基于多保真度代理模型的装备系统参数估计方法

Parameter estimation method of equipment system based on multi-fidelity surrogate model

系统参数估计是武器装备系统设计研制中的关键一环。针对复杂系统设计中逐渐突显的系统复杂性激增、需求动态性变化、响应的快速性、缺乏高精度数据等问题,提出一种基于多保真度代理模型的单装备系统参数估计方法。首先,该方法基于传统Kriging模型对低保真度数据建立代理模型。然后,将该模型作为全局趋势模型,插入高保真度数据建立多保真度代理模型,得到同时满足高精度和高效率双重要求的参数估计方法。最后,以飞机机翼系统设计中的最大升力系数参数估计问题为例,验证了所提方法在装备设计研制中的可行性和有效性,作为设计人员的决策支撑。

System parameter estimation is a key part in the design and development of weapon equipment systems.Aiming at the problems of the increasing system complexity,dynamic changes in demand,fast response,and lack of high-precision datas in the design of complex systems,a parameter estimation method for the equipment systems based on multi-fidelity surrogate model is proposed.Firstly,this method establishes a proxy model for low-fidelity data based on the traditional Kriging model.Then,this model is used as a global trend model for establishing a multi-tidelity proxy mode by inserting high-fidelity data,and a parameter estimation method that meets both the requirements of high accuracy and efficiency is obtained.Finally,the parameter estimation of the maximum lift coefficient in the design of aircraft wing systems is taken as an example to verify the feasibility and effectiveness of the proposed method in equipment design and development,which servased as a decision support for designers.