A statistical-based examination on wind turbines’ bi-static scatterings at 1 GHz frequency

Abstract

This paper aims to introduce a statistical modeling approach for wind turbines’ (WTs) bi-static scattering characteristics in free-space environments. By using the analytical signal model proposed in the literature for the complex bi-static scattering of WTs, the statistical features of the long-term magnitude and phase variation for different bi-static scenarios (i.e. several discrete receiver locations for a fixed-point transmitter) are characterized with the help of the distribution-fitting process for several probability densities via maximum-likelihood estimation. The results of distribution-fitting process are interpreted with the help of goodness-of-fit considerations such as minimum square error, Kolmogorov–Smirnov Test, Kuiper’s V Test, and Watson’s Test, in order to qualify the suitability of each considered probability density. Moreover, by carrying out the examination by including a polynomial series approximation approach on the estimated parameters of two best-matching densities (i.e. -stable and Student’s t), the outputs are rendered as to be valid for any receiver locations. In addition to the insightful outcomes and conclusions for the considered electronic system scenario including the operating frequency (i.e. 1 GHz) and orientations, the investigation presented in this paper also introduces a simple, fast-responding methodology of data pool generation, distribution-fitting and performance assessment for the WTs’ bi-static scattering effects.

Keywords:

• Bi-static scattering
• goodness-of-fit
• wind turbines

Notes

No potential conflict of interest was reported by the authors.

1 In order to perform the numerical evaluation of scattering terms including infinite series as in Equations (5(5) )–(7(7) ) within affordable margins of compromised computation accuracy and computation time costs, the terms with the indices satisfying have been considered.