Photovoltaic (PV) grid-connected system is nowa-days a well established industry. For PV inverters, European (EURO) efficiency and California Energy Commission (CEC) efficiency have been defined, as Figures of Merit (FoM) allowing easy PV inverter efficiency estimation, while including consideration on PV variability. This paper explores the applicability of weighted efficiencies for off-grid H2production system with DC-DC connection of the PV source, for both DC-DC converter, and electrolyzer. The study considers precise modeling of the system, and evaluation of its operation and efficiency over real scenario, with environmental data of three locations considered, with minute-based resolution, allowing precise simulation. Evaluation of EURO and CEC efficiencies for DC-DC converter and electrolyzer demonstrate the accuracy of the FoM with error lower than 1.23% and 3.70% respectively. It is finally recommended the use of weighted efficiencies for the evaluation of off-grid H2production systems, allowing fast and accurate comparison between different electrolyzer and DC-DC converter technologies while designing new plants.

Traditional linear controllers, such as the proportional-integral (PI) control employed in power converters, are prevalent across various power electronics applications. The efficacy of linear controllers is intricately tied to the specific operating point for which the control is designed. A misalignment between the designed operating point and the actual operating conditions results in performance degradation of the control system, thereby compromising the optimal operation of the converters. To address this disadvantage, a feedback linearization method applied to classical DC/DC converter topologies, including the buck, boost, buck-boost, and non-inverting buck-boost configurations, is presented in this paper. In contrast to alternative linear control approaches, feedback linearization enables comprehensive control of the converters across all operating points. Furthermore, it facilitates easy tuning of controller parameters, such as proportional and integral gains for a PI controller. Simulation and experimental results are employed to demonstrate the efficacy of the proposed method.

Multilevel converters emerge as a promising technology for high-power medium-voltage applications due to their different advantages. On the other hand, model predictive control (MPC) is one of the control techniques that has been widely used in different applications because of its features, such as fast transient response and flexibility to include nonlinearities. The finite control set MPC (FCS-MPC) approach for a seven-level topology converter reduces the number of calculations significantly, nevertheless depends on the circuital model making it sensitive under uncertainties. On the other hand, model-free predictive control (MF-PC) has gained attention due to its accurate prediction without previous knowledge of the power system. This paper proposes MF-PC to predict the current for each phase for the seven-level topology converter. Moreover, simulation results demonstrate superior performance and feasibility of the developed control method compared to the FCS-MPC approach applied to a seven-level topology converter.