Estimation Model of Electricity Generation with the Wind in the Colombian Caribbean Region Using Probability Distribution Functions

Abstract

The reduction in the availability of fossil fuels and global warming has led to alternatives in recent years for the generation of electric power. One of the biggest booms is the electric generation with wind turbines. This article presents an analysis of the characteristics of wind in two areas of Colombia in order to determine a statistical model and to verify the viability of implementing a system for obtaining energy through wind power. The two areas are located in the departments of Cesar (Alfonso Lopez) and Magdalena (Simon Bolivar), in the Colombian Caribbean area, which has the greatest potential for this type of project. It has been found out that the Alfonso Lopez station has a maximum operating period of 739 consecutive hours, with a probability of 10.9716% and it will operate continuously for a time greater than or equal to 9.3748 hours. The high asymmetry coefficient ensures a greater number of durations, higher than 9.3748 hours, while the Simon Bolivar station has a total number of useful series of 158918 with 28969 stops in the 2009-2013 period, with a probability of 4.7348% that it operates for a time greater than or equal to 5.4858 hours. It has been found out that the dispersion values are the lowest in the latter case, indicating that there is little variety in the duration of operation, from which it can be obtained that the Alfonso Lopez station has a greater wind potential than the Simon Bolivar one, so the construction of this wind farm is recommended in Magdalena zone. With the proposed model, an estimation of the potential can be made and control logics based on SPC can be proposed.