Statistical and spectral analysis of wind power: Fractional oscillation dynamics

Abstract

Time-dependent changes of the wind speed, as for example in Hera Campus (East Timor), are analysed by the statistical and the autocorrelation function in time domain and by the frequency spectrum (frequency domain) using the Fast Fourier Transform (FFT). The wind speed can be modelled using the Weibull distribution function. The autocorrelation function in time domain shows roughly a non-exponential decay with periodicity. The power spectrum shows two peaks and nearly 1/f a nature at high frequencies, close to the Kolmogorov prediction with α = 5/3. A Cole-Davidson type generalisation of wind dynamics, originating from the fractional dynamics of oscillation, is different from the dynamics of tides.

Keywords

• wind power
• statistical analysis
• Weibull probability distribution
• autocorrelation function
• Kolmogorov spectrum