The neutral point clamped (NPC) three-level grid-tied converter is the key equipment connecting renewable energy and power grids. The current sensor fault caused by harsh environment may lead to the split of renewable energy. The existing sensor fault-tolerant methods will reduce the modulation ratio index of the converter system. To ensure continuous operation of the converter system and improve the modulation index, a model predictive control method based on reconstructed current is proposed in this paper. According to the relationship between fault phase current and a voltage vector, the original voltage vector is combined and classified. To maintain the stable operation of the converter and improve the utilization rate of DC voltage, two kinds of fault phase current are reconstructed with DC current, normal phase current and predicted current, respectively. Based on reconstructed three-phase current, a current predictive control model is designed, and a model predictive control method is proposed. The proposed method selects the optimal voltage vector with the cost function and reduces time delay with the current reconstruction sector. The simulation and experimental results showthat the proposed strategy can keep the NPC converter running stably with one AC sensor, and the modulation index is increased from 57.7% to 100%.

Modern induction motor (IM) drives with a higher degree of safety should be equipped with fault-tolerant control (FTC) solutions. Current sensor (CS) failures constitute a serious problem in systems using vector control strategies for IMs because these methods require state variable reconstruction, which is usually based on the IM mathematical model and stator current measurement. This article presents an analysis of the operation of the direct torque control (DTC) for IM drive with stator current reconstruction after CSs damage. These reconstructed currents are used for the stator flux and electromagnetic torque estimation in the DTC with space-vector-modulation (SVM) drive. In this research complete damage to both stator CSs is assumed, and the stator current vector components in the postfault mode are reconstructed based on the DC link voltage of the voltage source inverter (VSI) and angular rotor speed measurements using the so-called virtual current sensor (VCS), based on the IM mathematical model. Numerous simulation and experimental tests results illustrate the behavior of the drive system in different operating conditions. The correctness of the stator current reconstruction is also analyzed taking into account motor parameter uncertainties, especially stator and rotor resistances, which usually are the main parameters that determine the proper operation of the stator flux and torque estimation in the DTC control structure.

In modern drive systems, the aim is to ensure their operational safety. Damage can occur not only to the components of the motor itself but also to the power electronic devices included in the frequency converter and the sensors in the measurement circuit. Critical damage to the electric drive that makes its further exploitation impossible can be prevented by using fault-tolerant control (FTC) algorithms. These algorithms are very often combined with diagnostic methods that assess the degree and type of damage. In this paper, a fault classification algorithm using an artificial neural network (ANN) is analysed for stator phase current sensors in AC motor drives. The authors confirm that the investigated classification algorithm works equally well on two different AC motors without the need for significant modifications, such as retraining the neural network when transferring the algorithm to another object. The method uses a stator current estimator to replace faulty sensor measurements in a vector control structure. The measured and estimated currents are then subjected to a classification process using a multilayer perceptron (MLP), which has the advantage of a small structure size compared to deep learning structures. The uniqueness of the method lies in the use of data in the training set that are not dependent on the parameters of a specific motor. Four types of current sensor faults were studied, namely total signal loss, gain error, offset, and signal saturation. Simulations were performed in a MATLAB/SIMULINK environment for drive systems with an induction motor (IM) and a permanent magnet synchronous motor (PMSM). The results show that the algorithm correctly evaluates the type of damage in more than 99.6% of cases regardless of the type of motor. Therefore, the results presented here may help to develop universal diagnostic methods that will work on a wide variety of motors.