Transfer learning-based fault location with small datasets in VSC-HVDC

Shang, Boyang; Luo, Guangzhao; Li, Meng; Liu, Yinglin; Hei, Jiaxin

doi:10.1016/j.ijepes.2023.109131

Cited by 8 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fast-fault detection is applied, include on pole-and-line for both the single and double lines HVDC protection [11], and a commutation failure prevention model is designed by optimizing the value of resistance to minimize extinction angle in HVDC [12], a transient current is suppressed by supplementary control (SC) in rectifier HVDC [13], active disturbance rejection the SC is used to dampen oscillation in HVDC-line [14], and a novel SC is also applied to damp oscillation on maximum available power of extreme-weak AC-system [15]. Artificial intelligent (AI) applications in electrical and energy systems are introduced as follows: Machine learning is used to protect fault detection and classification scheme [16], and to search the fault location in HVDC systems [17]. Moreover, an artificial neural network (ANN) is used for fault detection and classification of the LCC-HVDC [18], to minimize transient stability for variation operating point [19], robust stability power on transmission [20] and direct current microgrid of shipyard system [21].…”

Section: Introductionmentioning

confidence: 99%

Transient response mitigation using type-2 fuzzy controller optimized by grey wolf optimizer in converter high voltage direct current

Ginarsa,

Nrartha,

Muljono

et al. 2024

IJECE

View full text Add to dashboard Cite

Long high voltage direct current (HVDC) transmission link is commonly used to transmit electrical energy via land or under-sea cable. The long HVDC avoids reactive power losses (RPL) and power stability problems (PSP). On the contrary, the RPL and PSP phenomena occur in long high voltage alternative current-link (HVAC) caused by the high reactive component in the HVAC-link. However, the HVDC produces a high and slow transient current response (TCR) on the high value of the up-ramp rate. Interval type-2 fuzzy (IT2F) control on converter-side HVDC is proposed to mitigate this TCR problem. The IT2F is optimized by grey wolf optimizer (GWO) to adjust input-output IT2F parameters optimally. The performance of IT2F-GWO is assessed by the minimum value of integral time squared error (ITSE), peak overshoot, and settling time of the TCR. The IT2FC-GWO performance is validated by the performance of IT2F control that is optimized by genetic algorithm (IT2F-GA) and proportional integral (PI) controller. Simulation results show that the IT2F-GWO performs better with small ITSE, low peak overshoot, and shorter settling times than competing controllers.

show abstract