Open Access

A real-time monitoring and fault diagnosis method for underground mine electrical automation equipment combined with edge computing

Fuyong Guo

Fuyong Guo

Shandong Province Nuclear Industry 273 Geological BrigadeYantai, China
Search for this author on
Sciendo|Google Scholar
Guo, Fuyong
  
Mar 24, 2025
A real-time monitoring and fault diagnosis method for underground mine electrical automation equipment combined with edge computing's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 24, 2025
Received: Oct 22, 2024
Accepted: Feb 14, 2025
DOI: https://doi.org/10.2478/amns-2025-0776
Keywords
© 2025 Fuyong Guo, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Intelligent monitoring system architecture
Intelligent monitoring system architecture

Figure 2.

GRU neuron structure
GRU neuron structure

Figure 3.

Fault diagnosis framework
Fault diagnosis framework

Figure 4.

The effect of iteration number on model accuracy
The effect of iteration number on model accuracy

Figure 5.

Fault corresponding label
Fault corresponding label

Figure 6.

Textual Model prediction accuracy
Textual Model prediction accuracy

Node parameters of electrical automation equipment

Equipment circuit Test node Device configuration parameter
Reactive compensation/kvar Power flow overload coefficient Voltage limit coefficient Voltage amplitude/V
10-11 12 0.6343 0.5745 0.608 7.3198
12-14 13 0.7262 0.6288 0.6514 7.8069
25-50 34 0.8729 0.6306 0.7284 7.3801
15-20 17 0.8748 0.9253 0.9434 7.7455
21-46 42 1.1387 1.1481 1.0001 7.1032
34-59 48 1.1699 1.1561 1.1738 7.1342
25-46 29 1.3885 1.209 1.2639 7.4435
48-58 54 1.6808 1.3002 1.6448 7.1834
15-30 25 2.3264 1.6182 1.9221 7.1785
1-10 5 2.4568 1.6386 2.1993 7.0999
45-55 52 2.7133 1.9797 2.2305 7.2747
20-25 23 2.7522 1.9974 2.8971 7.3891
25-30 27 2.933 1.9992 2.8987 7.2243
28-38 35 2.9366 2.5323 2.9199 7.2859
1-15 7 2.9845 2.9079 2.9816 7.5853

Model hyperparameter Settings

Name of parameter Parameter values
Enter data pixel size 96*96
Convolution kernel size 5*5
Number of convolution kernel 256
Learning rate 0.002

Comparison of the number of diagnosis faults

Group Control group Experimental group Number of defect
Test point 1 5 41 60
Test point 2 32 37 60
Test point 3 11 46 60
Test point 4 2 43 60
Test point 5 45 52 60
Test point 6 18 35 60
Test point 7 23 53 60
Test point 8 26 49 60
Test point 9 37 48 60
Test point 10 31 57 60
Test point 11 3 36 60
Test point 12 19 47 60
Language:
English
Publication timeframe:
1 times per year
Journal Subjects:
Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other
Journal RSS Feed