【摘要】 针对飞行器系统不确定性参数多、模型结构复杂等特点,提出一种基于多项式混沌展开的不确定性量化方法。由一种基于多维多项式混沌展开近似模型(Polynomial chaos expansion,PCE)作为代理模型对飞行器模型进行近似,来减少不确定性参数经过飞行器动态系统传播的计算量。在近似的代理模型基础上进行Sobol全局灵敏度指标计算。通过飞行器动态模型实例展示了多项式混沌展开方法在基于方差分解的全局灵敏度分析的有效性。更多还原

【Abstract】 In order to study uncertainty quantification of aircraft, which has the characteristics of many uncertainty parameters, complex model structures, and so on, an uncertainty quantification method based on polynomial chaos expansion is proposed. A method based on multi-dimensional polynomial chaos expansion(PCE) is used as an agent model to approximate aircraft so as to reduce the amount of calculation of transmitting by the vehicle dynamic system from uncertainty parameters. The Sobol index of the global sensitivity is calculated based on the approximate agent model. An example is presented to demonstrate the effectiveness of the polynomial chaos expansion method in global sensitivity analysis based on variance decomposition.更多还原