning methods such as a generation countermeasure network and the like are adopted, implicit association relations are quantitatively extracted according to simulation historical data, and the function mapping relations between the implicit association relations and the combined model credibility index values are mined, so that the combined model credibility index values C are calculated by integrating the connection mechanism and the implicit association relations between sub modelsimg。
Credibility index value C of integrated modelimThe calculation formula is as follows:
Cim＝(Cimd+Cimg)/2。
reliability index value C of support elementsThe calculation method comprises the following steps:
supporting elements according to uncertainty and maturity respectivelyCarrying out credibility evaluation, and integrating the credibility evaluation result to obtain a credibility index value C of the support elements。
Including uncertainty CuAnd maturity Cma(ii) a The digital twin system is a complex system, and uncertainty is inevitably introduced in the whole life cycle processes of construction, operation and the like, so that certain risk is brought to the operation of the system; therefore, a trusted digital twin system must be tough against the risk of uncertainty; the principle of the credible evaluation of the uncertainty analysis is to ensure that the construction process of the digital twin unit model carries out perfect uncertainty management and minimize the failure probability of the model caused by uncertainty risks. The smaller the probability of failure, meaning the smaller the uncertainty, CuThe smaller the value of (A), CuHas a value range of [0,1 ]](ii) a Starting from the process analysis aiming at the uncertainty analysis, the adequacy and the completeness degree of uncertainty in the construction and operation process of the digital twin system are evaluated and taken as the key reference of the operation toughness of the digital twin system.
For the maturity, when the same digital twin system is evaluated and perfected for multiple times under various application scenes and different requirements, the system maturity is gradually improved, so that the system has higher reliability when meeting the requirements of scenes which are not practiced in the same category; the credibility is a single numerical value evaluated aiming at a specific scene and a specific requirement under a certain time slice, and the maturity is the accumulated credible experience observed from the angle of the whole time axis of system evolution; the higher the maturity of the system, the higher its confidence in the applicable field, CmaThe higher the value, the range of [0,1 ]]。
Reliability index value C of support elementSComprises the following steps:
Cs＝(Cma-Cu+1)/2。
obtaining the data credibility index value CdAnd a reliability index value C of the basic unit modelbuAnd a reliability index value C of the evolution unit modelruAnd an integrated model reliability index value CimAnd a support element reliability index value CS。
The calculation method for aggregating a plurality of reliability index values and outputting a reliability calculation value comprises the following steps:
CT＝(4*(αT1Cd+αT2Cbu+αT3Cru+αT4Cim)+Cs/5)；
wherein, CTCalculating a value for the confidence of time T; alpha is alphaT1、αT2、αT3、αT4Respectively, data reliability index values CdAnd a reliability index value C of the basic unit modelbuAnd a reliability index value C of the evolution unit modelruAnd an integrated model reliability index value Cim. The corresponding weights, the sum of which is 1; the support element is an indirect influence on the reliability, and the weight thereof is set to 20%.
Calculated reliability value C at output time TTAnd then, the method further comprises the following steps:
judging whether the user requirement is changed, namely, the function expectation of the digital twin by the user in the same life cycle stage is changed, if so, re-evaluating from the user requirement analysis, at the moment, the digital twin is also possible to be reconfigured, reconstructed and the like, therefore, the evaluation of the basic model part is also performed again, returning to the step 2), and outputting the credibility calculation value C of the time T +1T+1。
If no change exists, whether the digital twin reaches the evolution condition or not needs to be judged; the evolution condition is composed of time drive and event drive; when the accumulated unexploited duration reaches a threshold value or certain characteristics of the model cannot be matched with new data, the digital twin model evolves, and the reliability is updated at the moment; in this case, since the basic model has not been adjusted, only the evolution unit model, the integration model, and the support element may be evaluated for a new round, and the process returns to step 3), and only the digital twin operating in the closed loop may be evaluated, and the reliability calculation value C at the time T +1 may be outputT+1。
Example 2
The embodiment of the invention discloses an equipment-oriented digital twin dynamic credibility calculation system, which comprises:
a determination module: for determining the life cycle stage characteristics of the model.
An analysis module: the method is used for analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes.
A calculation module: and the method is used for carrying out reliability calculation on the plurality of evaluation indexes to obtain a plurality of reliability index values.
A polymerization module: the reliability calculation value is used for aggregating a plurality of reliability index values and outputting the reliability calculation value.
A judging module: and the user demand judging module is used for judging whether the user demand changes or not after the credibility calculation value is output.
