Black box systems identification with estimation

Abstract

In a black box system the internal parameters are not observables between input-output relationships. An approach is made knowing as estimation, based on the second probability moment. These results with a functional error and the innovation process are needed into identification structure. This process is known as an adaptive filter. Achieving a good convergence level with respect to the reference signal illustratively exemplified into simulation. The DC motor dynamical behavior is described by an adaptive filter without knowing the internal operation. Results compared with a methodology considered in.

Keywords

• digital filter
• estimator
• functional error
• identification
• stochastic gradient