Application of the Estimation-Before-Modeling (EBM) System Identification Method to the High Angle of Attack/Sideslip Flight of the T-2C jet Trainer Aircraft. Volume 2. Simulation Study Using T-2C Wind Tunnel Model Data

Abstract : This report presents the results of a feasibility study of the Estimation-Before-Modeling (EBM) method for aerodynamic parameter identification in the stall/post stall flight regimes. The feasibility study is conducted by processing synthetic flight data generated by exciting a wind tunnel model of the Navy's T-2C, a light jet trainer aircraft. It is carried out under realistic conditions by using the controls and the initial conditions of sixteen actual T- 2C flight test data records to generate the sixteen synthetic maneuvers used in the study. The synthetic data are corrupted by realistic measurement noise levels representative of the actual. The EBM method is a two stage process. In the first stage, the corrupted data are processed one maneuver at a time using an extended Kalman-Bucy filter/Bryson-Frazier smoother to obtain estimated values of the states and of the forces and moments acting on the aircraft. In the second stage, subspace modeling together with a step-wise multiple linear regression technique is used to process the estimated values from all sixteen maneuvers together in parallel to identify a global state/control dependent model of the three force coefficients and the three moment coefficients. The identified global model compares well with the true values of the T-2C wind tunnel model. The EBM method is demonstrated to accurately identify the high nonlinearities of the T-2C wind tunnel model. The prediction accuracy of the identified global model is shown to be excellent by a comparison of the responses for a new maneuver and for the original sixteen maneuvers.