Search

This article introduces a concept based on a step-up/down series partial power converter to perform droop control in dc grid-forming converters for buildings. Being responsible for processing all the power between the dc building and the grid, the high efficiency of this converter is essential. This study investigates the feasibility of combining two high-performance solutions—series-resonant dc-dc (DCX) converter, which features a constant voltage transfer ratio of 700/350 V and a partial power postregulator (P3R) that follows the droop control curves for prosumer dc buildings. In this application, the DCX is required to ensure isolation between ac and dc parts as required by standards for residential dc installations while also providing the necessary voltage matching. A 5-kW prototype was built and tested to evaluate the performance and effectiveness of the concept. The results show that P3R can provide the required output voltage regulation with an efficiency of around 99%, resulting only in minimal impact on the overall system efficiency. The realized dc grid-forming converter comprising DCX and P3R features an overall efficiency of 96.5% at rated power, with a very flat efficiency curve, presenting a promising solution instead of frequency, or duty-cycle controlled SRC, which results in lower efficiency at high power.

Power transformers are critical for grids, facilitating transmission, distribution, and voltage conversion. However, its insulation degrades over time, instigating catastrophic failures. This article investigates the emerging trends of artificial intelligence (AI) driven transformer prognostic health management (TPHM) strategies to mitigate such risks, emphasizing AI’s role in overcoming traditional methodology’s computational limitations. Hybrid approaches enhance dissolved gas analysis accuracy, while frequency response analysis is interpreted using machine learning. Deep-learning automates feature-weight assessment in partial-discharge and insulation analysis, eliminating manual health-index formulation. The implications and strategies for optimizing AI-driven TPHM efficiency are discussed, providing valuable trends in fostering a prognostic diagnosis ecosystem.

This article puts forward the analysis and design of single-phase bridgeless Cuk-derived dc–dc converter for a novel vehicle-to-vehicle (V2V) charge transfer technique. V2V charging adds flexibility to electric vehicle (EV) charging process and aids in mitigating range anxiety. The proposed V2V charge transfer technique has only one power conversion stage and only one onboard charger is used to charge the battery. The charger is capable of both grid-to-vehicle and V2V charging. The converter design is in discontinuous conduction mode which reduces the control complexity. The converter has a lower component count rendering the charger lightweight, power dense, and cost effective. The overall converter efficiency is high owing to low switching losses. This work has been compared to several state-of-the-art V2V topologies and configurations. The complete steady-state analysis and converter design operating in dc charging mode have been presented. Moreover, to validate the analysis, design, and applicability of the converter for dc V2V charge transfer, PSIM simulation, and laboratory-based hardware prototype have been built and the experimental results are provided.

Power conditioning systems are performing a key role in smart grid operation. The increase in renewable energy installation leads to an increase in the power electronic converters installation for distributed generation, rural electrification, and grid integration. As the numbers are increasing, the reliability of the grid is impacted mainly due to the power conditioning weak link. Thus, there is a need to improve the reliability and efficiency of overall power processing. One way to achieve this is to improve health by smart monitoring and make the controller fault-tolerant. For example, approaches that monitor the health of the devices along with fault-tolerant control architecture can be implemented in the controller to detect failures and to get timely alarm signals. For this, relevant data gathering and analysis are extremely critical. Machine learning (ML) and deep learning (DL) are approaches that analyze the data, learn from the data, and then apply it during the decision-making process. The Special Issue on Machine Learning Techniques in Power Electronics invites the articles related to data gathering/analysis and improvements of reliable operation of power conditioning and renewable energy resources with applications to the smart grid.