Model-predictive control (MPC) has emerged as a promising control method in power electronics, particularly for multiobjective control problems such as multilevel inverter (MLI) applications. Over the past two decades, improving the performance of MPC and tackling its technical challenges, such as computational load, modeling accuracy, cost function design, and weighting factor selection, have attracted great interest in power electronics. This article aims to discuss the current state of MPC strategies for MLI applications, describing the significance of each challenge with the reported effective solutions. Through this review, the MPC methods are categorized into two groups: direct MPC (without modulator) and indirect MPC (with modulator). The recent advances of each category are presented and analyzed, focusing on direct MPC as the most applied method for MLI topologies. In addition, some of the important concepts are experimentally validated through a case study and compared under the same operating conditions to evaluate the performance and highlight their features. Finally, the future trends of MPC for MLI applications are discussed based on the current state and reported developments.

Multilevel converters (MLCs) are widely recognized for their exceptional benefits and have emerged as the preferred choice for medium- and high-power/voltage applications. Their usage has also been extended to low-power applications to overcome issues associated with high switching frequencies and electromagnetic interference (EMI) commonly encountered in two-level converters. Common dc-link MLCs have received particular attention in industry due to their ability to eliminate the need for bulky and inefficient transformers and rectifiers, making them a compelling option for different applications, primarily medium- and high-power/voltage applications such as wind turbine (WT) power conversion systems. Furthermore, common dc-link topologies are required for back-to-back (BTB) configurations, as they facilitate the use of a shared dc-link between the rectification and inversion stages. Despite their popularity, there is currently no comprehensive review article dedicated to common dc-link topologies. This paper addresses this gap by presenting a comprehensive review of common dc-link MLCs, covering their topological evolution, features, topologies comparison, modulation techniques, control strategies, and industrial application areas. Additionally, future perspectives and recommendations are discussed to provide researchers and engineers with a better understanding of the potential applications and advantages of these converters.

Model-predictive control is a constrained optimization control method with superior performance than linear methods for multivariable and multiobjective control of power converters. Nonetheless, its performance is limited by model uncertainties and measurement noise. This study tackles this challenge by proposing a new hybrid parallel-cascade extended state observer (PC-ESO) with two key advantages: 1) higher disturbance rejection than the conventional linear ESO and cascade ESO (CESO) at low bandwidth and 2) better noise suppression than the conventional ESO. PC-ESO's time-domain structure and comprehensive frequency-domain analysis are presented. Furthermore, PC-ESO is applied to improve the transient disturbance rejection of CESO through a novel structurally adaptive ESO (SAESO) algorithm. The proposed SAESO provides both high-frequency noise suppression and better disturbance rejection than CESO and cascade-parallel ESO. Finally, the proposed methods are experimentally validated by the model-free predictive control of a grid-connected power converter.

Efficiency is a crucial concern for dual active bridge (DAB) converters. Although many studies on novel modulation schemes have attempted to improve the efficiency, achieving the optimal balance between current stress and zero-voltage-switching (ZVS) ability remains challenging. In addition, due to the wide operating range of DABs, small-signal models have to be setup at each equilibrium point to purpose a high performance of liner controllers. Existing works mainly employ the slow-loop to prevent oscillations during transitions, which degrades the dynamic and bandwidth. To address the current state of research, this article proposes a triple-phase-shift modulation combined with model predictive control (MPC) for DABs. Herein, a designed cost function initially calculates shift-angles. Subsequently, a proposed optimization segment for ZVS checks and revises the ZVS ability to select an optimal set of shift angles. The current stress is minimized on the premise that the wide range of ZVS is guaranteed. Meanwhile, the dynamic performance is improved through the nonlinearity of MPC. Finally, a 1.4-kW scaled-down prototype is built, and the experimental comparisons with other schemes verify the effectiveness and superiority of the proposed method.

This article proposes an ultralocal predictive surface-mounted permanent magnet synchronous motor (SPMSM) model-based predefined time sliding mode predictive speed control (UL-SMPC) to achieve exceptional disturbance rejection and tracking performance in SPMSM systems. First, an ultralocal predictive SPMSM model is given, incorporating a time-varying disturbance term and an adaptive control gain. Second, the control gain and disturbance term of the SPMSM model are decoupled and identified, respectively. A control gain optimizer is devised to estimate the control gain, and a predefined time reaching law-based generalized proportional integral observer (PT-GPIO) is developed to estimate the disturbance for each control period. The parameter tuning principles for the control gain optimizer and PT-GPIO are analyzed. Third, a cost index is defined using the predefined time reaching law-based sliding mode surface. Finally, UL-SMPC is synthesized by minimizing the cost index. Experimental results verify the outstanding robustness and tracking performances of the proposed method.

Model predictive control (MPC) usually suffers from high computational complexity when it comes to modular multilevel converters (MMCs). Some researchers have attempted to use a modulated approach to reduce the computational burden and improve the control performance. But these methods do not consider the actual physical limitations of the control system, and therefore the control performance degrades at high modulation indices or transients. To solve this problem, a modulated MPC with bound-constrained quadratic programming has been proposed. With this method, the optimal solution of the control problem can be obtained, ensuring a better control performance under high modulation index conditions or in transients. Finally, a comparative experiment with the conventional modulated MPC methods has been carried out. The experimental results validate that the proposed method can achieve superior performance when the MMC operates at high modulation index, transients, and low frequencies.