A novel distributed unit transient protection algorithm using support vector machines

Integrated busbar protection scheme utilizing a numerical technique based on coherence method

“…However, there is always a need to develop innovative and efficient methods of busbar protection. References [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] present some modern techniques, which utilize sophisticated algorithms, to provide fast busbar protection and reliable performance during CT saturation. Some busbar protections based on Travelling Wave (TW), which used the transient fault information to avoid CT saturation effects and improve the speed and sensitivity of the protective relay [7][8][9].…”

Fast transient‐based detection of busbar faults employing improved morphological gradient

“…However, CT is powerful in transforming high‐frequency current signals up to 50 kHz [7]. In other words, protective algorithms based on transient components using current and voltage waves and/or voltage signals are impractical and only and just only, current signal with the sampling frequency not larger than 100 kHz is the best input for algorithms that are based on transient components. Criterion #5: Rule‐based approach : The attempts for transient‐based busbar protection schemes, only represented by simulation results, can be divided into pattern‐recognition based [8, 9, 15] and rule‐based approaches [3–5, 16]. Inherently, the former approach requires massive simulations for training; while, generally, it cannot guarantee 100% accuracy in detecting faults of different conditions and discriminating internal and external faults. In Table 1, according to the aforementioned requirements, the comparison between some busbar protection algorithms that are based on transient components as well as the proposed algorithm has been shown.…”

“…Criterion #5: Rule‐based approach : The attempts for transient‐based busbar protection schemes, only represented by simulation results, can be divided into pattern‐recognition based [8, 9, 15] and rule‐based approaches [3–5, 16]. Inherently, the former approach requires massive simulations for training; while, generally, it cannot guarantee 100% accuracy in detecting faults of different conditions and discriminating internal and external faults.…”

Initial travelling wavefront‐based bus zone protection scheme

“…Moreover, any change in busbar architecture requires retraining of the algorithms. In the past, machine learning-based schemes for busbar protection [15][16][17][18][19][20] have been presented. Although these schemes provide high classification accuracy for diversified fault scenarios, it would be difficult to achieve probabilistic predictions because of the abrupt behaviour of the kernel or cost functions.…”