DOE-ANOVA Analysis to Estimate the Effect of Ambient Temperature, Pressure and Humidity on Surface Wind Speed

Abstract

The traditional assessment of wind resource considers the ambient factors like temperature, pressure and humidity as a secondary parameter, and focus on the statistically evaluation of wind magnitude and direction. Nevertheless, the evaluation requires a deeper analysis of the above ambient factors, because they are thermodynamically linked, and the energy flux among them will affect the wind dynamics. This research aims to analyse the interactions of temperature, pressure, and humidity variables with the wind speed with an inverse approach, it means that wind will be controlled by the ambient variables. We conducted a statistical analysis using data from a meteorological station located in the University of the Coast in Barranquilla, Colombia. The methodology is based on a Design of Experiments-Analysis of Variance (DOE-ANOVA) with a 3 2 factorial design. The estimated effects from the DOE-ANOVA results were utilized to generate standardized effect equations, for evaluating the response of wind speed during changes of the studied ambient variables (factors). The results evidenced that the applied methodology provided information about the non-linear interactions of the analysed variables, and the standardized effect equations were tested against a liner regression model with satisfactory results.