A combined aerodynamic parameter identification method for missing test data

适用于部分试验数据缺失的气动参数辨识方法

School of Astronautics, Northwestern Polytechnical University, Xi'an 710072, China

Received: 10 June 2022

Abstract

In order to solve the problem that some test data cannot be measured or the measurement is difficult, a combined aerodynamic parameters identification method for missing test data is proposed. In this method, the aerodynamic parameter identification problem is modified into an optimization problem. The initial value of the flight state and the aerodynamic parameter interpolation table are used as design variables, and the motion equation of the aircraft including all aerodynamic parameters is used as a model to construct an objective function containing multiple pieces of test data information. In the optimization, the aerodynamic parameter database and the identification results of the existing methods are used as the prior knowledge. The initial value of the unmeasured data is fitted as the reference value. Then, the feasible sample selection method is designed. Finally, the differential evolution algorithm is used to solve the problem. The proposed method is used to process 264 pieces of test data, and the results show that compared with the existing aerodynamic parameter identification methods, the proposed identification method can obtain all aerodynamic parameters with higher accuracy and can inversely calculate and obtain unmeasured flight test data practical engineering significance.

摘要

为解决试验中部分数据缺失或难以测量的问题, 提出了一种适用于部分试验数据缺失的气动参数联合辨识方法。该方法将气动参数辨识问题转为优化问题, 以飞行状态初值和气动参数插值表为设计变量, 使用包含全部气动参数的弹道模型, 构建包含多条数据的目标函数。优化中以气动参数数据库和现有方法辨识结果为先验知识, 拟合出未测量数据的初值作为基准值, 设计了可行样本选取方法, 利用差分进化算法进行求解。应用所提方法处理264条试验数据, 结果表明相比于现有气动参数辨识方法, 所提方法能辨识全部气动参数, 准确度更高, 且能反算出未测量的飞行试验数据, 具有实际工程意义。