REAL-TIME PREDICTION OF UNSTEADY AERODYNAMICS - APPLICATION FOR AIRCRAFT CONTROL AND MANEUVERABILITY ENHANCEMENT

The capability to control unsteady separated how fields could dramatic ally enhance aircraft agility, To enable control, however, real-time p rediction of these flow fields over a broad parameter range must be re alized, The present work describes real-time predictions of three-dime nsional unsteady separated flow fields and aerodynamic coefficients us ing neural networks, Unsteady surface-pressure readings were obtained from an airfoil pitched at a constant rate through the static stall an gle, All data sets were comprised of 15 simultaneously acquired pressu re records and one pitch angle record, Five such records and the assoc iated pitch angle histories were used to train the neural network usin g a time-series algorithm, Post-training, the input to the network was the pitch angle (alpha), the angular velocity (d alpha/dt), and the i nitial 15 recorded surface pressures at time (t(0)), Subsequently, the time (t + Delta t) network predictions, for each of the surface press ures, were fed back as the input to the network throughout the pitch h istory, The results indicated that the neural network accurately predi cted the unsteady separated flow fields as well as the aerodynamic coe fficients to within 5% of the experimental data, Consistent results we re obtained both for the training set as well as for generalization to both other constant pitch rates and to sinusoidal pitch motions. The results clearly indicated that the neural-network model could predict the unsteady surface-pressure distributions and aerodynamic coefficien ts based solely on angle of attack information, The capability for rea l-time prediction of both unsteady separated flow fields and aerodynam ic coefficients across a wide range of parameters in turn provides a c ritical step towards the development of control systems targeted at ex ploiting unsteady aerodynamics for aircraft maneuverability enhancemen t.