The penetration of PVs into the power grid is increasing day by day, as they are more economical and environment-friendly. However, due to the intrinsic intermittency in solar radiation and other meteorological factors, the generated power from PVs is uncertain and unstable. Therefore, accurate forecasting of power generation is considered one of the fundamental challenges in power system. In this paper, a deep-learning model based on two-stage attention mechanism over LSTM is proposed to forecast a day-ahead PV power. In addition, the Bayesian optimization algorithm is applied to obtain the optimal combination of hyper-parameters for the proposed deep-learning model. Various input features that can affect the PV power generation such as solar radiation, temperature, humidity, snowfall, albedo etc. are considered and their impact with respect to the attention mechanisms on the forecasted PV power is analyzed. The input consists of data from 21 PVs installed at different geographical locations in Germany. The proposed model is compared with state-of-the-art models such as LSTM-Attention, CNN-LSTM, and Ensemble model for day-ahead forecasting. The model is also compared with various single attention mechanisms such as Input-attention, SNAIL, Raffel, and Hierarchical attention etc. The proposed model outperforms the traditional methods in terms of accuracy, hence proving its efficiency. Forecasting Skill (FS) score of the proposed model is 0.4813 whereas 0.4427 is for the Ensemble model, which is the best among other state-of-the-art models. Root Mean Square (RMSE) and Mean Absolute Error (MAE) of the proposed model is 0.0638 and 0.0324 respectively, whereas those of the Ensemble model are 0.0685 and 0.0369 respectively.

Current transformer (CT) saturation is one of the significant problems for protection engineers. If CT saturation is not tackled properly, it can cause a disastrous effect on the stability of the power system, and may even create a complete blackout. To cope with CT saturation properly, an accurate detection or classification should be preceded. Recently, deep learning (DL) methods have brought a subversive revolution in the field of artificial intelligence (AI). This paper presents a new DL classification method based on unsupervised feature extraction and supervised fine-tuning strategy to classify the saturated and unsaturated regions in case of CT saturation. In other words, if protection system is subjected to a CT saturation, proposed method will correctly classify the different levels of saturation with a high accuracy. Traditional AI methods are mostly based on supervised learning and rely heavily on human crafted features. This paper contributes to an unsupervised feature extraction, using autoencoders and deep neural networks (DNNs) to extract features automatically without prior knowledge of optimal features. To validate the effectiveness of proposed method, a variety of simulation tests are conducted, and classification results are analyzed using standard classification metrics. Simulation results confirm that proposed method classifies the different levels of CT saturation with a remarkable accuracy and has unique feature extraction capabilities. Lastly, we provided a potential future research direction to conclude this paper.

Current transformer saturation has a negative effect on the operation of IEDs, resulting in their malfunction. Here, we present a technique to compensate for saturated waveforms using Bayesian Deep Neural Network (BDNN) comprising Deep Neural Network (DNN) and Bayesian optimization (BO). DNN, that utilizes stacked denoising autoencoder (SDAE) and Backpropagation (BP), is employed to optimize deep learning structure. Unlike the conventional neural network, which is a shallow network or random-initialize weights, the SDAE calculates optimal weights for each hidden layer and BP uses them to fine-tune which yields results with high performance for CT saturation compensation. To improve the empirical search of training hyperparameters, Bayesian optimization is adopted to decide training-related vectors such as batch size, learning rate, and number of neurons. Finally, the performance of the proposed approach was evaluated on an overhead transmission line which is imported from PSCAD/EMTDC with the different scenarios of fault inception angle, remnant flux, and voltage system. Therefore, numerical cases of saturation were comprehensively evaluated to demonstrate the performance of the proposed algorithm. A comparative analysis was shown to demonstrate that the proposed BDNN is superior to artificial neural network (ANN), and least square error (LES) technique.

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.

In this paper we propose a new method to evaluate the effects of nationwide conservation voltage reduction (CVR) on peak-load shaving, using substation operating results management system (SOMAS) data. Its evaluation is based on a national CVR factor, which is defined as the weighted average of CVR factors associated with all transformer banks and weighting coefficients are determined by the reconstructed loads corresponding to each transformer bank. To make use of the data resulting from nationwide CVR without installing additional measuring devices, we adopt a linearized static-load model with a linearizing parameter. SOMAS data are used to evaluate the effects of nationwide CVR on peak-load shaving in the Korean power system. Evaluation results show that the national CVR factor of the Korean power system has small values in the summer season and large values in the winter season. This means that the effect of nationwide CVR on peak-load shaving in the Korean power system presents stronger benefits during winter months.

This paper1 proposes a frequency estimation algorithm which effectively cancels the effect of unbalance in a three-phase power system. Precise frequency estimation under unbalance conditions is very difficult to achieve since a negative-sequence component distorts the spectrum of the phasor of three-phase voltage signals. In this paper, two parameters of the negative-sequence component are analytically compensated by introducing a modified periodogram, i.e., the phase parameter is removed, and the magnitude parameter is converted to a scaling factor. Rapid and precise frequency estimation is achieved by applying an iterative spectrum peak searching method to the modified periodogram. To verify the performance of the proposed algorithm, computer simulations were performed with various unbalanced and transient reflected signals. The results of the simulations confirm the accuracy and rapid convergence performance of the proposed algorithm.