Identifying Stochastic Wind Speed Model Parameters Using Autocovariance

Local wind speed estimation is important for various aircraft applications. Existing work has explored the use of different stochastic modeling techniques such as Random Walk, Gauss-Markov or Ornstein-Uhlenbeck for predicting wind speed. Autoregressive models have shown the capability for accurately predicting wind speed. This work explores the determination of stochastic parameters for a first-order autoregressive model using autocovariance. A data set containing wind speed data at 12 different locations was used to calculate the parameters of the first-order autoregressive model and then the model was validated with respect to real measurement data. Agreement between the simulated and real measurement data demonstrates the feasibility of determining autoregressive model parameters for wind speed using autocovariance.