当仿真因子数目较多,而真实系统观测数据相对稀少时,仿真模型的验证工作难以进行.针对仿真因子空间庞大,对应于每一组仿真因子配置,有且仅有对真实系统在相应工作条件下的一次观测这一极端情形,提出一种基于最优拉丁超立方设计(OMLHD,Or...当仿真因子数目较多,而真实系统观测数据相对稀少时,仿真模型的验证工作难以进行.针对仿真因子空间庞大,对应于每一组仿真因子配置,有且仅有对真实系统在相应工作条件下的一次观测这一极端情形,提出一种基于最优拉丁超立方设计(OMLHD,Orthogonal-Maximin Latin Hypercube Designs)的仿真模型验证方法.利用仿真验证试验的优化设计技术对仿真因子配置进行筛选,合理的选择、安排验证试验;采用p值分析方法解决验证试验中的数据一致性分析问题;基于逆概率分布定理的思想,对各组验证试验一致性分析结果进行综合分析.以某导弹制导仿真模型验证为例,对方法的性能进行了讨论.结果表明该方法能够实现在整个因子空间上对仿真模型的可信性进行验证的目的.展开更多
A new method based on cross-spectrum estimation for the verification and validation of computer simulation models is expounded in accordance with the characteristics of missile systems. The new method can expose the d...A new method based on cross-spectrum estimation for the verification and validation of computer simulation models is expounded in accordance with the characteristics of missile systems. The new method can expose the differences between two time processes in several aspects and can also give quantitative analysis results about the statistical consistence between them. An application to an actual anti-tank missile system simulation is presented and the calculated results confirm the effectiveness of the method. The approach can also be employed to verify the simulation models of other dynamic systems.展开更多
文摘当仿真因子数目较多,而真实系统观测数据相对稀少时,仿真模型的验证工作难以进行.针对仿真因子空间庞大,对应于每一组仿真因子配置,有且仅有对真实系统在相应工作条件下的一次观测这一极端情形,提出一种基于最优拉丁超立方设计(OMLHD,Orthogonal-Maximin Latin Hypercube Designs)的仿真模型验证方法.利用仿真验证试验的优化设计技术对仿真因子配置进行筛选,合理的选择、安排验证试验;采用p值分析方法解决验证试验中的数据一致性分析问题;基于逆概率分布定理的思想,对各组验证试验一致性分析结果进行综合分析.以某导弹制导仿真模型验证为例,对方法的性能进行了讨论.结果表明该方法能够实现在整个因子空间上对仿真模型的可信性进行验证的目的.
文摘A new method based on cross-spectrum estimation for the verification and validation of computer simulation models is expounded in accordance with the characteristics of missile systems. The new method can expose the differences between two time processes in several aspects and can also give quantitative analysis results about the statistical consistence between them. An application to an actual anti-tank missile system simulation is presented and the calculated results confirm the effectiveness of the method. The approach can also be employed to verify the simulation models of other dynamic systems.