According to the method and the system, through analysis, data evaluation, basic unit model evaluation, evolution unit model evaluation, integrated model evaluation and support evaluation of user requirements, credibility related elements of the digital twins can be comprehensively and systematically investigated, the finally calculated credibility can fully reflect complexity, dynamics, interactivity and uncertainty of the digital twins, and higher reference value is provided for users. Meanwhile, dynamic value calculation is carried out through multilayer dynamic judgment and time binding of a data body and the reliability, factors which can change the digital twin reliability are intensively studied, including requirement change, time sequence drive, evolution event drive and the like, so that the reliability of the digital twin which emphasizes dynamic interaction also has effectiveness, and the actual physical significance of the reliability is enhanced.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. An equipment-oriented digital twin dynamic credibility calculation method is characterized by comprising the following steps of:
determining the life cycle stage characteristics of the model;
analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes;
carrying out reliability calculation on the plurality of evaluation indexes to obtain a plurality of reliability index values;
and aggregating a plurality of reliability index values and outputting a reliability calculation value.
2. The equipment-oriented digital twin dynamic credibility calculation method according to claim 1, wherein in step 3), the plurality of evaluation indexes are data, a basic unit model, an evolution unit model, an integration model and a support element;
the data and the basic unit model are used for evaluating a basic model and data which are not accessed to an actual system; the evolution unit model, the integration model and the support element are used for evaluating digital twins running in a closed loop and need to input a data sample at a time T;
obtaining the data credibility index value CdAnd a reliability index value C of the basic unit modelbuAnd a reliability index value C of the evolution unit modelruAnd an integrated model reliability index value CimAnd a support element reliability index value CS。
3. The equipment-oriented digital twin dynamic credibility calculation method according to claim 2, wherein the method is characterized in thatIn step 3), a confidence index value C of the datadThe calculation method comprises the following steps:
dividing the data into a data source and a data body, respectively carrying out credibility evaluation on the authority of the data source and the fidelity and the practical value of the data body, and synthesizing the credibility evaluation result to obtain a credibility index value C of the datadThe formula is as follows:
Cd＝Ad*(αd1Fd+αd2Vd)；
wherein C isdAs a reliability index value of the data, αd1、αd2The weight coefficient is determined according to the relative contribution of the corresponding index to the user demand, the sum of the two weights is 1, Ad、Fd、VdCredibility index values corresponding to authority, fidelity and use value indexes respectively, and the value range of the credibility index values is [0,1 ]]。
4. The equipment-oriented digital twin dynamic credibility calculation method according to claim 2, wherein in step 3), the credibility index value C of the basic unit model isbuThe calculating method comprises the following steps:
aiming at the subject field view, the basic unit model is divided into different types of sets C according to the subject field to which the basic unit model belongs, wherein the sets C are composed of part or all of the subject fields of acoustic Sc electric Ec gas Ac liquid Hc kinetic Mc thermal Tc and are represented as follows:
Figure FDA0003519467600000021
after the division is finished, aiming at the basic unit model of each specific category, a multi-index hierarchical evaluation method in the design field obtains a credibility index value C under a subject field viewbud；
Aiming at a model characteristic view, according to the difference of the characteristics of a digital twin unit model, dividing the digital twin unit model into a first model category, a second model category, a third model category and a third model category, and according to the three types of models, designing and developing different credible evaluation algorithms, wherein the formula is as follows:
Cbum＝αm1Bm+αm2Wm+αm3Gm；
wherein, CbumThe reliability index value under the model characteristic view is obtained; alpha is alpham1、αm2、αm3Respectively providing contribution weights of the first model, the second model and the third model to the use requirement under the user specified scene, wherein the sum of the three terms is 1; b ism、Wm、GmReliability index values of the first model, the second model and the third model are respectively;
reliability index value C of basic unit modelbuComprises the following steps:
Cbu＝(Cbud+Cbum)/2。
5. the equipment-oriented digital twin dynamic credibility calculation method according to claim 2, wherein in step 3), the credibility index C of the evolution unit modelruThe calculation method comprises the following steps:
analyzing a dynamic data assimilation core technology means, dividing the core technology means of evaluation of the evolution unit model into three aspects of 'fast', 'accurate' and 'stable', respectively evaluating the credibility of the evolution unit model, and synthesizing the credibility evaluation result to obtain a credibility index value C of the evolution unit modelruThe calculation formula is as follows:
Cru＝αr1Fr+αr2Ar+αr3Sr；
