Particle Swarm Optimization method for estimation of Weibull parameters: a case study for the Brazilian northeast region

Abstract

In this paper the application of the Particle Swarm Optimization (PSO) method to estimate the Weibull parameters for wind resources in the Brazilian Northeast Region (BRNER) is reported. For the present research, wind speed data from three 80 m towers installed at different sites in the region were collected. The measuring periods for each tower site are: February 2012 to January 2013 for Maracanaú, August 2012 to July 2013 for Parnaíba, and May 2012 to March 2013 for Petrolina. Aiming to compare with the PSO performance, five numerical methods are applied to calculate the Weibull distribution parameters. Best performance for all analyzed sites is achieved by the PSO method, with a correlation higher than 99% and an error close to zero. PSO proves to be a valuable technique for characterization of the particular wind conditions found in the BRNER.