Improvement Algorithm for Wind Farm Supervision Based On Proportional Distribution

Abstract

Nowadays, the research related to the wind farms is oriented to the development of improved supervision algorithm to manage the active and reactive powers as well as to provide an ancillary system. This paper proposes an enhancement PD (proportional distribution controller) algorithm for wind farm supervision. This algorithm combines a conventional PD algorithm with the prediction of the wind power generator, by using Artificial Neural Network (ANN). In fact the prediction power permits to determine the maximum active and reactive powers, which represents the PD regulator limits. Hence, the estimation of aerodynamic power, which represents major problems of the conventional PD algorithm, can be easily avoided. The performance of the proposed algorithm is verified through simulation results considering a wind farm of three generators (1.5 MW).