Prospects of wind energy production in the western Fiji — An empirical study using machine learning forecasting algorithms

Electricity market in Fiji Islands are evolving. Accurate wind power forecasts are beneficial for wind plant operators, utility operators, and utility customers. An accurate forecast makes it possible for grid operators to schedule the economically efficient generation to meet the demand of electrical customers. This paper describes a feasibility study undertaken to forecast the potential of wind energy within the context of Rakiraki area which belongs to Western Division in Fiji by using forecasting algorithms. The daily wind speed data we consider from Fiji Meteorological Service within the time frame from 29th of August 2012 until the 30th of December 2016 and analyze to forecast wind speed to see the possibility of wind energy production in Fiji. Forecasting algorithms are tested with the dataset and it is clearly observed that Randomizable Filtered Classifier algorithm has forecasted exceptionally well. This study would encourage potential investors in giving them near to actual forecasted wind data for a feasibility study of their investment into wind energy farming to meet the demand of renewable energy production in Fiji.