Study of different parameters estimation methods of Weibull distribution to determine wind power density using ground based Doppler SODAR instrument

Abstract

The effectiveness of nine different numerical methods is examined for calculating the parameters of Weibull distribution at three different heights 80 m, 100 m and 120 m to estimate wind power density. The measurement campaign was conducted at Kayathar, Tamil Nadu, India. The time series wind data were recorded using SecondWind Triton SODAR (Sound Detection and Ranging) instrument. Firstly, the fidelity assessment of SODAR measurement was examined. The aim of this study is to identify the more accurate method for computing wind power density of a selected region. The performance of the selected methods is judged based on goodness of fit test i.e. Root Mean Square Error Test (RMSE), Coefficient of Determination (R2), Mean Absolute Percentage Error (MAPE), and Chi-square Test (X2). Wind power densities are estimated with the help of estimated parameter values. This study proposes an approach to utilize SODAR and also aims to examine the accuracy of SODAR measurement by comparing the results with cup anemometer, in an attempt to establish adequate criteria for an effective utilization of SODAR for wind resource assessment. The results suggest that SODAR may be used as an alternative to meteorological mast. Keywords: Weibull parameters, Wind frequency distribution, Wind speed, Probability distribution function, Statistical analysis