Parametric Data-Driven Reduced Order Models With State Consistence for Aeroelastic Analysis

Abstract

Abstract This paper presents a framework to develop data-driven parametric reduced order models (PROMs) for aeroelastic (AE) analysis of flexible vehicles within a broad flight envelop. It is based on the separate domain and mode shape perturbation method. The flight envelop is first partitioned by multiple grid points, on each of which an aerodynamic ROM (AeroROM) is constructed using system identification (SYSID) techniques to capture dependence of the generalized aerodynamic force on the generalized displacement using data generated by high-fidelity CFD simulation. Then AeroROMs not on the grid point are obtained by interpolating those at neighboring grid points. Two interpolation schemes, i.e., the output-based interpolation and coefficient interpolation are developed. The parametric AeroROM is then coupled with the mode-based structural ROMs to enable integrated AE analysis under various flight conditions. The state consistence enabled by different SYSID techniques and performance of both ROM interpolation methods are also investigated. For the first time, it is found that the autoregressive exogenous ROM allows state consistence and direct model coefficient interpolation. The ROMs exhibit excellent agreement with CFD simulations (< 3% relative error) and orders of magnitudes speedup. The effort opens up new opportunities for parametric AE analysis and flight control design.