.png){kind=icon}
.png){kind=icon}
Return
A Grid Fault Type Identification Method Based on Novel Features of <italic>LCL</italic>-Type Inverter
.png){kind=icon}
PRE
AI
J
H
Y
D
Z
W
P
DOI：10.1109/jsen.2026.3706395.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The fault characteristics of the new power system are different from those of the traditional power system based on synchronous generators. While existing grid fault type identification methods have incorporated inverter dynamics and control interactions to varying degrees, they predominantly treat the inverter’s filter as a passive component whose response is either compensated or suppressed. The potential of utilizing the filter’s inherent transient resonance characteristics as a direct and autonomous source of fault information remains unexplored. Therefore, this article proposes a grid fault type identification method based on the resonant information of the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LCL</i> filter carried by the grid-connected inverter. Firstly, the resonance characteristics of voltages and currents within a 10 ms data window are extracted by three-point interpolation fast Fourier transform (TP-IpFFT). Secondly, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LCL</i> resonance characteristics of the grid-connected inverter under different fault types are analyzed. Finally, the multibranch 1-D convolutional neural network (MB-1D-CNN) method was used to train and test data under different fault conditions such as fault type, fault phase, transition resistance, and fault distance. The verification results show that the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LCL</i> filter of the grid-connected inverter can provide characteristic basis for grid fault type identification.
Keywords:
Deep learning
grid fault type identification
LCL resonance characteristics
three-point interpolation fast Fourier transform (TP-IpFFT)
Journal
IF:
4.5
Papers:
2.1W
Citations:
7.3W
Organization
Cited Papers