2022
DOI: 10.1016/j.epsr.2022.108073
|View full text |Cite
|
Sign up to set email alerts
|

A hybrid approach for fault location in power distributed networks: Impedance-based and machine learning technique

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
5

Citation Types

0
20
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
8
1
1

Relationship

1
9

Authors

Journals

citations
Cited by 32 publications
(20 citation statements)
references
References 28 publications
0
20
0
Order By: Relevance
“…The traditional double-ended traveling wave method is only applicable to pure overhead lines or pure cable lines with the same wave speed, and cannot be used in mixed lines with different wave speeds. For this reason, the literature [1][2][3], proposed the single-end traveling wave method for fault ranging, This method is simple in principle and requires less measurement information and is economical, but it is not applicable in active distribution networks with complex situations; proposed the use of the time difference between the line mode component and the zero mode component for ranging, which can cause large errors because the zero mode component is severely attenuated during line transmission and the zero mode wave speed is not easily accessible; the literature [4][5] proposed a hybrid line fault location algorithm based on the time midpoint method, which determines the time midpoint location by detecting the time when the fault occurs at the fault detection point, and then selects the search direction for one-by-one projection, which is more complicated to operate; Narges Rezaee Raveshd et al [6] proposed a zone detection method based on wavelet analysis, which achieves the fault zone by detecting the timefrequency characteristics of the reflected pulse detection, but the accuracy of wavelet analysis is easily affected by wavelet bases. In addition, this method has different time-frequency analysis effects of signals under different wavelet decomposition scales, and the choice of wavelet decomposition scale is also an important factor affecting the time-frequency analysis of reflected waves.…”
Section: Introductionmentioning
confidence: 99%
“…The traditional double-ended traveling wave method is only applicable to pure overhead lines or pure cable lines with the same wave speed, and cannot be used in mixed lines with different wave speeds. For this reason, the literature [1][2][3], proposed the single-end traveling wave method for fault ranging, This method is simple in principle and requires less measurement information and is economical, but it is not applicable in active distribution networks with complex situations; proposed the use of the time difference between the line mode component and the zero mode component for ranging, which can cause large errors because the zero mode component is severely attenuated during line transmission and the zero mode wave speed is not easily accessible; the literature [4][5] proposed a hybrid line fault location algorithm based on the time midpoint method, which determines the time midpoint location by detecting the time when the fault occurs at the fault detection point, and then selects the search direction for one-by-one projection, which is more complicated to operate; Narges Rezaee Raveshd et al [6] proposed a zone detection method based on wavelet analysis, which achieves the fault zone by detecting the timefrequency characteristics of the reflected pulse detection, but the accuracy of wavelet analysis is easily affected by wavelet bases. In addition, this method has different time-frequency analysis effects of signals under different wavelet decomposition scales, and the choice of wavelet decomposition scale is also an important factor affecting the time-frequency analysis of reflected waves.…”
Section: Introductionmentioning
confidence: 99%
“…The development of AI has led to the emergence of data-driven and AI-based methods for fault location in active distribution networks [17,17]. These methods use historical fault knowledge models that consider multiple factors and integrate data collected from supervisory control and data acquisition (SCADA) systems for more accurate results.…”
Section: Introductionmentioning
confidence: 99%
“…This method presents a new impedance matrix operation procedure for studying distribution systems in zones that span multiple subsystems. Tavoosi et al combined the impedance method and deep neural network to propose a new method for fault location [7]. The required time for fault location with the new method is less than 6 seconds, with an accuracy rate of 99%.…”
Section: Introductionmentioning
confidence: 99%