Analysis of Wind Characteristics using ARMA & Weibull Distribution

Abstract

A wind energy conversion system (WECS) converts the available wind potential at a particular wind site into electrical energy to meet the ever-increasing power demand. Assessing the adequacy of a WECS requires the consideration of some factors one of them is random nature of wind resources at the site location. So a model has to be developed in order to portray the intermittent characteristics of the wind. Here efforts have been made to present two popular methods to portray the behaviour of wind speed, one-time series autoregressive moving average (ARMA) model and other Weibull distribution. The fitting to the wind speed data using those methods follow their own different procedures, and the accuracies are checked indifferent approaches. In case of ARMA model, the accuracy is checked using F-criterion following Box-Jenkins guidelines. Weibull distribution fitting is checked through its parameter estimation using some statistical analysis.