Voltage sag source location based on multi-layer perceptron and transfer learning

The existing voltage sag source localization only utilizes one simulation type or measurement data based on sampled waveform data. Moreover, it requires more storage space and transmission channels cannot combine the advantages of simulated data, resulting in poor applicability model. Hence, this paper proposes a locating method transfer learning from information event list combined with grid structure Firstly, location features are extracted degree impact, network structure, sag-type characteristics that characterize position source, they collectively used as inputs to Then, is build multi-classification model multi-layer perceptron line number classification number, employed fine-tune parameters achieve learning. Finally, achieved trained correctness proposed verified through actual specific area East China.