Condition Assessment and Analysis of Bearing of Doubly Fed Wind Turbines Using Machine Learning Technique

Condition monitoring of wind turbines is progressively increasing to maintain the continuity clean energy supply power grids. This issue great importance since it prevents from failing and overheating, as most with doubly fed induction generators (DFIG) are overheated due faults in generator bearings. Bearing fault detection has become a main topic targeting optimum operation, unscheduled downtime, maintenance cost turbine generators. Wind equipped condition devices. However, effective reliable still faces significant difficulties. As majority health techniques primarily focused on single operating condition, they unable effectively determine turbines, which results unwanted downtimes. New strategies for data analysis were incorporated into this research, given large amount variety data. The development new model temperature DFIG bearing versus speed identify false alarms key innovation work. research aims analyze parameters bearings using SCADA k-means clustering training. variables k obtained by elbow method that revealed three classes (k = 0, 1, 2). Box plot visualization used quantify points. average rotation measurement found be primary indicators characterize normal or irregular conditions. In order evaluate performance model, an assessment indices also executed. ultimate goal study able use SCADA-recorded provide advance warning failures issues.