Posters

Abstract

Floating TLP-based met masts is a relatively new concept for offshore wind resource assessment. Short, non-hub-height met masts together with a lidar is already an established measurement setup for onshore resource assessment. TLPs unavoidable motion is transferred to all wind sensors adding thus an additional uncertainty to measurements. This paper aims to assess the additional uncertainty, introduced to cups, ultrasonics and vanes, due to the motion of the TLP platform, by means of wind tunnel tests. Recently, FloatMast® deployed such a setup in the Aegean Sea (65m depth) and completed a 1-year measurement campaign. The on-board met mast (reaching 40m asl) was equipped with typical wind sensors (cups, vanes, ultrasonic), as well, as motion sensors. Data from this campaign were used to identify, for all wind sensors, typical motion patterns and tailor accordingly wind tunnel tests. TLPs are structures presenting low stiffness in sway, surge and yaw and high stiffness in heave, roll and pitch. As far as, wind measurements are considered, TLPs are characterized by: a) very low pitch and roll angles and b) horizontal motion in resonance with wave excitation. Considering the above, tests are performed for one dimensional reciprocating translation with primary characteristics: a) the motion velocity variation and b) the motion period. The motion amplitude is a dependent parameter and although limited by the wind tunnel capacity (up to 1m peak-to-peak), still addresses a large part of the target window. Experiments were performed at CRES wind tunnel, which is used for accredited anemometer calibrations, at four wind speeds (4 m/s, 8 m/s, 12 m/s and 16 m/s) for the cup and ultrasonic anemometers and at one (8 m/s) for the wind vane. All three sensors were prior calibrated according to MEASNET guidelines and then, the average deviations of the moving versus static test cases were examined. Considering deviations of mean values, it appears that the applied motion cases cause deviations that remain below each sensor calibration uncertainty. The induced turbulence intensity due to the motion, for all the test cases, is also examined. Higher turbulence intensities values are measured for the cup anemometer relatively to the ultrasonic for the same oscillations and this is attributed to the sensors different measuring methods. Based on the application range of these tests and considering a conservative approach for the uncertainty estimation, we propose a generalized uncertainty estimation for TLP structures as a function of wind speed. Concluding, wind tunnel tests can quantify the introduced uncertainty on anemometers, due to the typical motion patterns of a TLP met-mast.