Mechanical condition monitoring of on-load tap-changers using chaos theory & fuzzy C-means algorithm

Mechanical failure is the main fault of transformer on-load tap-changers (OLTC), and part electrical failure, to a certain degree, is originally caused by mechanical malfunctions. In order to detect the hidden mechanical faults of OLTC timely and effectively, this paper presents the fuzzy C-Means (FCM) algorithm to recognize the distribution patterns of reconstructed vibration signals, which are closely related to the operation of OLTC and chaotic. First, Cao's method is applied to obtain the embedding dimension and delay time to reconstruct the phase space. Then the corresponding vibration signals in high dimension space are analyzed based on the proposed method. The results have indicated that the centroids distributions of each condition vary greatly, and the different lengths and angles of centroid vectors can be regarded as the judgment criterion quantitatively. Thus the method will provide the theoretical reference and practical guidance for the fault detection of OLTC.