Abstract
Recurrent neural network has been known for its dynamic mapping and better suited for nonlinear dynamical system. Nonlinear controller may be needed in cases where the actuators exhibit the nonlinear characteristics, or in cases when the structure to be controlled exhibits nonlinear behavior. The feedforward network with static characteristic usually uses a tapped delay input to control a nonlinear dynamic system. In the recurrent network, on the other hand, the dynamic behavior of the nonlinear system can be captured by the internal loop in its neurons and thus, a better system estimation and control can be expected using this control structure. In this paper, a multilayer perceptron diagonal recurrent neural network (DRNN) based control structure is employed to improve the performance of feedforward structure for Active Noise Control (ANC) systems where the nonlinearity occurs in the actuators. A comparison of DRNN with feedforward network is presented to highlight the improvement made by the recurrent structure.
