Wind turbine fault diagnosis dataset (FAST– NREL 5MW)

This repository contains the measurement signals used during the development and validation of a wind turbine fault diagnosis methodology. The dataset includes six fault scenarios of a wind turbine benchmark. Each file fault_i.mat contains the time series of the available sensor measurements corresponding to fault condition i. A healthy operating condition is also provided. The MATLAB live script read_Sensor_data.mlx is included for visualization and inspection of the sensor behavior and fault effects. They are provided to allow readers to understand the characteristics of the signals used to design and tune the diagnostic method. The signals were generated using the wind turbine benchmark proposed by Odgaard, Stoustrup and Kinnaert (2013), based on the NREL 5-MW reference wind turbine implemented in FAST. The FAST simulation model is not distributed. Only processed measurement signals required for diagnosis are provided. Pérez-Pérez, E. J., Puig, V., Santos-Ruiz, I., Gúzman-Rabasa, J. A., & Valencia-Palomo, G. (2026). Wind turbine fault diagnosis using structural analysis and optimized-Rectangular GPR interval estimation. Control Engineering Practice, 172, 106940. https://doi.org/10.1016/j.conengprac.2026.106940本仓库包含用于开发与验证某风力发电机组故障诊断方法的测量信号。 该数据集涵盖某风力发电机组基准模型的六种故障工况。每个fault_i.mat文件均包含对应故障工况i的可用传感器测量时序数据，同时还提供了健康运行工况的相关数据。 附带了MATLAB实时脚本read_Sensor_data.mlx，用于可视化与检视传感器行为及故障影响，以便读者理解用于设计与调优诊断方法的信号特征。 本数据集所用信号基于Odgaard、Stoustrup与Kinnaert于2013年提出的风力发电机组基准模型生成，该模型依托于在FAST中实现的NREL 5-MW参考风力发电机组。 FAST仿真模型未随本仓库分发，仅提供诊断所需的已处理测量信号。