Open Access

High-Dimensional Feature Optimization and Real-Time Prediction Model with Support Vector Machines for Fault Diagnosis of Electrical Equipment

Lei Li

Lei Li

Shijiazhuang Insitute of Railway TechnologyShijiazhuang, China
Search for this author on
Sciendo|Google Scholar
Li, Lei
 and 
Yanling Gao

Yanling Gao

Hebei Vocational College of Rail TransportationShijiazhuang, China
Search for this author on
Sciendo|Google Scholar
Gao, Yanling
  
Mar 19, 2025
High-Dimensional Feature Optimization and Real-Time Prediction Model with Support Vector Machines for Fault Diagnosis of Electrical Equipment's Cover Image
About this article
Previous Article
Next Article
Cite
Download Cover
Published Online: Mar 19, 2025
Received: Oct 25, 2024
Accepted: Feb 03, 2025
DOI: https://doi.org/10.2478/amns-2025-0480
Keywords
© 2025 Lei Li et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Geometric diagram of principal component analysis
Geometric diagram of principal component analysis

Figure 2.

ISFD-POPS model
ISFD-POPS model

Figure 3.

MC-HBT classifier
MC-HBT classifier

Figure 4.

ISFD-POPS model flow
ISFD-POPS model flow

Figure 5.

The signal amplitude changes in the time of failure F1
The signal amplitude changes in the time of failure F1

Figure 6.

The signal amplitude changes when the F7 occurs
The signal amplitude changes when the F7 occurs

Figure 7.

The signal amplitude changes when the F8 occurs
The signal amplitude changes when the F8 occurs

Figure 8.

The signal amplitude changes when the F10 occurs
The signal amplitude changes when the F10 occurs

Figure 9.

The signal amplitude changes when the F14 occurs
The signal amplitude changes when the F14 occurs

Figure 10.

Structure of extreme learning machine
Structure of extreme learning machine

Figure 11.

Prediction model
Prediction model

Figure 12.

H2 prediction
H2 prediction

Figure 13.

The actual data of the training set is compared to the forecast data
The actual data of the training set is compared to the forecast data

After using the model presented in this article

Prediction fault type
Actual fault type f1 f7 f8 f10 f14
f1 523 0 27 0 0
f7 0 550 0 0 0
f8 63 0 205 5 277
f10 78 0 134 275 63
f14 65 0 278 32 175

The prediction of ammonia in different dimensions

Dimension m Parameter C Parameter g Parameter ε Parameter optimization error
2 6 220.3 0.000945 0.25
3 25.6 125.4 0.08 0.18
4 9.5 255.1 0.000945 0.344
5 72.14 46.8 0.145 0.0051
6 55.31 91.3 0.000945 0.00241
7 53.64 75.6 0.00088 0.00575

Transformer fault characteristic gas data(μL/L)

Time H2 C2H6 C2H2 C2H4 CH4
Supplementary data 22.1 110.1 0 67.6 32
2022/08/11 22.3 109.2 0 67.8 31.6
2022/09/01 21.6 107.8 0 66.7 31.7
2022/09/08 21.1 106.7 0 66.6 31.2
2022/10/11 19.9 106.5 0 64.5 30.6
Supplementary data 20.2 104.8 0 64.2 30.9
2022/12/25 18.2 103.9 0 62.6 30.5
Supplementary data 19.1 104.7 0 62.9 30.5
Supplementary data 18.8 103.7 0 63.4 30.4
2023/01/01 18.6 102.4 0 62.5 30.7
Supplementary data 18.7 102.7 0 62.5 31
2023/02/12 18.7 102.5 0 61.9 30.2
Supplementary data 19 103.9 0 62.5 31.6
2023/04/12 19.3 103.5 0 63 31.5
2023/05/01 19.5 103.8 0 63.2 31
Supplementary data 19.6 101.9 0 63.5 29.5
Supplementary data 19.9 102.5 0 63.2 31
2023/07/02 21.2 104.5 0 64.5 31.5
2023/08/11 21.2 105.1 0 64.2 31.5
2023/10/11 21.6 108.6 0 64.4 30
2024/01/01 21.9 110.6 0 65.3 32.6
2024/01/15 22 115.2 0 65.1 33.3

Test results

Name Time Actual value SVR/BP Predictive value RE% of SVR/BP MAPE% of SVR/BP
H2 1.39 20.51 20.937 20.902 4.12 5.33 4.04 5.14
2.01 20.59 20.96 20.72 -3.99 5.32
C2H6 0.67 111.8 111.503 111.812 0.66 -0.079 1.925 0.301
1.89 113.52 112.57 112.43 1.91 1.46
C2H4 1.48 64.41 64.12 64.95 1.145 -2.2 0.9975 1.735
2.31 64.89 64.285 64.89 -0.59 2.9
CH4 0.98 31.86 32.122 31.924 2.5 3.6 2.502 3.43
2.47 31.6 31.74 32.18 1.184 -3.19

Before using the model presented in this article

Prediction fault type
Actual fault type f1 f7 f8 f10 f14
f1 550 0 0 0 0
f7 0 550 0 0 0
f8 195 0 95 86 174
f10 256 0 81 138 75
f14 193 0 140 55 162
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