Dynamic assessment and prediction of equipment status in 220kV substations by big data and AI technology

This file contains the 177 data samples which have been collected from literatures, it has been divided into 150 data  sample for training and 27 samples for testing. The 150 data samples have been divided into 15 groups, in which every groups contain two types of power transformer faults, without repetition by using the combination of subset of set method then every group is inputted the FCM algorithm for clustering. The last step it contains the final clustering data  set which contains 30 samples. Every fault will have 5 samples.