wherein, CruThe reliability index value of the evolution unit model is obtained; alpha is alphar1、αr2、αr3The weighted values corresponding to the indexes of fast, quasi and stable are respectively, and the sum of the weighted values is 1; fr、Ar、SrThe reliability index values corresponding to the rapidity index, the accuracy index and the stability index are respectively.
6. Root of herbaceous plantThe equipment-oriented digital twin dynamic credibility calculation method according to claim 2, wherein in step 3), the credibility index value C of the integrated model isimThe calculation method comprises the following steps:
aiming at an integrated model formed by fusing multidisciplinary models of the same entity, a disciplinary identification mechanism of the model is used, and a relational network construction method among the multidisciplinary models is used according to a complex knowledge structure among the multidisciplinary models; unified quantitative representation is carried out on the multidisciplinary model fusion relation by using a multidisciplinary model fusion effect evaluation method based on knowledge, and then a credibility index value C of the integrated model is solved according to a multidisciplinary fusion model credibility quantitative calculation method based on the credibility and the multidisciplinary fusion relation of the unit modelimd；
Aiming at a larger-granularity integrated model aggregated by a plurality of models, quantifying a model dependency relationship and a program according to a coupling relationship, an interoperation process and dynamic time sequence constraint among submodels in a model transparent combination mode, and deriving the reliability of the combined model by adopting a Bayesian network; under the model non-transparent combination mode, the reliability index value C of the combined model is calculated by integrating the connection mechanism and the implicit incidence relation among the sub-modelsimg；
Credibility index value C of integrated modelimThe calculation formula is as follows:
Cim＝(Cimd+Cimg)/2。
7. the method for calculating the digital twinning dynamics reliability as claimed in claim 2, wherein in step 3), the reliability index value C of the supporting element is calculatedsThe calculation method comprises the following steps:
carrying out credibility evaluation on the support elements according to the uncertainty and the maturity respectively, and synthesizing the credibility evaluation result to obtain a credibility index value C of the support elementss；
Uncertainty confidence index value CuProbability of model failure for uncertainty risk, CuHas a value range of [0,1 ]]The smaller the probability of failure, the smaller the uncertainty, CuThe smaller the value of (c);
investigating and evaluating a confidence index value C of a maturity accumulated from a confidence experience according to a whole time axis perspective evolved from the systemma，CmaHas a value range of [0,1 ]]The higher the maturity, CmaThe greater the value of (a);
reliability index value C of support elementSComprises the following steps:
Cs＝(Cma-Cu+1)/2。
8. the equipment-oriented digital twin dynamic credibility calculation method according to claim 2, wherein in step 4), the calculation method for aggregating the plurality of credibility index values and outputting the credibility calculation value is as follows:
CT＝(4*(αT1Cd+αT2Cbu+αT3Cru+αT4Cim)+Cs/5)；
wherein, CTCalculating a value for the confidence of time T; alpha is alphaT1、αT2、αT3、αT4Respectively are the data reliability index values CdThe reliability index value C of the basic unit modelbuThe reliability index value C of the evolution unit modelruAnd the reliability index value C of the integrated modelim. The corresponding weights, the sum of which is 1; the support element is an indirect influence on the reliability, and the weight thereof is set to 20%.
9. The equipment-oriented digital twin dynamic reliability calculation method according to claim 2, wherein in step 4), the reliability calculation value C at the time T is outputTAnd then, the method further comprises the following steps:
judging whether the user requirement is changed or not, if so, returning to the step 2), and outputting the reliability calculation value C of the time T +1T+1；
If no change exists, returning to the step 3), evaluating only the digital twin operated by the closed loop, and outputting a credibility calculation value at the time T +1CT+1。
10. An equipment-oriented digital twin dynamic credibility calculation system, comprising:
a determination module: the method comprises the steps of determining life cycle stage characteristics of a model;
an analysis module: the system is used for analyzing the user demand based on the life cycle stage characteristics to obtain a plurality of evaluation indexes;
a calculation module: the reliability calculation module is used for carrying out reliability calculation on the evaluation indexes to obtain a plurality of reliability index values;
a polymerization module: the reliability index value is used for aggregating a plurality of reliability index values and outputting a reliability calculation value;
a judgment module: and the user demand judging module is used for judging whether the user demand changes or not after the credibility calculation value is output.
CN202210177451.7A 2022-02-24 2022-02-24 Equipment-oriented digital twin dynamic credibility calculation method and system Pending CN114580162A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210177451.7A CN114580162A (en) 2022-02-24 2022-02-24 Equipment-oriented digital twin dynamic credibility calculation method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210177451.7A CN114580162A (en) 2022-02-24 2022-02-24 Equipment-oriented digital twin dynamic credibility calculation method and system

