A Dual-Path Neural Network for High-Impedance Fault Detection
High-impedance fault detection poses significant challenges for distribution network maintenance and operation. We propose a dual-path neural network for high-impedance fault detection. To enhance feature extraction, we use a Gramian Angular Field algorithm to transform 1D zero-sequence voltage sign...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/2/225 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588048755326976 |
|---|---|
| author | Keqing Ning Lin Ye Wei Song Wei Guo Guanyuan Li Xiang Yin Mingze Zhang |
| author_facet | Keqing Ning Lin Ye Wei Song Wei Guo Guanyuan Li Xiang Yin Mingze Zhang |
| author_sort | Keqing Ning |
| collection | DOAJ |
| description | High-impedance fault detection poses significant challenges for distribution network maintenance and operation. We propose a dual-path neural network for high-impedance fault detection. To enhance feature extraction, we use a Gramian Angular Field algorithm to transform 1D zero-sequence voltage signals into 2D images. Our dual-branch network simultaneously processes both representations: the CNN extracts spatial features from the transformed images, while the GRU captures temporal features from the raw signals. To optimize model performance, we integrate the Crested Porcupine Optimizer (CPO) algorithm for the adaptive optimization of key network hyperparameters. The experimental results demonstrate that our method achieves a 99.70% recognition accuracy on a dataset comprising high-impedance faults, capacitor switching, and load connections. Furthermore, it maintains robust performance under various test conditions, including different noise levels and network topology changes. |
| format | Article |
| id | doaj-art-351ccecc3d4847e48f059e59987a8dec |
| institution | Kabale University |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-351ccecc3d4847e48f059e59987a8dec2025-01-24T13:39:48ZengMDPI AGMathematics2227-73902025-01-0113222510.3390/math13020225A Dual-Path Neural Network for High-Impedance Fault DetectionKeqing Ning0Lin Ye1Wei Song2Wei Guo3Guanyuan Li4Xiang Yin5Mingze Zhang6College of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaCollege of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaCollege of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Electrical and Control Engineering, North China University of Technology, Beijing 100144, ChinaBeijing Institute of Metrology, Beijing 100020, ChinaSchool of Electrical and Control Engineering, North China University of Technology, Beijing 100144, ChinaState Grid Jilin Electric Power Research Institute, Changchun 130015, ChinaHigh-impedance fault detection poses significant challenges for distribution network maintenance and operation. We propose a dual-path neural network for high-impedance fault detection. To enhance feature extraction, we use a Gramian Angular Field algorithm to transform 1D zero-sequence voltage signals into 2D images. Our dual-branch network simultaneously processes both representations: the CNN extracts spatial features from the transformed images, while the GRU captures temporal features from the raw signals. To optimize model performance, we integrate the Crested Porcupine Optimizer (CPO) algorithm for the adaptive optimization of key network hyperparameters. The experimental results demonstrate that our method achieves a 99.70% recognition accuracy on a dataset comprising high-impedance faults, capacitor switching, and load connections. Furthermore, it maintains robust performance under various test conditions, including different noise levels and network topology changes.https://www.mdpi.com/2227-7390/13/2/225high-impedance faultGramian angular fieldparallel networkCrested Porcupine Optimizer |
| spellingShingle | Keqing Ning Lin Ye Wei Song Wei Guo Guanyuan Li Xiang Yin Mingze Zhang A Dual-Path Neural Network for High-Impedance Fault Detection Mathematics high-impedance fault Gramian angular field parallel network Crested Porcupine Optimizer |
| title | A Dual-Path Neural Network for High-Impedance Fault Detection |
| title_full | A Dual-Path Neural Network for High-Impedance Fault Detection |
| title_fullStr | A Dual-Path Neural Network for High-Impedance Fault Detection |
| title_full_unstemmed | A Dual-Path Neural Network for High-Impedance Fault Detection |
| title_short | A Dual-Path Neural Network for High-Impedance Fault Detection |
| title_sort | dual path neural network for high impedance fault detection |
| topic | high-impedance fault Gramian angular field parallel network Crested Porcupine Optimizer |
| url | https://www.mdpi.com/2227-7390/13/2/225 |
| work_keys_str_mv | AT keqingning adualpathneuralnetworkforhighimpedancefaultdetection AT linye adualpathneuralnetworkforhighimpedancefaultdetection AT weisong adualpathneuralnetworkforhighimpedancefaultdetection AT weiguo adualpathneuralnetworkforhighimpedancefaultdetection AT guanyuanli adualpathneuralnetworkforhighimpedancefaultdetection AT xiangyin adualpathneuralnetworkforhighimpedancefaultdetection AT mingzezhang adualpathneuralnetworkforhighimpedancefaultdetection AT keqingning dualpathneuralnetworkforhighimpedancefaultdetection AT linye dualpathneuralnetworkforhighimpedancefaultdetection AT weisong dualpathneuralnetworkforhighimpedancefaultdetection AT weiguo dualpathneuralnetworkforhighimpedancefaultdetection AT guanyuanli dualpathneuralnetworkforhighimpedancefaultdetection AT xiangyin dualpathneuralnetworkforhighimpedancefaultdetection AT mingzezhang dualpathneuralnetworkforhighimpedancefaultdetection |