Wind driven generator fault mode identification method and system based on multi-feature fusion

The invention discloses a wind driven generator fault mode identification method and system based on multi-feature fusion. The method comprises the following steps: acquiring SCADA data during operation of a generator from a data acquisition system of a wind power plant; coding different fault data, making labels, and taking the labels as a model data set; according to the obtained fault data set with the labels, feature weight sorting is carried out on different faults through grey relational degree analysis, and according to the forward search principle, a plurality of features with the highest weight are selected as feature vectors corresponding to the faults; carrying out a fault diagnosis experiment by adopting a plurality of feature quantities, and extracting a simplest feature combination when the fault recognition rate reaches the highest under a specific fault; and finally, taking a union set from the simplest feature combination of different fault defects of the wind driven generator, fusing the fea