This paper presents a novel independent model predictive control (MPC) strategy for a Multilevel Inverter (MLI) with an LC output filter and a split DC-link. The core innovation lies in the independent yet coordinated control of the AC and DC sides of the inverter using two distinct MPC techniques. The AC side is managed via a Modulated Model Predictive Control (M2PC), ensuring a fixed switching frequency and improved harmonic performance, while the DC-link capacitor voltages are regulated through a Finite Control Set MPC (FCS-MPC), leveraging the redundant voltage vectors to maintain voltage balance without influencing AC performance. Unlike conventional predictive control approaches that rely on a unified cost function incorporating both AC and DC variables, with carefully tuned weighting factors, this method decouples the control objectives. This not only eliminates the need for weight tuning but also significantly reduces computational complexity and enhances controller robustness. The proposal simplifies the design and implementation of each controller and allows for independent optimization of dynamic behavior on each side of the inverter. Results demonstrate substantial improvements in total harmonic distortion (THD), steady-state error (SSE), and neutral point voltage stability when compared to traditional M2PC controllers with joint cost functions. Furthermore, the modular structure of the proposed controller enables intuitive constraint inclusion, such as current limiting, and exhibits excellent transient behavior without sacrificing steady-state quality.

sponsorship: This work is part of the project 2020-EXT-007, "MIRA-ESTE: Specific, innovative microgrids solutions (accounting for environmental, social, technological and economic aspects) for isolated rural areas of Ecuador" from the Research Group of Propagation, Electronic Control, and Networking (PROCONET) of Universidad de las Fuerzas Armadas ESPE and KU Leuven. This work has been developed with the support of VLIR-UOS and the Belgian Development Cooperation (DGD) under the project EC2020SIN322A101. (Universidad de las Fuerzas Armadas ESPE|2020-EXT-007, KU Leuven, VLIR-UOS, Belgian Development Cooperation (DGD)|EC2020SIN322A101)

Several studies have been conducted on Hybrid Power Generation Systems (HPGS) to reduce the existing gap in access to electricity, especially in isolated and difficult access areas. The appropriate use of HPGS allows to power a community, reduces environmental pollution, and improves the performance of a system that only uses Fossil Fuel Generators (FFG). However, despite the efforts made in several countries, there are still isolated areas that remain without electricity due to several factors, such as the high cost of generation, transmission, and energy distribution. For this reason, this work seeks to generate new viable solutions to improve access to electricity especially in isolated areas of Ecuador, although the tool can be applied to any community without constant electricity access. For this purpose, the modeling of an HPGS is proposed, such system is composed of an FFG, a photovoltaic generator, and a hybrid energy storage system (i.e., electrical and thermal). A control algorithm for the FFG is used to adapt the motor rotational speed by using a minimum fuel consumption to supply the load profile. The results are obtained from simulations with real data of three case studies. A reduction in fuel consumption, greenhouse gas emissions, and loss power from renewable energy is demonstrated.

A fuzzy logic-based energy management system is proposed to allow the power exchange (PE) between the storage systems of multiple interconnected residential microgrids. The adjacent microgrids are connected to the utility network and comprise a solar roof-top photovoltaic system and an Energy Storage System (ESS). The proposed strategy for PE considers the battery state-of-charge and the rate of power change of each microgrid. The strategy's objective is to use the excess energy in the ESS of one microgrid to charge the ESS of another one that has a power shortage. Therefore, it is possible to optimize the energy generated by the renewable energy system and reduce the peaks and fluctuations in the power profile that each microgrid exchanges with the mains. Simulations and comparison results under different scenarios with and without PE using up to five interconnected microgrids, demonstrates the effectiveness of the proposed PE management while preserving the batteries' life.

This paper presents the comparison of two modulated model predictive control (M2PC) strategies (the classical M2PC approach, and the M2PC with optimized overmodulation) applied to a three-phase three-level voltage source inverter (3Ph-3L-VSI). The control strategies are tested under various scenarios: no-load, a linear load connection, and a nonlinear load connection. Although each algorithm has its advantages, simulations show that the overmodulated version presents a substantially better performance than the classical version since the classical M2PC creates gaps between switching regions, making rough transitions between them. In contrast, the overmodulation version can produce a continuous transition between those regions, resulting in lower THD and control effort.

During the last century, population growth, together with economic development, has considerably increased the energy demand and, although renewable energies are becoming an alternative, still total energy supply is mainly non-renewable, causing well-known negative effects such as pollution and global warming. On the other hand, technological advances have allowed the development of increasingly efficient distributed generation systems and the emergence of microgrids, whose studies have been focused on architecture, elements, and objectives of the associated energy management strategies. In this regard, energy management strategies based on a Fuzzy Logic controller have been developed for electro-thermal microgrids where parameter optimization has been carried out through heuristic procedures of trial and error with acceptable results but involving a high computational cost. To solve the aforementioned drawbacks, in the present work the use of Cuckoo Search optimization nature-inspired algorithm that allows the adjustment of Fuzzy Logic controller parameters and ensures a higher quality of energy management is proposed. Obtained results show encouraging outcomes for the use of these meta-heuristic optimization algorithms.

In this paper, an energy management strategy based on microgrid power forecasting is applied to a residential grid-connected electro-thermal microgrid with the aim of smoothing the power profile exchanged with the grid. The microgrid architecture under study considers electrical and thermal renewable generation, energy storage system (ESS), and loads. The proposed strategy manages the energy stored in the ESS to cover part of the energy required by the thermal generation system for supplying domestic hot water to the residence. The simulation results using real data and the comparison with previous strategy have demonstrated the effectiveness of the proposed strategy.

This paper extends a previous design of an energy management strategy based on Fuzzy Logic Control (FLC) for a residential grid-tied microgrid including a Hybrid Renewable Energy System (HRES) and an Energy Storage System (ESS). The main goal of the proposed design is to improve the grid power profile while satisfying the constraints established by the ESS. The strategy extension includes generation and demand forecasting in order to predict the future behavior of the microgrid. According to the forecast error and the battery State-of-Charge (SOC) the proposed strategy increases/decreases the grid power with the purpose of smoothing the power exchanged with the grid. The performance of the proposed strategy is verified through comparative numerical simulations using real data measured at the microgrid of the Public University of Navarre (UPNa).

An improvement of fuzzy logic controller strategy for a grid-tied domestic microgrid is presented. The fuzzy logic controller takes onto account the microgrid power flux trend in order to minimize the power profile exchanged with the main grid while preventing at the same time the battery life time against deep cycles of charging and discharging. As well as in previous work, the fuzzy logic controller is designed to minimize a set of quality indices. The benefits of the new approach are highlighted through numerical simulations using real generation and consumption measured data.

A fuzzy logic controller design for the battery management of a grid-tied real domestic microgrid including 6 kWp of PV and 6kW wind power renewable generators is presented. The fuzzy controller parameters design is undertaken on the basis of the minimization of a set of quality indices such as the battery dynamic range and the power variation range exchanged with the grid. The features of the proposed approach are highlighted with yearly real data by numerical simulations of the power profile injected into the grid and the battery state of charge.