Application of automation technology in power system protection and control
, und
29. Sept. 2025
Über diesen Artikel
Online veröffentlicht: 29. Sept. 2025
Eingereicht: 26. Jan. 2025
Akzeptiert: 10. Mai 2025
DOI: https://doi.org/10.2478/amns-2025-1131
Schlüsselwörter
© 2025 Lirong Xiao, Fugen Shu and Taiping Wu, published by Sciendo.
This work is licensed under the Creative Commons Attribution 4.0 International License.
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.
Figure 9.
Figure 10.
Figure 11.
Figure 12.
Training results under different parameters
| No. | First filter size | N | Second filter size | Nr | Number of iterations | Training set error rate | Verify the set error rate |
|---|---|---|---|---|---|---|---|
| 1 | 10×10 | 12 | 10×10 | 32 | 60 | 0.5381% | 0.5793% |
| 2 | 10×10 | 20 | 10×10 | 32 | 60 | 0.4877% | 0.5595% |
| 3 | 10×10 | 12 | 10×10 | 20 | 60 | 0.6762% | 0.8187% |
| 4 | 10×10 | 12 | 10×10 | 32 | 40 | 0.5391% | 0.6679% |
| 5 | 10×10 | 12 | 10×10 | 32 | 10 | 5.0672% | 9.4767% |
| 6 | 6×6 | 12 | 10×10 | 32 | 60 | 0.4993% | 0.5282% |
| 7 | 18×18 | 12 | 10×10 | 32 | 40 | 0.5988% | 0.6996% |
| 8 | 10×10 | 12 | 6×6 | 20 | 40 | 0.6796% | 0.7791% |
| 9 | 6×6 | 20 | 10×10 | 32 | 40 | 0.4789% | 0.4992% |
Power system voltage waveform data
| Type | Before optimization | Post-optimization | Difference value | Rate of change |
|---|---|---|---|---|
| Maximum value /KV | 36.17 | 34.39 | 1.78 | 4.92% |
| Minimum value /KV | -35.04 | -34.84 | 0.2 | 0.57% |
| Valid value /KV | 22.35 | 24.39 | 2.04 | 9.13% |
Test result
| Category | NF | 1PG | 1NG | 1PN | 2PG | 2NG | 2PN | 3PG | 3NG | 3PN | Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|---|
| NF | 6000 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100.00% |
| 1PG | 4 | 5924 | 21 | 4 | 15 | 8 | 10 | 7 | 1 | 6 | 98.73% |
| 1NG | 0 | 7 | 5946 | 6 | 7 | 10 | 3 | 10 | 9 | 2 | 99.10% |
| 1PN | 4 | 6 | 2 | 5967 | 1 | 1 | 5 | 6 | 2 | 6 | 99.45% |
| 2PG | 6 | 7 | 4 | 10 | 5951 | 1 | 3 | 3 | 9 | 6 | 99.18% |
| 2NG | 10 | 9 | 4 | 10 | 1 | 5957 | 0 | 5 | 1 | 3 | 99.28% |
| 2PN | 4 | 8 | 8 | 9 | 10 | 2 | 5937 | 10 | 6 | 6 | 98.95% |
| 3PG | 9 | 4 | 9 | 7 | 5 | 6 | 2 | 5951 | 6 | 1 | 99.18% |
| 3NG | 1 | 8 | 0 | 1 | 2 | 3 | 5 | 3 | 5976 | 1 | 99.60% |
| 3PN | 6 | 2 | 6 | 2 | 1 | 1 | 8 | 4 | 3 | 5967 | 99.45% |
| Total | 99.29% | ||||||||||
Output results of different fault types
| Fault type | Output result | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| NF | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 1PG | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 1NG | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 1PN | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
| 2PG | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 |
| 2NG | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |
| 2PN | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 |
| 3PG | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
| 3NG | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| 3PN | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
Various types of prediction accuracy, recall rate, F1 value
| Category | Accuracy | Recall | F1 |
|---|---|---|---|
| NF | 100.00% | 98.32% | 99.15% |
| 1PG | 98.73% | 96.68% | 97.70% |
| 1NG | 99.10% | 95.05% | 97.03% |
| 1PN | 99.45% | 98.09% | 98.77% |
| 2PG | 99.18% | 99.37% | 99.28% |
| 2NG | 99.28% | 97.02% | 98.14% |
| 2PN | 98.95% | 96.36% | 97.64% |
| 3PG | 99.18% | 98.54% | 98.86% |
| 3NG | 99.60% | 98.37% | 98.98% |
| 3PN | 99.45% | 97.97% | 98.70% |
| Total | 99.29% | 97.53% | 98.39% |