No abstract

No abstract

In this work, black-box model identification and output prediction for unknown sampled-d � ta rr: ini n: um phase system has been achieved. Feedforward neural network (multilayer perceptron) is used .for system l?entlficatlOn. Unscented Kalman Filter (UKF) online determine weights of neural netw ? rk � nd � redlcts outP � t III ? pen-l � op sampled-data configuration. Magnetic levitation and DC motor model has been IdentIfied m computer sImulatIOns usmg the presented black-box identification and prediction scheme.

No abstract