Adaptive Feed-Forward Neural Network for Wind Power Delivery

Abstract

This paper describes a grid connected wind energy conversion system. The interconnecting filter is a simple inductor with a series resistor to minimize three-phase current Total Harmonic Distortion (THD). Using the Recursive Least Squares (RLS) Estimator, an online grid impedance technique is proposed in the stationary reference frame using the Recursive Least Squares (RLS) Estimator. An Adaptive Feedforward Neural (AFN) Controller has also been developed using the inverse of the system to improve the performance of the current Proportional-Integral controller under dynamical conditions and provide better DC link voltage stability. The neural network weights are computed in real-time using the controller sample time, making the system highly compliant to abrupt changes in grid conditions. In the proposed technique, the varying inductance of the grid can be estimated and utilized in an adaptive feed-forward neural network for improving and smoothening of the power delivery to the grid from a wind energy system.