Field Measurements of Wind Loads on Heliostats in an Operational CSP Tower Plant

Introduction: Wind-induced loads can lead to heliostat surface deformations and tracking deviations that impact both the structural integrity and optical performance of the heliostat field in a concentrating solar power (CSP) tower plant. Dynamic wind loads can deform the heliostat's support structure and the reflecting mirrors, depending on a heliostat's location in the field, its orientation, and the characteristics of the approaching turbulent wind. Furthermore, the heliostat's structural design must be optimized for typical and extreme wind conditions to avoid costly overdesign and structural failure. Wind loads in a heliostat field are commonly estimated based on wind tunnel tests and numerical simulations, however localized wind conditions and resulting wind loads in the field can significantly deviate from idealized conditions. To understand wind loading in real-world conditions, NLR conducted field measurements of wind and resulting structural loads on heliostats in an operating CSP tower plant. We equipped three heliostats: two at the edge of the collector field, and one in the interior, with an array of load sensors for measuring forces and moments on the support structure, mirror deformations and dynamic accelerations. One main goal is to understand how static and dynamic wind loads, such as lift forces and hinge moments, vary between heliostats at the edge and in the interior of the heliostat field. Another focus is the effect of turbulent wind fluctuations on load fluctuations and wind-induced tracking deviations of heliostats in the field edge and interior over the operating range of frequencies and amplitudes. Methodology: Wind measurements instrumentation is installed on 3 masts adjacent to the 3 instrumented heliostats, with each mast (inflow mast on north-western field edge shown in Fig 1) containing 3 ultrasonic anemometers at heights of 2.75m, 5.5m and 11m measuring 3D wind speed and temperature, temperature and humidity sensors at 2.65m and 11m and a cup anemometer at 15m. Three individual heliostats (two at the field edge and one in the field interior) are instrumented with 8 bending and torsional strain gages, 3 differential pressure sensors (orange double triangles) between mirror facets, 4 accelerometers (blue circles) in the heliostat surface corners, 2 dynamic tilt inclinometers on the torque tube (purple circles) and 2 laser displacement sensors (red triangles) on the mirror support frame near the top and bottom edges of the heliostat surface, as shown in Fig 1. All wind and load measurements are sampled at 20 Hz for the measurement period from October 2024 to April 2025. Load measurements are processed for calculation of drag and lift forces, bending and torsional moments on the pedestal, torque tube and mirror support frame. The corresponding deviations in tracking angles and accelerations of the mirror surface and mirror support frame are measured using inclinometers, laser displacement sensors and tri-axial accelerometers. Localized wind flow approaching each heliostat is measured by adjacent 11m wind masts containing three 3D ultrasonic anemometers for analysis of wind profiles and atmospheric turbulence characteristics. Results: Fig 2 shows the time series of one-minute average inflow wind speed, wind direction, temperature and the lift forces and differential pressures measured on the three instrumented heliostats during a northerly wind period on October 28, 2024. While the field is in operation during the morning, there is significant variation in the lift forces on the pedestal from the field edge to the interior. The lift forces become particularly unsteady as the wind speed increases above 10 m/s, which triggered H3 to be stowed just before midday (vertical dashed line). The lift forces on H1 and H2 decrease significantly once they are stowed around 12:30 (vertical dashed-dotted line), however the lift force on H3 remains larger than H1 and H2 while they are all stowed. Time series data collected at 20 Hz of pressure differentials, torque tube torques (hinge moments), pedestal torques (azimuth moments) and localized support frame bending moments will be analyzed for H1 and H2 at the field edge and H3 in the field interior and their non-dimensional aerodynamic coefficients will be compared with wind tunnel experiments.