Switching-function-based detailed equivalent model for numerically efficient electromagnetic transient simulation of multilevel power converter systems

Electromagnetic transient (EMT) simulation has become an essential tool for the analysis and design of modern power electronic systems. However, achieving efficient and accurate simulation of complex converter topologies remains challenging, as conventional EMT simulation involves establishing the detailed companion models of a large number of switching devices and other circuit components. Although a small simulation time step can achieve numerical accuracy, it also results in tremendous computational burden and significantly reduce simulation efficiency. A newly proposed simulation approach, known as the switching-function-based detailed equivalent model (SFB-DEM), has been developed to improve computational efficiency while maintaining simulation accuracy. Nevertheless, the implementation of this modeling approach has primarily been applied to one-terminal simple power converter topologies such as DC/AC half-bridge two-level converters, and its effectiveness for more complex multilevel converter systems has not yet been fully investigated. This work extends the SFB-DEM approach to a back-to-back AC/DC/AC five level converter and develops a set of predetermined voltage and current insertion factors based on semiconductor switching functions to ease computational burden for numerically efficient simulation. To demonstrate the improvement in simulation efficiency, the proposed SFB-DEM is implemented in MATLAB environment, and a detailed model (DM) of the investigated power converter topology is constructed using the widely-used Simulink/SimPowerSystems toolbox. The comparison results in case studies have proved that the proposed SFB-DEM outperforms the Simulink-based DM in terms of simulation efficiency while maintaining numerical accuracy. Under open-loop and closed-loop control conditions, the simulation efficiency has been improved by approximately 20-fold and 40-fold, respectively.