Publications (1)

Publication Number Publication Date
CN114580162A true CN114580162A (en) 2022-06-03

Family

ID=81775388

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210177451.7A Pending CN114580162A (en) 2022-02-24 2022-02-24 Equipment-oriented digital twin dynamic credibility calculation method and system

Country Status (1)

Country Link
CN (1) CN114580162A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115759509A (en) * 2022-11-11 2023-03-07 北京航空航天大学 Complex system-level digital twin operation virtual-real consistency determination and interaction method
CN116432323A (en) * 2023-06-14 2023-07-14 南京航空航天大学 Aircraft structure digital twin credibility assessment method based on Bayesian network

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115759509A (en) * 2022-11-11 2023-03-07 北京航空航天大学 Complex system-level digital twin operation virtual-real consistency determination and interaction method
CN115759509B (en) * 2022-11-11 2023-10-31 北京航空航天大学 Complex system level digital twin operation virtual-real consistency judging and interacting method
CN116432323A (en) * 2023-06-14 2023-07-14 南京航空航天大学 Aircraft structure digital twin credibility assessment method based on Bayesian network
CN116432323B (en) * 2023-06-14 2023-09-29 南京航空航天大学 Aircraft structure digital twin credibility assessment method based on Bayesian network

Similar Documents

Publication Publication Date Title
Bøegh A new standard for quality requirements
Rezaeian et al. Simulation of synthetic ground motions for specified earthquake and site characteristics
CN114580162A (en) Equipment-oriented digital twin dynamic credibility calculation method and system
Seo et al. On the value of outlier elimination on software effort estimation research
CN116150897A (en) Machine tool spindle performance evaluation method and system based on digital twin
Howick et al. Building confidence in models for multiple audiences: the modelling cascade
Hahn The conundrum of verification and validation of social science-based models
García et al. A survey on ontology metrics
He et al. Big data-oriented product infant failure intelligent root cause identification using associated tree and fuzzy DEA
Lesieur et al. Meta-modeling for urban noise mapping
Zhang et al. A mathematical model of performance-relevant feature interactions
Ernst et al. Arch-comp 2022 category report: Falsification with ubounded resources
Capizzi et al. Efficient control chart calibration by simulated stochastic approximation
Agiollo et al. Towards quality-of-service metrics for symbolic knowledge injection
Kumar et al. A Fuzzy Logic Approach to Measure Complexity of Generic AspectOriented Systems.
Kota et al. A method for estimating the degree of uncertainty with respect to life cycle assessment during design
Le et al. A generative network model for product evolution
CN109522639B (en) Portable concrete dam material dynamic parameter testing system and testing method
Zhang et al. Physics-informed deep learning for structural vibration identification and its application on a benchmark structure
Abrahao et al. Assessing the reproducibility and accuracy of functional size measurement methods through experimentation
Zhang et al. A combinational QoS-prediction approach based on RBF neural network
Leyffer A complementarity constraint formulation of convex multiobjective optimization problems
saagar Ponnusamy et al. A simulation fidelity assessment framework
Lin et al. Application of permutation entropy and factorization network in the classification of congenital heart disease
Hahn The conundrum of verification and validation of social science-based models redux

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