Modelos equivalentes de parques eólicos usando algoritmos genéticos

Abstract

This work presents a genetic algorithm-based methodology that determines aggregated dynamic models of both squirrel cage induction generator (SCIG) and double fed induction generator (DFIG), presenting different electrical and mechanical parameters. The technique is based on a multi-objective optimal formulation solved by a genetic algorithm to minimize the quadratic error of the active power and reactive power between flue equivalent single-generator model and the investigated wind farm. The influence of flue wind farm equivalent model on flue dynamic behavior of synchronous generators in flue power system, are also investigated by using the proposed method. The approach is tested on a 10MW wind farm consisting of 4 wind turbines (2 x 2MW and 2 x 3MW) when both SCIG and DFIG are alternately integrated on the infinite bus and IEEE 14-bus power system. The results obtained using the detailed dynamic model for the wind farm representation are compared against those obtained with flue proposed aggregated model to evaluate the accuracy and the computational cost of the proposed model.