Development of Unified High-Fidelity Flight Dynamic Modeling Technique for Unmanned Compound Aircraft

Abstract

This study presents the unified high-fidelity flight dynamic modeling technique for compound aircraft. The existing flight dynamic modeling technique is absolutely depended on the experimental data measured by wind tunnel. It means that the existing flight dynamic model cannot be used for analyzing a new configuration aircraft. The flight dynamic modeling has to be implemented when a performance analysis has to be performed for new type aircraft. This technique is not effective for analyzing the performance of the new configuration aircraft because the shapes of compound aircraft are very various. The unified high-fidelity flight dynamic modeling technique is developed in this study to overcome the limitation of the existing modeling technique. First, the unified rotor and wing models are developed to calculate the aerodynamic forces generated by rotors and wings. The revolutions per minute (RPM) and pitch change with rotation direction are addressed by rotor models. The unified wing model calculates the induced velocity by using the vortex lattice method (VLM) and the Biot–Savart law. The aerodynamic forces and moments for wings and rotors are computed by strip theory in each model. Second, the performance analysis such as propeller performance and trim for compound aircraft is implemented to check the accuracy between the proposed modeling technique and the helicopter trim, linearization, and simulation (HETLAS) program which is validated. It is judged that this study raises the efficiency of aircraft performance analysis and the airworthiness evaluation.

Extended Hierarchical Kriging Method for Aerodynamic Model Generation Incorporating Multiple Low-Fidelity Datasets

“…The LHS method was employed to create sampling plans for the HF and LF analyses. The HF dataset encompassed 80 sample points produced through CFD analysis using ANSYS-Fluent 2020 R2 software, while the LF datasets were generated using the HETLAS [52], AVL [53], and XFLR5 [54] analysis tools, resulting in three LF datasets, each comprising 1300 sample points. It is worth noting that while HETLAS, AVL, and XFLR5 are considered low-fidelity analysis tools, they offer the advantage of producing a large volume of data points, spanning a wide spectrum of flight conditions at a relatively modest computational cost.…”