• M ESWAR MURALI
• Dr. R. Madhu

In this paper, a feedforward control scheme with disturbance prediction capability is proposed which improves its performance under accurate prediction, but it deteriorates due to the inevitable prediction error. A recently developed repetitive controller is being used in this paper. A time-base generator has been used as the input signal for neural approximator. The dynamic optimization problem (DOP) of shaping the control signal is solvable with the help of a function approximator. Here the feed-forward neural network (FFNN) is used as the function approximator. A constant-amplitude constant-frequency voltage-source inverter along with an output LC filter, is assumed to be equipped with a disturbance load current sensor. This enable the implementation of disturbance feed-forward path as part of the non-repetitive subsystem. The robustness to a measured noise is also tested alongside. The comparison is based on the results of numerical experiments.

Full state feedback control, Repetitive Control, Artificial neural network, Dual Feedforward Neural Network.

"DISTURBANCE PREDICTION FOR TIME VARYING SYSTEMS USING FEEDFORWARD NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.711-718, June 2019, Available :http://www.jetir.org/papers/JETIR1907J91.pdf

"DISTURBANCE PREDICTION FOR TIME VARYING SYSTEMS USING FEEDFORWARD NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp711-718, June 2019, Available at : http://www.jetir.org/papers/JETIR1907J91.pdf