Cascading of diode bypassed transistor‐voltage‐source units in multilevel inverters

Abstract

Multilevel inverters are gaining interest in industry, especially for renewable energy sources, as a means to reduce output distortion and also reduce electromagnetic interference effects. However, multilevel inverters have a high-transistor component count. In order to reduce the number of power transistors within a multilevel inverter, a bypass diode technique is suggested for two multilevel inverter topologies. In the first inverter configuration, referred to here as the diode-bypassed multilevel dc-link inverter, one H-bridge is required for the entire single-phase system, with one transistor and one bypass diode needed for each voltage source used in the inverter. Each voltage source contributes in supplying power to the load in both the positive and negative half cycles of the output waveform. In the second configuration, referred to here as the diode-bypassed neutral point inverter, one transistor and one bypass diode is required for each voltage source, plus two additional transistors for the entire single-phase inverter system. For this configuration, each dc source contributes only to one half-cycle of the output waveform, either positive or negative. For both configurations, it is possible to ensure even power distribution among all the voltage sources (or compensate for imbalance in the state of charge of the dc sources). Experimental results suggest the feasibility of constructing the proposed inverter topologies and simulations provide preliminary power-efficiency data.