The purpose of this paper is to remove the exponentially decaying DC offset in fault current waveforms using a deep neural network (DNN), even under harmonics and noise distortion. The DNN is implemented using the TensorFlow library based on Python. Autoencoders are utilized to determine the number of neurons in each hidden layer. Then, the number of hidden layers is experimentally decided by comparing the performance of DNNs with different numbers of hidden layers. Once the optimal DNN size has been determined, intensive training is performed using both the supervised and unsupervised training methodologies. Through various case studies, it was verified that the DNN is immune to harmonics, noise distortion, and variation of the time constant of the DC offset. In addition, it was found that the DNN can be applied to power systems with different voltage levels.

This paper proposes a technique using Energy Storage (ES) system to reduce power losses in distribution system where Photovoltaic (PV) is connected with grid, and mitigate voltage rise and voltage drop simultaneously. The sensitivity analysis of system loss is used to define the location of ES with the minimum energy losses. A modified IEEE 34-bus test system was simulated by Matlab/Simulink. Simulation results show that optimal siting and sizing of ES can help keeping voltages within the range and also minimizing energy losses.

To make a correct decision during normal and transient states, the signal processing for relay protection must be completed and designated the correct task within the shortest given duration. This paper proposes to solve a dc offset fault current phasor with harmonics and noise based on a Deep Neural Network (DNN) autoencoder stack. The size of the data window was reduced to less than one cycle to ensure that the correct offset is rapidly computed. The effects of different numbers of the data samples per cycle are discussed. The simulations revealed that the DNN autoencoder stack reduced the size of the data window to approximately 90% of a cycle waveform, and that DNN performance accuracy depended on the number of samples per cycle (32, 64, or 128) and the training dataset used. The fewer the samples per cycle of the training dataset, the more training was required. After training using an adequate dataset, the delay in the correct magnitude prediction was better than that of the partial sums (PSs) method without an additional filter. Similarly, the proposed DNN outperformed the DNN-based full decay cycle dc offset in the case of converging time. Taking advantage of the smaller DNN size and rapid converging time, the proposed DNN could be launched for real-time relay protection and centralized backup protection.

Current transformer saturation is a key issue for power systems because it negatively affects the operation of relays, resulting in the malfunction of protection devices. Recently, deep learning has been used in many academic fields with promising results. Here, we present a technique to compensate for saturated waveforms using deep learning to reconstruct an undistorted waveform. The optimal structure was obtained using pre- and fine-tuning mechanisms, which yielded good performance and initialized the optimum weight during the pre-training stage. The deep learning parameters were determined using particle swarm optimization prior to training. Finally, deep learning performance was evaluated using newly introduced conditions that were not observed during the training stage.

This paper presents a novel method for the removal of dual decaying DC offsets (DDCOs) to enhance discrete Fourier transform-based phasor estimation. The method proposed in this paper uses the sum of even samples and the sum of odd samples from the input signal to remove the AC components, thereby precisely estimating the primary and secondary DDCOs. The fluctuations induced by DCCOs present in the output of traditional DFT methods are eliminated by using the estimated DCCOs. The performance of the method was evaluated in terms of both mathematical-generated signals and fault current signals from a 154 kV Korean overhead transmission system. PSCAD/EMTDC was used to generate fault current signals at various fault distances and angles. The results show that the proposed method can estimate the phasor of the fundamental frequency component accurately regardless of the primary and secondary DCCOs. The paper concludes by describing the hardware implementation in a prototype unit that considers the real-world requirements of power system protection.

Inspec keywords: learning (artificial intelligence); discrete Fourier transforms; fault currents; phase estimation; neural nets; power system protection; power system faults; power engineering computing; power consumption Subjects: Integral transforms; Power engineering computing; Other topics in statistics; Other topics in statistics; Power system protection; Integral transforms; Neural computing techniques