Four-level hybrid-clamped inverter is a newly proposed topology that can operate under a wide voltage range without switches connected in series. However, when it is applied in medium voltage high power conversions, the flying capacitors in each phase will occupy a huge volume and a high switching frequency is required to restrain the voltage ripples. In order to overcome this drawback, a four-level active NPC inverter is discussed in this paper, which consists of only 6 switches and no diodes or flying capacitors are required. In order to balance the neutral-point voltages under the full power factor and modulation index range, a capacitor voltage balance method based on carrier-overlapped PWM is proposed in this paper. The upper and lower dc-link capacitor voltages are balanced by zero-sequence voltage injection and the central dc-link capacitor voltage is balanced by adjusting the duty cycles of switching signals slightly. Simulation and experimental results are presented to confirm the validity of this method.

The neutral-point voltage unbalance problem holds back the extensive application of neutral-point clamped (NPC) multilevel converters with more than three voltage levels in industry. Traditional phase-disposition PWM (PDPWM) or nearest-three-vectors (NTV) modulation cannot achieve the voltage balance over the full modulation indexes and load power factors when the voltage level is higher than three. To solve this problem, a novel generalized carrier-based PWM method named carrier-overlapped PWM (COPWM) for NPC multilevel converters is proposed in this paper, which is proved to satisfy the volt-second balance principle. With this modulation method, the average values of all the neutral-point currents equal to zero in a fundamental period, and so all the dc-link capacitor voltages can be naturally balanced in the ideal and steady-state conditions. In order to simplify its implementation, an equivalent multi-reference modulation method with only one triangular carrier is also derived. Simulation and experimental results on a five-level diode-clamped inverter are presented to verify the proposed modulation method.

Neutral-point voltage imbalance is one of the main problems of five-level neutral-point clamped (NPC) inverters that impedes their industrial application. In order to solve this problem, a closed-loop neutral-point voltage balancing method based on carrier-overlapped PWM (COPWM) is proposed in this paper. The four DC-link capacitor voltages are balanced by three steps: voltage balancing between the two outer capacitors, voltage balancing between the two inner capacitors, and voltage balancing between the two inner and the two outer capacitors. The relationship between the neutral-point currents and the reference voltage is studied, and the top and the bottom DC-link capacitor voltages are balanced by zero-sequence voltage injection. The two inner capacitor voltages are balanced by adjusting the width of switching signals. Simulation and experimental results are presented to confirm the validity of this method.

Neutral-point clamped multilevel converters with more than three voltage levels are hard to balance the DC-link capacitor voltages, which limits their application in industry. In order to solve the neutral-point potential balancing problem of four-level neutral-point clamped converters, a novel carrier-overlapped PWM method is proposed in this letter, which satisfies the volt-second balance principle and can achieve the natural voltage balance of three DC-link capacitors under ideal and steady states. Experimental results are presented to confirm the validity of this PWM method.

In order to increase the output voltage levels and reduce the isolated dc sources, a symmetrical hybrid nine-level inverter for high-speed motor drive applications is presented in this paper. Each phase of this inverter is composed of a five-level dc/dc converter cell and an H-bridge cell. The dc/dc converter is operated at high frequency with low-voltage devices and the H-bridge is operated at fundamental frequency with high-voltage devices. The three phases are connected to a common dc-link and each drives an isolated winding of an open-winding motor. The operating principles and modulation method are introduced. A detailed analysis of the average currents through the flying capacitor and neutral point of the dc-link is presented and a capacitor voltage balancing method based on phase-shifted pulse-width modulation is proposed. Simulation and experimental results are presented to demonstrate the feasibility of this topology and control method.

The focus of this article is to introduce the concept of an online reinforcement learning (RL) solution and to propose a novel finite control-set model predictive control framework subject to system uncertainties, which possesses the excellent applicative potential for power converter systems with unknown perturbations. In this framework, the control task is performed by incorporating an adaptive neural network approximation-based RL and neural predictor-based predictive current control solution. To be more precise, a critic neural network is responsible for learning a strategic utility function online, and an actor network is developed to derive control behaviors by approximating the unknown model dynamics and optimizing the learned utility function obtained from the critic network. Compared to previous works, it not only attenuates the inherent issues of system uncertainties and unknown disturbances, but also provides a flexible framework and allows the enhancement of control property. Furthermore, by deploying the Lyapunov approach, it shows that all signals in the closed-loop system are uniformly ultimately bounded. Finally, numerical simulation and experiments validate our theoretical findings.