Fault diagnosis of wind turbine gearbox based on kernel fuzzy c-means clustering

Conventional methods for fault diagnosis need a process of supervised training based on historical samples of known fault. However, it is time-consuming and costly to collect all kinds of known fault samples. In practice, it is lack of complete known samples for supervised training. These methods fail to diagnose a new or unknown fault. In this paper, a method based on kernel fuzzy c-means clustering (KFCM) was proposed to diagnose the known and unknown faults for fault diagnosis of wind turbine gearbox. At first, the samples of known samples were classified by KFCM and the class centre of each known fault was acquired. Similarity parameters in kernel space between new data samples and known class centres were employed in this paper. Thereafter similarity parameters were calculated for diagnosing whether the new data samples belong to knows faults. The proposed method was applied in fault diagnosis of wind turbine gearbox. The results show that the proposed method can diagnose both the known faults and unknown faults accurately and effectively.