Simulation and Fault Detection Techniques for Multilevel Inverters used in Smart Grids

A smart grid includes a variety of intelligent control systems for power generation and also for increasing the energy efficiency of the system. These systems include smart power converters, but also renewable energy and highly efficient resources. Multilevel inverters are considered modern and basic elements of high voltage power supply systems; however, the failure of electronic power components strongly affects the reliability of the entire system. The important goal in defining and dimensioning of smart grids is the measuring and determining of the boundary conditions and operating regimes of the power components. This paper presents an innovative technique for the fault detection of power components for a three-phase multilevel inverter, which is one of the typical components of a smart grid. Single, double and triple switch failures can be diagnosed by this method. The detection mechanism is based on the analysis of the stator current, with the possibility to differentiate between the simple and multiple switching failures. The paper also approaches multilevel converter models, which are systematically made from basic components and use configurable subsystems to switch between architectures of different converter topologies. Also, these models can achieve the harmonic analysis to verify the system correctness and can simulate different operational scenarios to test the functioning of the control system.