Identification of vibration–speed curve for hydroelectric generator unit using statistical fuzzy vector chain code and support vector machine

The faults of hydroelectric generator unit have many different symptoms, and the relation between vibration amplitude and rotating speed is an important diagnosis criterion. In this article, a novel vibration–speed curve identification approach based on statistical fuzzy vector chain code and support vector machine is proposed and applied to identify the relation between vibration amplitude and rotating speed. In the identification process, the shape features of vibration–speed curve are extracted by statistical fuzzy vector chain code at first and then input to support vector machine to identify the type of vibration–speed curve; furthermore, the fault type of hydroelectric generator unit can be determined. Compared with the previous methods, statistical fuzzy vector chain code shows the advantages of low feature dimension, simple calculation, invariance to scaling and translation, and sensitive variance to rotation. The results of identification and comparative experiments show that the proposed method is more effective and efficient and can identify the vibration–speed curve with satisfactory accuracy.