Direct Adaptive Inverse Control via Fractional Least Mean Square

Abstract

This work aims to perform the performance analysis of the Fractional Least Mean Square (FLMS) algorithm in the Direct Adaptive Inverse Control (DAIC) design, in terms of convergence speed and steady-state Mean Square Error (MSE), for the controller weight vector. The controller, obtained through inverse identification of the plant model, is based on a Finite Impulse Response (FIR) adaptive filter. To obtain non-conservative results, the performance analysis was performed in the presence of a sinusoidal reference signal and sinusoidal disturbance signal. As an increment of complexity to the DAIC design, the plant model is non-minimum phase.