Simulation Reliability Evaluation Method And System Development Of Complex Simulation System

In general,simulation credibility evaluation of complex systems is a time-consuming process,classic simulation model credibility evaluation methods have certain quantitative requirements for the simulation data and real data of the model,however,with the presentation and development of multi-resolution modeling(MRM)and simulation reuse models in complex systems,the classic simulation credibility evaluation method is sometimes difficult to complete the credibility evaluation of multi-resolution modeling and simulation reuse models.In the context of joint operations,the difficulty of obtaining low-resolution model data in MRM can be imagined,because of its characteristics,the simulation reuse model urgently needs to complete the overall credibility evaluation of the reuse model in a short time,then to complete the simulation model reuse.Therefore,it is very urgent to study the credibility evaluation method of low-resolution models in multi-resolution models and the overall credibility evaluation method of simulation reuse models to improve the credibility evaluation efficiency of complex systems.The main work of the thesis are as follows:(1)Proposing a method for credibility assessment of multi-resolution simulation models for aggregation-disaggregation.In the environment of joint operations,low-resolution models are very difficult to obtain real data,and unable to calculate model credibility using classic credibility evaluation algorithms,while obtaining real data from high-resolution models is relatively easy.Therefore,this paper studies the aggregation-disaggregation method in multi-resolution modeling,and proposes the accuracy propagation formula of multi-resolution simulation model under aggregation-disaggregation.Using small sample expansion technology based on bicubic interpolation to expand a small number of current combat simulation data samples,after comprehensively considering the two factors of model accuracy and consistency,a semi-parametric model is used to solve the credibility calculation method of the low-resolution model,and a multi-resolution simulation model credibility evaluation method for aggregation and disaggregation are proposed.(2)Proposing the credibility evaluation method for simulation reuse model based on bias propagation.Even if there are more complete sub-model simulation credibility evaluation data or evaluation results in the simulation reuse model,it still takes a long time and money to use the classic credibility evaluation algorithm to evaluate the credibility of the reuse model.On the basis of above,this paper proposes to use the simulation credibility of the sub-model to calculate the overall credibility of the simulation reuse model.According to the different application environments of the simulation reuse models,this paper divides simulation reuse models into physical-oriented simulation reuse models and numerical solution-oriented simulation reuse models,then introduces their respective modeling methods in detail.Using the theory of deviation propagation and the degree of data type matching between models,a credibility evaluation method for simulation reuse models based on deviation propagation under two types of simulation reuse models is proposed.(3)Based on the above methods,a credibility evaluation system for the simulation of complex systems is implemented.The functional modules,main classes and database design are introduced in detail.The credibility evaluation process of simulation model based on multi-resolution modeling and the credibility evaluation process of simulation reuse model are showed by two cases.