Hydroelectric generating sets fault diagnosis based on DDAGSVM

Hydroelectric generating sets (HGS) are main generating equipments in the electrical power system, and there is a strong demand on their reliable and safe operation. This paper aims to set up an efficient and high accuracy model of HGS fault diagnosis by support vector machine (SVM) method. So far as now, the constructing N-class SVMs is still an unsolved research problem. The paper presents a new constructing method (the Decision Directed Acyclic Graph Support Vector Machine, DDAGSVM) in HGS fault diagnosis at the first time in order to obtain better results. The DDAGSVM operates in a kernel-induced feature space and uses two-class maximal margin hyperplanes at each decision node of the DDAG. The last example results show that DDAGSVM is substantially faster to train and evaluate than either standard algorithms (Pairwise, Maxwin) in HGS fault diagnosis, while maintaining comparable accuracy to both of algorithms.