Synthetic-Wind-Farm-Stream-No-Repair

# Synthetic Wind Turbine Condition Monitoring (No Repair) ## Context This dataset contains synthetic, multivariate time-series data representing the operational stream of a wind farm. In this 'No Repair' version, faults are allowed to persist without maintenance interventions. It is designed for predictive maintenance, anomaly detection, and fault classification. ## Target Variables * **`fault_labels`**: Multiclass target specifying the exact nature of the fault (e.g., yaw_misalignment, pitch_misalignment, temperature, healthy). * **`is_drifted`**: singleclass target specifying when dataset starts to drift. ## Attributes * **Environmental**: `wind_speed`, `air_density`, `wind_direction`, `ambient_temp`. * **Operational**: `active_power`, `rotor_speed`, `nacelle_direction`, `nacelle_position`, `pitch_blade_angle_1/2/3`. * **Fault Indicators**: `is_drifted`, `fault_severity`, `fault_temperature`, `fault_pitch_misalignment`, `fault_yaw_misalignment`.