Microgrids, which are small electric power systems, accommodate different distributed generations and energy storage system. When faults occur outside or inside a microgrid, the static switch between bulk power system and microgrid must immediately open the circuit. This paper presents a fault detection, classification, and localization method that is based on a multiresolution analysis of the discrete wavelet transform (DWT) and a Taguchi-based artificial neural network (ANN). The differences of wavelet entropies of the three-phase fault voltages, three-phase fault currents, and the wavelet entropies of the neutral fault current are inputs to the ANN, which is trained using orthogonal datasets that are obtained from Taguchi's experiments. The ANN identifies the faulty phase and the location of the fault. The proposed DWT and ANN are implemented in a Renesas RX62T microcontroller, which is verified by chip-in-the-loop simulation with a real-time digital simulator (OP5600, Opal Lab) and a 50 kVA static switch to show the applicability of the proposed method.

The operations of power systems are becoming more challenging on account of the high penetration of renewable power generation, including photovoltaic systems. One method for improving the power system operation involves making accurate forecasts of day-ahead solar irradiation, enabling operators to minimize uncertainty in managing the balance between generation and load. To overcome the limitations of Long Short-term Memory (LSTM) in a one-dimensional forecasting problem, this work proposes a novel method in forecasting solar irradiation by encoding time-series data into images using the Gramian Angular Field and the Convolutional LSTM (ConvLSTM) network. The pre-processed data become a five-dimensional input tensor that is perfectly suitable for ConvLSTM. The ConvLSTM network uses convolution operations in its input-to-state transition and state-to-state transition. The network thus enables time-series forecasting by a feature-rich approach, which ultimately provides competitive forecasting performance despite the use of a small dataset. The proposed method was evaluated in day-ahead solar irradiation forecasting using a univariate dataset of Global Horizontal Irradiation (GHI) data from Fuhai in Taiwan. The proposed method was resampled using 5×2-fold cross-validation, and assessed using the Wilcoxon signed-rank test to determine the statistical significance of the result. It outperformed benchmark methods such as Autoregressive Integrated Moving Average (ARIMA), Convolutional Neural Network cascaded with Long Short-term Memory (CNN-LSTM), and LSTM cascaded with a fully-connected (FC) network.

Load flow studies are crucial in investigations of operation and planning problems in the power systems. Traditional methods for determining load flow are based on deterministic approaches. However, the parameters of a power system (such as load and renewable power generation) may be uncertain. An exact probabilistic load flow (PLF) study requires a long CPU time due to many convolution computations involving probability density functions. This paper proposes a novel PLF method that is based on Taguchi's orthogonal arrays. The proposed method utilises a few deterministic load flow solutions that are obtained using Taguchi's method to estimate the means and standard deviations of bus voltages, phase angles, line flows, and other metrics. A load flow calculation corresponds to an experiment in Taguchi's method. An optimal experiment is also specified by considering the largest deviation from the nominal load flow solution. A 25‐bus standalone power system and a modified Institute of Electrical and Electronics Engineers (IEEE) 118‐bus system are tested. The simulation results show that the proposed method not only requires fewer deterministic load flow solutions to perform PLF analysis than the traditional point‐estimate method but also yields accurate means and standard deviations of bus voltages and line flows.