Adaptive Inverse Control Synthesis Subject to Sinusoidal Disturbance for Non-Minimum Phase Plant via FVSS-NLMS Algorithm

In this paper, an Adaptive Indirect Inverse Control (IAIC) methodology based on the Finite Impulse Response (FIR) Filter is proposed, such that the controller is represented by an adaptive FIR Filter. The estimate of the weights vector of FIR Filter can be performed through an adaptive algorithm based on stochastic gradient descent, such that the performance of IAIC is influenced by the performance of update of the weights vector, in terms of convergence speed and steady-state Mean Square Error (MSE), that, consequently, is influenced by the step size of an adaptive algorithm. Aiming to present a proposal to solve this problem, a new version of NLMS algorithm is proposed, with the adapted step size through Mamdani Fuzzy Inference System (MFIS). The proposed algorithm was evaluated in the IAIC syhnthesis and applied in non-minimum phase plant, in the presence of a sinusoidal disturbance signal added to the control signal.