NEURAL-NETWORK PREDICTION AND CONTROL OF 3-DIMENSIONAL UNSTEADY SEPARATED FLOWFIELDS

Using artificial neural networks (ANN), one approach to the control of unsteady aerodynamics is to develop real-time models which, given the actuator control signals, anticipate the unsteady flowfield wing inte ractions. These models of flow-wing interactions can then be used as t he foundation upon which to develop adaptive control systems. This art icle supports this concept using three-dimensional unsteady surface pr essure topologies collected from a rectangular wing pitched through th e static stall angle at seven nondimensional pitch rates. A neural net work model of the unsteady surface pressures was developed by training an ANN on five of these seven data sets. Following training, the only inputs required for the model were instantaneous angle of attack and angular velocity. These network-predicted unsteady surface pressure ti me histories were compared directly to the experimental pressure data. Then, a neural network controller for the wing motion history was dev eloped using the pressure model. The results indicated that the contro ller actuator signals reliably yielded motion histories that generated the measured lift to drag ratio (L/D) time histories. Further, the re sults suggest that for any desired L/D requirement optimized motion hi stories can be generated.