Application of extreme learning machine for series compensated transmission line protection

Kalman filter-based method for Online Sequential Extreme Learning Machine for regression problems

“…Thus, ELM has a much faster training process with better performance in some cases when compared with traditional neural networks (Huang et al, 2015). This has been a motivation for the application of ELM to many practical problems, for instance, image classification (Luo and Zhang, 2014), multistep-ahead time series prediction (Grigorievskiy et al, 2014) and fault section identification in transmission lines (Malathi et al, 2011).…”

Bearing fault diagnosis based on intrinsic time-scale decomposition and extreme learning machine

“…Compared to the traditional neural network, ELM improves the network generalization ability and the learning speed [5,6]. More specific theories of this method are given in Refs [7][8][9][10].…”

Fault location determination for transmission lines with different series-compensation levels using transient frequencies

“…MOVs are circuit elements that offer capacitors protection against high voltages [1,[3][4][5][6][7][8]. In the seriescompensation model, the line current is divided into two components: the capacitor and the MOV currents.…”

“…In this case, the capacitor needs protection and therefore control circuits are intended to prevent the capacitor from being damaged in this situation. Commonly proposed protection circuits include MOVs and air gaps with RL damping [5][6][7][8]. …”

“…Although series compensation is provided by capacitors, a protective circuit is required in the case of failure, when the voltage on the capacitor may exceed its normal value, damaging it. Metal oxide varistors (MOVs) and air gaps are used as typical protective circuits [5][6][7][8].…”