Model predictive control (MPC) for modular multilevel converter (MMC) systems has drawn attention among researchers in recent years due to its straightforward implementation, ability to control multiple objectives in a single cost function, and excellent dynamic response. Even though MPC seems promising for the MMC, it suffers from an excessive increase in computational complexity when the number of submodules (SMs) per arm (N) is increased. The computational load can be reduced by decoupling the SM capacitor voltage control from the cost function and balancing them in an external voltage sorting algorithm. In this paper, a further reduction of computations in the MPC is achieved by the preselection of control options that can satisfy the control objectives in the next sampling time. The preselection algorithm generates a greatly reduced number of control options to be computed by a straightforward MPC loop at each sampling time. In addition, the MPC together with the preselection algorithm can generate 2N + 1 output levels in the MMC while maintaining a low dv/dt in the output voltage and suppressing the circulating currents. The steady-state and dynamic performance of the MMC operating with the proposed method is verified by simulation and experimental results.

The wide variety of arc faults induced by different load types renders residential series arc fault detection complicated and challenging. Series dc arc faults could cause fire accidents and adversely affect power systems if not promptly detected. However, in practical power systems, they are difficult to detect because of a low arc current, absence of a zero-crossing period, and various abnormal behavior based on different types of power loads and controllers. In particular, conventional protection fuses may not be activated when they occur. Undetected arc faults could cause false operation of power systems and potentially lead to damage to property and human casualties. Therefore, it is imperative to develop a detection system for series arc faults in DC systems for the reliable and efficient operation of such systems. In this study, several typical loads, especially nonlinear and complex loads such as power electronic loads, were chosen and analyzed, and five time-domain parameters of the current—average value, median value, variance value, RMS value, and distance of the maximum and minimum values—were chosen for arc fault detection. Various machine learning algorithms were used for arc fault detection and their detection accuracies were compared.

Nearest-level control (NLC) is a popular technique used in modular multilevel converters (MMCs) with a large number of submodules (SMs) owing to the NLC’s flexibility and ease of implementation. However, in medium-voltage applications, MMCs contain a relatively low number of SMs, and the drawbacks of the NLC methods emerge, wherein the poor quality of output voltages and currents result in high total harmonic distortion, large ripples in SM capacitor voltages, and unsuppressed circulating currents. Several NLC methods have been proposed to handle these problems, but they do not satisfy all the control objectives simultaneously. This paper proposed a modified NLC capable of enhancing the output quality of MMCs with low number of SMs without deteriorating the control objectives. Unlike previously reported NLC methods, instead of directly calculating the numbers of SMs from the upper and lower arm voltage references, the difference and total number of SMs are obtained from the output voltage reference and circulating current control, respectively. Hence, the numbers of SMs in the upper and lower arms are acquired by simply solving a system of first-order two-variable equations. The simulated and experimental results for a single-phase MMC system were used to verify the appropriateness and effectiveness of the proposed modified NLC method.

As one of the most vulnerable components to temperature and temperature cycling conditions in power electronics converter systems in these application fields as wind power, electric vehicles, drive system, etc., power semiconductor devices draw great concern in terms of reliability. Owing to the wide utilization of power semiconductor devices in various power applications, especially insulated gate bipolar transistors (IGBTs), power semiconductor devices have been studied extensively regarding increasing reliability methods. This study comparatively reviews recent advances in the area of reliability research for power semiconductor devices, including condition monitoring (CM), active thermal control (ATC), and remaining useful lifetime (RUL) estimation techniques. Different from previous review studies, this technical review is carried out with the aim of providing a comprehensive overview of the correlation between various enhancing reliability techniques and discussing the corresponding merits and demerits by using 144 related up-to-date papers. The structure and failure mechanism of power semiconductor devices are first investigated. Different failure indicators and recent associated CM techniques are then compared. The ATC approaches following the type of converter systems are further summarized. Furthermore, RUL estimation techniques are surveyed. This paper concludes with summarized challenges for future research opportunities regarding reliability improvement.

A Lyapunov‐function‐based technique for finite control set predictive control is proposed to control the load currents of three‐phase voltage source inverters (VSIs). The developed technique, based on a discrete‐time model of a VSI, determines a control law using the Lyapunov function. Based on the Lyapunov stability analysis considering inevitable quantisation errors between the proposed control law and control actions selected from the inherent finite control set of the VSI, all signals of the closed‐loop dynamics are uniformly ultimately bounded and the current control errors converge to a neighbourhood of the origin. In addition to rendering the finite control set predictive control system globally stable, the proposed Lyapunov‐function‐based finite control set predictive control reduces the amount of calculations required to predict a future variable by half compared with the conventional finite control set predictive control, resulting in lower actuation time delay. Experimental results with three‐phase VSIs are presented to validate the proposed Lyapunov‐function‐based control method.

Finite control set model predictive control (MPC) method needs calculations corresponding to the total number of voltage vectors made by power converters. This operating principle of the MPC leads to an increase in the computational load and number of calculations, which dramatically increase proportional to the increase in the level of cascaded H‐bridge (CHB) inverters. However, it is difficult to consider all voltage vectors to find the optimal vector for high‐level CHB inverters because of the short sampling time. This study proposes a new MPC algorithm to reduce the number of calculations for multi‐level CHB inverters, in which only three voltage vectors are considered as optimal vectors regardless of the level of CHB inverters. Consequently, compared with the conventional MPC method for CHB inverters, the proposed MPC method requires fewer calculations and reduces the computational load without affecting the system performance and dynamic response.