A Robust Adaptive Unscented Kalman Filter for Floating Doppler Wind-LiDAR Motion Correction

Abstract

This study presents a new method for correcting the six degrees of freedom motion-induced error in ZephIR 300 floating Doppler Wind-LiDAR-derived data, based on a Robust Adaptive Unscented Kalman Filter. The filter takes advantage of the known floating Doppler Wind-LiDAR (FDWL) dynamics, a velocity–azimuth display algorithm, and a wind model describing the LiDAR-retrieved wind vector without motion influence. The filter estimates the corrected wind vector by adapting itself to different atmospheric and motion scenarios, and by estimating the covariance matrices of related noise processes. The measured turbulence intensity by the FDWL (with and without correction) was compared against a reference fixed LiDAR over a 25-day period at “El Pont del Petroli”, Barcelona. After correction, the apparent motion-induced turbulence was greatly reduced, and the statistical indicators showed overall improvement. Thus, the Mean Difference improved from −1.70% (uncorrected) to 0.36% (corrected), the Root Mean Square Error (RMSE) improved from 2.01% to 0.86%, and coefficient of determination improved from 0.85 to 0.93.

On Adaptive Unscented Kalman Filtering for Floating Doppler Wind-Lidar Motion Correction: Effect of the Number of Lidar Measurement Heights

“…A study on the FDWL motion-correction performance when using the UKF method [5] and in relation to the number of lidar measurement heights (Zephir T M 300) was presented. It was shown that, at a given height, the effect of sequentially measuring at N different heights is equivalent to down-sampling the wind vector at that height by the same factor.…”

“…So far, different methodologies have been presented towards this purpose, which either require access to the lidar internal line-of-sight (LoS) measurements [2] or to carry out the compensation statistically at a post-processing level [6]. Recently, an on-the-run FDWL motion-correction method which does not require accessing the lidar internal LoS measurements has been presented by the authors [5]. The method is based on an adaptive Unscented Kalman Filter (UKF) that takes into account the FDWL dynamics as well as the lidar wind-retrieval algorithm to estimate the motion-corrected wind vector.…”

Assessing Obukhov Length and Friction Velocity from Floating Lidar Observations: A Data Screening and Sensitivity Computation Approach

“…Horizontal wind speed (HWS) and direction at ten different measurement heights (from 90 to 315 m) were measured by a ZephIR TM 300 DWL installed on a platform at 20.88 m in height (hereafter called the mast-DWL). This lidar measured 50 Line of Sights (LoSs) at equally spaced azimuth angles (7.2 • azimuth step between LoSs) at a sampling rate of 50 Hz along a conical scan with elevation of 30 • [39,40]. Two Vaisala TM HMP155D were installed at 21 and 90 m, which provided temperature and humidity observations every 0.25 s. Air pressure measurements were recorded at 21 m and 90 m by two Vaisala TM PTB210 every 0.25 s as well.…”

“…Regarding the FDWL, it is well-known that motion-induced effects on the retrieved wind vector become prominent at a 1-s temporal resolution but are negligible at the 10 min level. For the latter case, motion effects increase fluctuations on the wind speed [21,39,61]. At the 10 min level, it was shown [62,63] that the spatial variation (SV) parameter (see below) can be used as a threshold to filter data as it represents a trade-off between key performance indicator (KPI) improvements and data availability.…”

“…Therefore, the SV can be understood as a turbulence parameter describing the variation degree of the radial wind speeds within the scan [64]. High SV values are associated to measurements for which the hypothesis of uniform wind during the lidar scan is no longer valid [39]. This leads to outlier measurements, which must be removed.…”