Full-condition Digital Twin Dataset for Wind Turbine in Real Scenarios

This dataset takes cooling fans operating under actual working conditions as the research object. Centering on the requirements of digital twin modeling and signal processing, it carries out dynamic, real-time and continuous physical signal acquisition. The experimental environment is characterized by severe aliasing of strong aerodynamic noise and bearing fault signals, with heavy noise interference, which is highly consistent with real industrial field scenarios. The dataset contains multi-type fault samples of SKF 6236 bearings: normal bearing (NM), outer race fault (OR), inner race fault (IR), rolling element fault (RF), and cage fault (CF). Different severity levels are set for each fault type. Some faulty bearings are equipped with nylon dust covers, while normal bearings retain the original metal dust covers. The rotational speed ranges from 150 rpm to 1500 rpm. Featuring real working scenarios, complex environments, strong noise, abundant fault types and coverage of different fault degrees, this dataset can truly simulate the operating status of bearings under actual working conditions, providing high-difficulty and high-authenticity data support for digital twin-oriented bearing fault diagnosis research.