This paper examines the recent advances made in the field of Deep Learning (DL) methods for the automated identification of Partial Discharges (PD). PD activity is an indication of the state and operational conditions of electrical equipment systems. There are several techniques for on-line PD measurements, but the typical classification and recognition method is made off-line and involves an expert manually extracting appropriate features from raw data and then using these to diagnose PD type and severity. Many methods have been developed over the years, so that the appropriate features expertly extracted are used as input for Machine Learning (ML) algorithms. More recently, with the developments in computation and data storage, DL methods have been used for automated features extraction and classification. Several contributions have demonstrated that Deep Neural Networks (DNN) have better accuracy than the typical ML methods providing more efficient automated identification techniques. However, improvements could be made regarding the general applicability of the method, the data acquisition, and the optimal DNN structure.

This work proposes a methodology to automate the recognition of Partial Discharges (PD) sources in Electrical Distribution Networks using a Deep Neural Network (DNN) model called Convolutional Autoencoder (CAE), which is able to automatically extract features from data to classify different sources. The database used to train the model is constructed with real defects commonly found in MV switchgear in service, and it also includes noise and interference signals that are present in these installations. PD sources consist of defective mountings, such as the loss of sealing cap of cable terminations, or an earth cable in contact with cable termination insulation. Four sources were replicated in a Smart Grid Laboratory and on-line measurement techniques were used to obtain the PD signal data. The Continuous Wavelet Transform (CWT) was applied to post-process the PD signal into a time-frequency image representation. The trained model predicts with high accuracy new data, demonstrating the effectiveness of the methodology to automate the recognition of different partial discharges and to differentiate them from noise and other interference sources.

The expanding integration of wind and photovoltaic (PV) energy is disrupting the power system planning processes. Their incorporation poses limitations to forecasting due to their inherent variability. This review compiles a total of ninety studies conducted and published between 2019 and 2025, presenting for the first time an integrated approach that simultaneously optimizes the generation, transmission, storage, and flexibility of resources given high ratios of renewable generation. We present a systematic taxonomy of conflicting optimization approaches—deterministic, stochastic, robust, and AI-enhanced optimization—outlining meaningful mathematical formulations, real-world case studies, and the achieved trade balance between optimality, scale, and runtime. Emerging international cooperation clusters are identified through quantitative bibliometric analysis, and method selection in practice is illustrated using a table with concise snapshots of case study excerpts. Other issues analyzed include long-duration storage, centralized versus decentralized roadmap delineation, and regulatory and market drivers of grid expansion. Finally, we identified gaps in the literature—namely, resilience, sector coupling, and policy uncertainty—that warrant further investigation. This review provides critical insights for researchers and planners by systematically integrating methodological perspectives to tackle real-world, application-oriented problems related to generation and transmission expansion models amid significant uncertainty.

Vehicle-to-Grid chargers (V2Gs) are currently considered a key component for supporting future power systems in primary and especially secondary frequency regulation. Most of the research works conducted so far on this topic are based on a vehicle-centric perspective, whereas global approaches where the impact caused by fleets of V2Gs is studied from the perspective of an entire electric grid remain quite unexplored. In particular, the effects of aggregating different amounts of V2Gs from the HV, MV and LV subgrids to participate in frequency control represents a novel aspect to be investigated, which is expected to possess a strong potential for further improving the aggregation processes. To this end, this work presents a “global working framework” developed to evaluate the impact on an entire electric system of V2Gs participation in frequency regulation. A reality-based electric network is considered, where two HV, three MV and ten LV subgrids are interconnected. The effects of four different percentages of V2Gs aggregated from the HV, MV and LV subgrids are evaluated for two V2G installed capacities, 10 MW and 20 MW. Results are finally analyzed, including frequency vs time response, rate of change of frequency, grid losses, etc., and valuable knowledge, insights and guidelines are obtained for guiding future grid-level aggregation optimizations of V2Gs participation in frequency regulation.

The high penetration of renewable energies will affect the performance of present protection algorithms due to fault current injection from generators based on power electronics. This paper explains the process followed for analyzing this effect on distance protection and the development of a new algorithm that improves its performance in such a scenario. First of all, four commercial protection relays were tested before fault current contribution from photovoltaic system and full converter wind turbines using the hardware in the loop technique. The analysis of results obtained, jointly with a theoretical analysis based on commonly used protection strategy of superimposed quantities, lead to a conclusion about the cause of observed wrong behaviors of present protection algorithms under a high penetration of renewables. According to these conclusions, a new algorithm has been developed to improve the detection of faulted phase selection and directionality on distance protection under a short circuit current fed by renewable energy sources.

Renewable power is expected to increase drastically in the coming years due to the energy transition. A large part of the newly installed generators will be connected to the power system through inverters and electronic converters, whose behaviour differs from the generators connected synchronously to the network. One of the main differences is the current contribution during symmetrical and asymmetrical faults which can affect protection systems. New grid codes establish requirements for fast current injection, but the converter’s maximum current limitations during faults make it difficult to establish control strategies for such current contribution. This paper studies the performance of faulted phase selector algorithm of a commercial relay under the current contribution from renewables before unbalanced faults. Two positive and negative sequence current injection strategies in compliance with new Spanish grid code requirements are proposed and tested under fault conditions in HiL (Hardware in the Loop). Test results show that the selected injection strategy affects the fault phase identification algorithm. Furthermore, the negative sequence injection requirements established in the new grid code improve the relay performance when line-to-line faults are applied, but they are not enough to identify all fault types.