UPATRAS Floating Wind Turbine Vibration Dataset for Damage Diagnosis under Varying Wind Conditions

This repository contains vibration measurements acquired from a lab-scale Floating Wind Turbine (FWT) developed at the University of Patras, Stochastic Mechanical Systems and Automation (SMSA) Laboratory. The dataset is intended for research on vibration-based structural health monitoring, damage detection and damage diagnosis under varying operating conditions and related signal-processing, machine-learning and AI applications. The experimental campaign considers six structural states: one healthy state and five early-stage damage scenarios. The damage scenarios arise from three damage types: connection degradation between tower and floater, added mass simulating potential ice accumulation, and blade cracks. The FWT operates under nine operating conditions defined by the combination of three wind directions and three wind speeds. For each structural state and operating condition, ten repeated measurements are provided, resulting in a total of 540 vibration signals. All measurements are acquired using a single uniaxial accelerometer mounted on the upper part of the tower, with sampling frequency fs = 1024 Hz. Each signal contains N = 30 720 samples, corresponding to 30 s of vibration data and a frequency bandwidth of [0 − 512] Hz. Each measurement is provided in both CSV and MAT formats.

If you use this dataset in your work, please cite the following publication: https://doi.org/10.3390/s25041170

Further details on the data are available in the README.pdf file included in this dataset.