Synthesizing Wind Speed Data

Windographer's wind speed data synthesis algorithm generates artificial wind speed data from three pieces of information:

The mean diurnal wind speed profile (mean speed for each hour of the day)
The Weibull k value
The autocorrelation factor

In the Synthesize Wind Speed Data tool window, if you set the level of detail to Advanced, you enter the 24-hour mean profile directly. In Simple or Intermediate mode, you enter it indirectly by specifying the mean wind speed, the diurnal pattern strength, and the hour of peak wind speed, from which Windographer generates a cosine-shaped 24-hour mean profile.

To generate synthetic wind speed data from these three pieces of information, Windographer performs the following steps:

It generates an autoregressive time series with the 1-hr autocorrelation coefficient that you have specified. This time series has a normal distribution.
It performs a probability transformation on the diurnal profile to transform it to the same normal distribution as the autoregressive time series.
It adds the transformed diurnal profile to each day of the autoregressive time series.
It performs a probability transformation to make the resulting time series conform to a Weibull distribution with the desired mean and k value.

The resulting wind speed time series has the desired mean diurnal pattern, conforms to the desired Weibull k, and exhibits the desired degree of autocorrelation.