Table 7 Parameter configurations for LVQ-ANN components (ANN1–ANN5).
From: Condition monitoring and fault diagnosis of power transformer based on non-invasive measurement
Network | Task | No. of Neurons | Learning Rate | Training/Testing Split | No. of Layers | Activation Functions |
|---|---|---|---|---|---|---|
ANN1 | Identify SHF | 3 | 0.1 | 50%: 50% (132/135) cases | 1 | ReLU (hidden), Sigmoid (output) |
ANN2 | Classify SEF, TTF | 7 | 0.07 | 1 | ReLU (hidden), Sigmoid (output) | |
ANN3 | Locate SHF | 12 | 0.05 | 3 | ReLU (hidden), Linear (output) | |
ANN4 | Locate TTF | 4 | 0.08 | 1 | SoftMax (output) | |
ANN5 | Locate SEF | 8 | 0.06 | 2 | Tanh (hidden), SoftMax (output) |
