Markov method for simulating non-Gaussian wind speed time series

This paper details a method which can be used to construct a wind simulator capable of generating wind time series with any distribution of hourly averages, exponentially decaying autocorrelation function, and a Gaussian realization of the turbulence. The method is based on a Markov random walk for hourly averages and an inverse hourly transform of the power spectrum to produce short-term turbulence. The Markov process is discussed in the first section, and the turbulence generator is covered in the second section. A description of the applications for which the model was developed follows.