Posters

Abstract

Lidars are now ubiquitous in wind resource assessment measurement campaigns. Accurately estimating the uncertainties of lidar measurements is critical for deriving the uncertainties of energy yield assessments and ultimately project financial risk. This presentation proposes the relevant uncertainties in lidar measurements and propagates these uncertainties through wind field reconstruction (WFR) algorithms used to estimate 1 Hz and 10-minute wind speeds. In flat terrain, these uncertainties include line-of-sight (LOS) uncertainty and pointing angle uncertainty. LOS uncertainty is a function of carrier-to-noise ratio (CNR) and are specific to heterodyne or homodyne lidar. Pointing angle uncertainty is a function of lidar manufacturing tolerances. These two uncertainties can be estimated using white box calibration techniques and distributions of verification campaign statistics. Following the Guide to the expression of uncertainty in measurement (GUM:1995), these uncertainties can be propagated through WFR algorithms to estimate 10-minute measurement uncertainties. In the case of scalar WFR, 1 Hz uncertainties are propagated via standard error of the mean to 10-minute uncertainties. In the case of vector WFR, the uncertainties are combined simultaneously with the 10-minute WFR averaging. In complex terrain, additional uncertainties exist in flow angle estimate uncertainty and speed inhomogeneity uncertainty, as well as uncertainties stemming from the elevation model data such as SRTM or ASTER, and the device's field installation. These uncertainties, associated with the computational fluid dynamics (CFD) simulations used to adjust lidar measurements, can be estimated using data collected during complex terrain measurement campaigns by collocated lidar and anemometry and the concurrent CFD modeling results. These uncertainties can be estimated and propagated following GUM:1995 or treated using Monte Carlo methods from Propagation of distributions using a Monte Carlo method (JCGM 101:2008). This framework can be used to estimate the uncertainties intrinsic to lidar, and as well to other remote sensing devices like sodar or radar. There is some overlap with the uncertainties derived from sensitivities to environmental parameters, but it is not 1-to-1. Derivations are presented for scalar, vector, and hybrid WFR for a pulsed, doppler beam swinging lidar in flat and proposals are presented for complex terrain. Comparisons with field data are presented from WindCube v2 lidar scalar WFR in flat terrain using manufacturing validation tests.