Table 3 Internal fault scenarios for training data.

From: Enhancing HVDC transmission line fault detection using disjoint bagging and bayesian optimization with artificial neural networks and scientometric insights

Transient period [1ms]

Collected samples

Fault resistance (Ω)

Fault distance (km)

Noise (db)

1st segment

189

0.01,25,50,…175, 200

1,10,20,…,180,190,198

20,25, 30

2nd segment

273

0.01,25,50,…,275, 300

1,10,20,…,180,190,198

20,25,30

3rd segment

357

0.01,25,50,…,375, 400

1,10,20,…,180,190,198

20,25,30

  1. Total faulty sample = 819 /each fault type, Healthy & External fault = 40 + 60 = 100/each segment type.
  2. Total training samples = [(819) × 3] + (100 × 3) = 2,757.
  3. Note that all these internal faults and external disturbances belong to DC link 12.
Back to article page