engleski

Application of a neural predictive controller in boost converter input current control

An application of a non-linear predictive controller in the cascade control structure of an boost converter is investigated. The neuro-predictive controller is realized as a non-linear optimizer, using Levenberg-Marquardt unconstrained optimization procedure. For the prediction of future process response two MLP neural networks are used. One network is used for modeling the converter dynamics and the other one for an on-line estimation of the converter?s input voltage. This structure ensures good performance in all operating regions and inherent compensation of ripples in converter?s input current, caused by variations of the input voltage. Advantages of the proposed control structure are demonstrated through the experimental comparison to a linear GPC with manually adjusted feed forward compensator.

boost converter; input current ripples compensation; MLP neural networks; predictive control.

nije evidentirano