Fault Line Selection Method Based on Transfer Learning Depthwise Separable Convolutional Neural Network

Table 2

Automated algorithm simulation process.

Algorithm	Fault simulation automation based on MATLAB/Simulink
1	Call the Simpower System Toolbox to build a 10 kV power distribution system model
2	Configure fault parameters: different fault lines, phases, impedances, and locations
3	Start the simulation and apply a fault when running for 0.02 s
4	Stop the simulation after running for 0.2 s
5	Read 36 channel current simulation data (362560) and save it as a mat file
6	Repeat steps 2–5 to generate 6000 samples required for each faulty feeder
7	Repeat step 6 to get 6 types of fault sample data, for a total of 36000 samples