Wind load design considerations for the elevation and azimuth drives of a heliostat

Abstract

This paper investigated the dynamic fluctuations of the high-frequency surface pressure and force measurements on an instrumented scale-model heliostat within a turbulent ABL generated in a wind tunnel. Peak aerodynamic load coefficients on the model heliostat calculated following the equivalent static wind load method were consistent with previous wind tunnel studies in the literature. The dynamic analysis of the hinge, azimuth and overturning moments in the current study showed that there are a range of critical load cases and heliostat configurations that need to be considered to investigate the dynamic loading on the elevation and azimuth drives of a heliostat. Quasi-steady variation of the fluctuating peak loads following a Gaussian distribution was found to under-predict the maximum hinge and overturning moments in operating and stow configurations. It is therefore recommended that the analysis of instantaneous loads on the elevation drive and pedestal foundation is carried out for an improved estimation of the heliostat design wind loads.

Stowing strategy for a heliostat field based on wind speed and direction

“…By applying two criteria of wind speed and direction to the stowing strategy of parts of the heliostat field with similar azimuth angle rather than the whole field, the overall energy yield can be increased during synoptic high-wind periods with a steady wind direction. The operating wind loads on the heliostats that continued operating during high-wind periods were estimated using the wind load coefficients [8] corresponding to azimuth-elevation angle configuration, and comparing with the maximum stow loads to ensure the structural integrity of the supporting pylon, torque tube, drives and foundation.…”

“…Standard practice in the operation of heliostat fields is to stow the entire field during the approach of a high-wind event, such as gust front when the wind speed exceeds this design wind speed. The peak wind loads on heliostats in some operating positions exceed the maximum stow load, depending on the azimuth-elevation configuration of the heliostat and the turbulence intensity and length scales in the atmospheric boundary layer (ABL) simulated in wind tunnel experiments on scale-model heliostats at the University of Adelaide [3,[7][8][9][10][11][12]. It has be shown that the maximum wind loading on the drives and supporting structure during a synoptic gust front with a steady wind direction only applies to a portion of the heliostats in the field, such as azimuth angles of 0° (wind impacting front of heliostat) and 180° (wind impacting back of heliostat) for the hinge moment and overturning moment, and azimuth angles of ±60° and ±120° for the azimuth moment [8].…”

“…The peak wind loads on heliostats in some operating positions exceed the maximum stow load, depending on the azimuth-elevation configuration of the heliostat and the turbulence intensity and length scales in the atmospheric boundary layer (ABL) simulated in wind tunnel experiments on scale-model heliostats at the University of Adelaide [3,[7][8][9][10][11][12]. It has be shown that the maximum wind loading on the drives and supporting structure during a synoptic gust front with a steady wind direction only applies to a portion of the heliostats in the field, such as azimuth angles of 0° (wind impacting front of heliostat) and 180° (wind impacting back of heliostat) for the hinge moment and overturning moment, and azimuth angles of ±60° and ±120° for the azimuth moment [8]. In this study, a strategy that stows heliostats based on a design wind speed and wind direction is investigated through a sensitivity analysis of the azimuth angles of individual heliostats within a field and their corresponding aerodynamic load coefficients derived in wind tunnel experiments.…”

“…2. Non-dimensional wind load coefficients on a scale-model heliostat [3,7,8] at a range of elevation ( ) and azimuth ( ) angles in the University of Adelaide wind tunnel.…”

Extreme Value Analysis for Peak Heliostat Wind Load Predictions

“…Analysis of transient load distributions on a scale-model heliostat in the University of Adelaide wind tunnel using the three-sigma approach adopted in the equivalent static wind load design methods implemented by Peterka and Derickson [1] was found to underpredict the maximum hinge moment about the central elevation axis and overturning moment about the base of the pedestal [2]. Further, positive skewness of the von Mises combined stress distributions on the pedestal and torque tube calculated from the measured loads resulted in the three-sigma peak of the Gaussian distribution being underestimated by 6% and 2%, respectively, in comparison with a cumulative probability of 99.7% of the best-fit Gaussian distribution [3].…”