Electric ships have been developed in recent years to reduce greenhouse gas emissions. In this system, inverters are the key equipment for the permanent-magnet synchronous motor (PMSM) drive system. The cascaded insulated-gated bipolar transistor (IGBT)-based H-bridge inverter is one of the most attractive multilevel topologies for modern electric ship applications. Usually, the fault-tolerant control strategy is designed to keep the ship in operation for a certain period. However, the fault-tolerant control strategy with hardware redundancy is expensive and slow in response. In addition, after fault-tolerant control, the ship’s PMSM may experience shock and overheating, and IGBT life is reduced due to uneven switching frequency distribution. Therefore, a stratified reconfiguration carrier disposition Sinusoidal Pulse Width Modulation (SPWM) fault-tolerant control strategy is proposed. The proposed strategy can achieve fault tolerance without any extra hardware. A reconfiguration carrier is applied to improve the fundamental amplitude of inverter output voltage to maintain the operation of the ship’s PMSM. In addition, the available states of faulty H-bridge are fully used to contribute to the output. These can improve the life of IGBTs by reducing and balancing the power loss of each H-bridge. The principles of the proposed strategy are described in detail in this study. Taking a cascaded H-bridge seven-level inverter as an example, simulation and experimental results verify that the proposed strategy, in general, has a potential future application on electric ships.

Grid‐connected inverters are the core equipment in the renewable power system. There are multiple current sensors which may affect the driving module of the switch in inverter. In multiple current sensor fault diagnosis, the coupling between fault components makes fault diagnosis difficult. This paper presents an offset fault diagnosis method of multiple current sensors based on the average modulation voltage model. Based on the influence of current sensor offset fault in the system, the average modulation voltage model is established in three‐phase stationary coordinates. The difference between the measured value of the current and the actual value is estimated through the model when the offset fault occurs. And then the fault is located. Experimental results show that the fault can be located accurately and fault tolerant control can be performed by this method when there are offset faults in multiple sensors simultaneously.

Three-phase grid-connected inverters have been widely used in the distributed generation system, and the current sensor has been applied in closed-loop control in inverters. When the current sensor offset faults occurs, partial fault features of multiple current sensors disappear from the closed-loop control grid-connected system, which leads to difficulties for fault diagnostics and fault-tolerant control. This paper proposes a fault tolerance method based on average current compensation mode to eliminate these adverse effects of fault features. The average current compensation mode compensates the average of the three-phase current to the αβ axis current to realize the fault feature reconstruction of the current sensor. The mode does not affect the normal condition of the system. Then, the data-driven method is used for fault diagnosis, and the corresponding fault tolerant control model is selected according to the diagnosis results. Finally, the experimental results show that the proposed strategy has a good fault tolerance control performance and can improve the fault feature discrimination and diagnostic accuracy.

The extensive convolution computations, repetitive data transfers, and storage in convolutional neural networks (CNNs) result in prolonged fault diagnosis time. To address this issue, an in-memory convolutional neural network (IMCNN) is proposed for fast open-circuit fault (OCF) diagnosis of cascaded H-bridge multilevel inverters (CHMLIs). To be specific, an extended matrixization encoding (EME) method is first designed to convert the original 1-D diagnostic voltage signal. Then, convolutional calculations are physically performed based on an in-memory feature extraction (IM-FE) using memristor crossbar arrays, with pooling operations used to extract fault features. The extracted features are then input into a fully connected layer, and the Adam algorithm is used to train the model parameters to counter resistance drift. The trained model is then used to identify fault types. Experimental results on the three-phase five-level grid-connected CHMLI platform validate the performance of the proposed approach.