High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load
High Impedance Fault (HIF) detection in distribution networks is challenging for protection engineers, mainly because HIFs possess unique characteristics, including non-linearity, asymmetry, randomness, and relatively low fault current levels compared to the feeder load current. In this regard, the...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
middle technical university
2024-06-01
|
| Series: | Journal of Techniques |
| Subjects: | |
| Online Access: | https://journal.mtu.edu.iq/index.php/MTU/article/view/1320 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832595154104483840 |
|---|---|
| author | Mohammed Naisan Allawi Ali Nasser Hussain Mousa K. Wali Daniel Augusto Pereira |
| author_facet | Mohammed Naisan Allawi Ali Nasser Hussain Mousa K. Wali Daniel Augusto Pereira |
| author_sort | Mohammed Naisan Allawi |
| collection | DOAJ |
| description |
High Impedance Fault (HIF) detection in distribution networks is challenging for protection engineers, mainly because HIFs possess unique characteristics, including non-linearity, asymmetry, randomness, and relatively low fault current levels compared to the feeder load current. In this regard, the study proposes an approach to detect HIFs in a radial distribution feeder based on the spectrum analysis of current signals at the substation bus. The proposed method comprises two stages: signal decomposition and feature extraction. Fast Fourier Transform (FFT) is utilized for signal decomposition, followed by feature extraction. These features are subsequently used as input to an artificial neural network (ANN) to distinguish HIF from non-HIF events, such as linear and non-linear load switching, capacitor bank switching, and transformer energization. The proposed method's efficacy is rigorously evaluated under various dynamic conditions, demonstrating that the method can detect and differentiate HIFs from non-fault events with a high detection rate and high accuracy of 99.3%, irrespective of the HIF location and fault resistance.
|
| format | Article |
| id | doaj-art-e14946650da44f4e944ac62c72e419a4 |
| institution | Kabale University |
| issn | 1818-653X 2708-8383 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | middle technical university |
| record_format | Article |
| series | Journal of Techniques |
| spelling | doaj-art-e14946650da44f4e944ac62c72e419a42025-01-19T10:58:55Zengmiddle technical universityJournal of Techniques1818-653X2708-83832024-06-016210.51173/jt.v6i2.1320High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear LoadMohammed Naisan Allawi0Ali Nasser Hussain1Mousa K. Wali2Daniel Augusto Pereira3Electrical Engineering Technical College, Middle Technical University, Baghdad, Iraq.Electrical Engineering Technical College, Middle Technical University, Baghdad, Iraq.Electrical Engineering Technical College, Middle Technical University, Baghdad, Iraq.Universidade Federal de Lavras, Lavras, Brazil High Impedance Fault (HIF) detection in distribution networks is challenging for protection engineers, mainly because HIFs possess unique characteristics, including non-linearity, asymmetry, randomness, and relatively low fault current levels compared to the feeder load current. In this regard, the study proposes an approach to detect HIFs in a radial distribution feeder based on the spectrum analysis of current signals at the substation bus. The proposed method comprises two stages: signal decomposition and feature extraction. Fast Fourier Transform (FFT) is utilized for signal decomposition, followed by feature extraction. These features are subsequently used as input to an artificial neural network (ANN) to distinguish HIF from non-HIF events, such as linear and non-linear load switching, capacitor bank switching, and transformer energization. The proposed method's efficacy is rigorously evaluated under various dynamic conditions, demonstrating that the method can detect and differentiate HIFs from non-fault events with a high detection rate and high accuracy of 99.3%, irrespective of the HIF location and fault resistance. https://journal.mtu.edu.iq/index.php/MTU/article/view/1320Frequency SpectrumFast Fourier TransformArtificial Neural NetworkPSDDistribution FeederHigh Impedance Fault |
| spellingShingle | Mohammed Naisan Allawi Ali Nasser Hussain Mousa K. Wali Daniel Augusto Pereira High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load Journal of Techniques Frequency Spectrum Fast Fourier Transform Artificial Neural Network PSD Distribution Feeder High Impedance Fault |
| title | High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load |
| title_full | High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load |
| title_fullStr | High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load |
| title_full_unstemmed | High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load |
| title_short | High Impedance Fault Detection in Distribution Feeder Based on Spectrum Analysis and ANN with Non-Linear Load |
| title_sort | high impedance fault detection in distribution feeder based on spectrum analysis and ann with non linear load |
| topic | Frequency Spectrum Fast Fourier Transform Artificial Neural Network PSD Distribution Feeder High Impedance Fault |
| url | https://journal.mtu.edu.iq/index.php/MTU/article/view/1320 |
| work_keys_str_mv | AT mohammednaisanallawi highimpedancefaultdetectionindistributionfeederbasedonspectrumanalysisandannwithnonlinearload AT alinasserhussain highimpedancefaultdetectionindistributionfeederbasedonspectrumanalysisandannwithnonlinearload AT mousakwali highimpedancefaultdetectionindistributionfeederbasedonspectrumanalysisandannwithnonlinearload AT danielaugustopereira highimpedancefaultdetectionindistributionfeederbasedonspectrumanalysisandannwithnonlinearload |