Microgrids experience significantly different fault currents in different operating scenarios, which make microgrid protection challenging. Existing intelligent protection schemes rely on the extraction of appropriate fault features using statistical parameters. The selection of these features is difficult in a microgrid because of its various operating scenarios. This study develops a convolutional neural network‐based intelligent fault protection strategy (CNNBIPS) for microgrids that inherently integrates the feature extraction and classification process. The proposed strategy is directly applicable to three‐phase (TP) current signals; thus, it does not require any separate feature extractor. In the proposed CNNBIPS, TP current signals sampled by the protective relays are used as an input to three different CNNs. The CNNs apply convolution and pooling operations to extract the features from the input signals. Then, fully connected layers of the CNNs employ the features to develop fault‐type, phase, and location information. To analyse the efficacy of the proposed design, we execute exhaustive simulations on a standard microgrid test system. The results confirm the effectiveness of the proposed strategy in terms of detection accuracy, security, and dependability. Moreover, comparisons with previous methods show that the proposed approach outperforms the existing microgrid protection schemes.

Islanding detection in power systems poses a challenge that can result in technical problems and safety hazards. This study examines a new passive islanding detection method for photovoltaic‐based distributed generation (DG) units. The modal signal of the three‐phase current and its envelope are calculated for a targeted DG unit. Variance in the autocorrelation of the modal current envelope (VAMCE) is used as an islanding detection criterion in the proposed method. The proposed method employs autocorrelation function (ACF) of modal current envelope (derived by Hilbert‐transform‐based approach) to extract the transient content. Then, the variance of the ACF is calculated, called criterion VAMCE to distinguish between islanded and normal operating conditions. The main advantage of the proposed technique is more relevant to its more accuracy and smaller non‐detection zone in comparison with the other similar passive techniques. Since the VAMCE has significant variation during islanding conditions and negligible change in normal conditions, it can be used for detection of islanding condition effectively. The method was evaluated through several simulations under different islanded and normal operating conditions. The results confirm that the proposed method has better accuracy and speed compared to other similar methods.

The increasing penetration of electric vehicles (EVs) in the distribution grid has established them as a prospective resource for ancillary services. These services require adequate control strategies for prompt and efficient operation. In this study, an energy management scheme (EMS) has been proposed to employ an off-board EV smart charger to support the grid during short-term variance of renewables and reactive load onset. The scheme operates by calculating power references for the charger instantaneously. The EMS incorporates a proportional power division methodology, proposed to allocate power references to the individual EVs connected to the charger DC-bus. This methodology considers the state-of-charge and battery sizes of the EVs, and it can aggregate energy from various types of EVs. The proposed scheme is compared with another power allocation method, and the entire EMS is tested under the scenarios of power mismatch and voltage sag/swell events. The results show that the proposed scheme achieves the goal of the aggregation of EVs at the charger level to support the grid. The EMS also fulfills the objectives of voltage regulation and four-quadrant operation of the smart charger.