A data-driven modulation strategy for a triple-active-bridge (TAB) DC-DC converter is developed to directly minimize the total semiconductor losses, including both conduction and switching components, in wide-bandgap (WBG) SiC MOSFET-based systems. An offline exhaustive parameter sweep in PLECS, integrating detailed device-level SiC loss models and enforcing zero-voltage switching (ZVS) constraints, generates a high-fidelity dataset. The resulting dataset is further augmented and used to train a compact, artificial neural network (ANN) that predicts loss-optimal duty-cycles in real time, eliminating the need for large multidimensional lookup tables and computationally intensive online searches. Simulation results demonstrate consistent loss reductions of up to 23% compared to conventional phase-shift modulation across a wide operating envelope spanning 200-1000V and 2-10kW. The trained ANN requires only 870 bytes of memory and achieves mean absolute prediction error below 0.8%, enabling practical, high-efficiency control of multiport converters with minimal computational overhead and facilitating scalable implementation in embedded real-time power electronic controllers.

Interconnecting DC microgrids (MGs) into a cluster enhances renewable energy utilization and improves power supply reliability by enabling power flow between them. Effective management of DC MG clusters requires a control system designed for rapid response to fluctuating renewable source behavior and load demands. Traditional control methods lack the ability to pre-specify desired system performance, including convergence time and transient/steady-state behavior. Therefore, this work explores a prescribed performance function-based predefined time (PPF-PDT) control for optimizing the power dispatch of interconnected DC MGs according to the user-assigned preplanned desired performance. This scheme comprises secondary and tertiary control layers to handle the optimal operation for individual MGs and interconnected MGs, respectively, using a dual-layer sparse cyber network. In each MG, the secondary control matches the incremental costs of all distributed generation units while stabilizing the MG’s average voltage to the assigned voltage level within an adjustable predefined settling time independent of initial states. Adopting PPF significantly enhances transient and steady-state behavior, ensuring the tracking errors remain within desired performance limits. Additionally, distributed tertiary controllers across multiple MGs adjust their voltage references to optimize the exchanged power among them within a user-assigned tunable settling time. A thorough Lyapunov analysis verifies the stability of the proposed control algorithm within the predefined time and confines the tracking errors within acceptable bounds. Extensive simulation and experimental studies confirm the feasibility of the control strategy under various conditions.

This article introduces a novel prescribed time-based method for analyzing the convergence of evolutionary game dynamics in an information lossy network. Traditional game theory limits players to two choices, i.e., either cooperation or defection. However, player behavior in real-world scenarios is often multidimensional and complex; therefore, this work employs a continuous action iterated dilemma that allows players to choose a wider range of strategies. Moreover, traditional convergence analysis often relies on Jacobian matrices, which entail complex derivations. In contrast, the proposed strategy employs a time generator-based protocol that achieves agreement between all the players at a prescribed time, explicitly set by the user through a time parameter within the protocol. A comprehensive Lyapunov analysis affirms the prescribed time convergence even when the network is exposed to information loss during data transfer. Numerical simulations illustrate that the proposed scheme leads to a faster agreement at the preassigned time and with a better resilience performance compared to existing methods.

Smart grid is the nexus of advanced information and communication technologies and legacy power systems. With increasing awareness on the vulnerabilities of smart grids to cyberattacks, cybersecurity is becoming a prime concern. Earlier, it was assumed that the power system communication protocols are very specialized and different, so the “security by obscurity” approach would be sufficient. However, with the standardization of communication protocols for power utilities and the emergence of the power market, this approach is no longer valid. IEC 62351 Standard has been published to provide security recommendations for different power system communication protocols including IEC 61850. IEC 61850 is emerging as the most promising and popular power system communication standard. Therefore, in this article, a detailed analysis of security threats, possible attacks, and security requirements for IEC 61850 communication is presented. Building on this, the security considerations presented in IEC 62351 for securing different IEC 61850 messages, such as generic object-oriented substation events (GOOSE), sampled values (SV), routable-GOOSE, routable-SV, and manufacturing message specification messages have been presented in great detail.

Wide-scale deployment of Smart Inverters (SIs) can only happen if their impacts on power systems can be clearly understood. For this reason, thorough power flow and system stability studies are required. Traditional power system simulation software does not include proper models for SIs. Furthermore, dynamic behavior of SIs is not very well known to develop such models. Consequently, hardware in the loop tests with digital real-time simulators seem to be the best option, due to their high fidelity. That being said, interface between the simulated and real-world plays a very significant role. Since the real world is sampled and these samples are utilized to map reality inside digital real-time simulator, any mistake may render the test unstable. On the other hand, real-time simulation has very strong timing requirements and this becomes a deciding factor on how much detail can be modeled. Since the bridge inverters have several components that operate in time steps that are much smaller than conventional power systems, digital realtime simulators have very limited capacity. Using simplified inverter models has been investigated in the past and shown to be acceptable in steady-state situations. This paper investigates use of such models for protection studies in low-voltage networks.

Recent developments in renewable energy and information technology (IT) fields made it easier to set up power systems at a smaller scale. This proved to be a turning point for developing first-access electricity systems for the underserved locations around the world. However, there are planning and operation challenges due to lack of past data on such places. Deployment of Internet-of-Things (IoT) devices and proliferation of smart infrastructures with additional sensors will lead to tremendous opportunities for gathering very useful data. For different stakeholders to access and manage these data, trusted and standardized mechanisms need to be in place. Storing proper data in a well-structured common format allows for collaborative research across disciplines, large-scale analytics, and sharing of algorithms and methodologies, in addition to improved customer service. Data standardization plays a more vital role in the context of electricity access in the underdeveloped countries, where there is no past data on generation or consumption as in utility grids. Data collected in a standard structure, being it for a short period of time, facilitate learning from the past experiences, monitoring the current projects, and delivering better results in the future endeavors. It will result in ways to better assist consumers and help the industry operate more efficiently by sharing data with different stakeholders. It can also enhance competition, thus making electricity accessible faster and to more people. The focus of this article is data standardization for first-access electricity systems, in general, and renewable energy-based microgrids, in particular. Different data sources and ways that the corresponding data can be exploited, technological and capacity constraints for storage of data, political and governance implications, as well as data security and privacy issues, are examined. This article is relevant to different stakeholders, such as investors, public utilities, nongovernmental organizations (NGOs), and communities. Using the data standardization approach developed here, it is possible to create a much-needed first-access electricity system database. This will provide an important resource for project developers and energy companies to assess the potential of a certain unelectrified site, estimating its demand growth in time and establishing universal control systems that can seamlessly communicate with different components.