Sistemi Elettronici di Gestione della Potenza per Applicazioni di Sorgenti Rinnovabili e di Accumulo Energetico

In the current race toward electrification, driven by the increasing adoption of sustainable energy sources, energy storage technologies, micro-grids and electric vehicles (EVs), power electronics plays a pivotal role. Efficient propagation of electrification, particularly in power electronic converters, necessitates the development of new optimization techniques and enhancements to existing technologies. This dissertation addresses these challenges by introducing novel modeling techniques, control strategies for existing converters, and improvements in the operation of dc-ac inverters. The first part of this dissertation focuses on the development of a novel modeling approach called energy-based modeling, which provides a more comprehensive description of dc-dc converter operations. This approach incorporates not only the canonical dynamics of duty cycles and voltages but also the dynamics of carrier displacement. Initially, the proposed modeling technique is applied to the four-switch buck-boost converter. Subsequently, the methodology is extended to buck and boost converters, where phase-shift effects have traditionally been overlooked. By applying the developed energy-based modeling to these converters, the study explores various control applications, leading to new regulation schemes and concepts that enhance converter stability and operational performance. The proposed modeling and control strategies are experimentally validated on a prototype. The second part of this dissertation addresses the optimization of inverters, focusing on the operational improvement of three-phase four-leg inverters with neutral voltages generated by split dc-link capacitors. This topology is increasingly adopted for its great performance under unbalanced load conditions, making it ideal for applications such as electric-vehicles-based uninterruptible power supplies. The optimization challenge is tackled through the application of multi-carrier modulations to modify the ripple interactions between phases and neutral currents. This approach reduces the current flowing through the dc split capacitors, potentially decreasing power losses and reducing capacitor volume. Initially, static phase-shift modulations are applied to selected carriers, providing a simple method to reduce current without requiring hardware or control modifications. Subsequently, dynamic phase-shift modulation is employed to cancel the first harmonic of the current ripple, achieving further current reduction at the cost of increased implementation complexity. In summary, this dissertation presents significant contributions to the fields of converter modeling and control, as well as inverter optimization. The novel energy-based modeling approach and associated control schemes offer enhanced stability and operational benefits for dc-dc converters, while the proposed multi-carrier modulation techniques improve the efficiency and performance of three-phase four-leg inverters.