Bayesian Analysis of Regional Wind Energy Potential

The Bayesian method of statistical analysis has been applied to the problem of site assessment of the wind speed distribution. A Rayleigh distribution is used for wind speeds of interest for energy conversion. The mean wind speed is selected as the random parameter and a prior distribution selected to enable computation of the Bayesian distribution of wind speed reflecting both existing regional data and sampling observations. The convergence of the Bayesian distribution to the Rayleigh with increasing data is illustrated in terms of a sufficient statistic of sampling. Utility functions are computed for the power generated by a wind turbine. These are then combined with the Bayesian wind speed distribution to demonstrate two types of decision analysis: (1) Selection of the optimal site for a given wind turbine; and (2) selection of the optimal turbine for a given site.