Model Development and Predictive Control of a Low-Inertia DC Solid-State Transformer (SST)

In this article, a dynamic model is developed for a low-inertia dc solid-state transformer (SST) formed by inductor-less modular multilevel converters (MMCs) based on phase-shifted square-wave modulation (PS-SWM). Unlike other SSTs, both power flow and circulating energy need to be controlled quickly and accurately to achieve high performance during both transients and steady state. The developed model explicitly defines the circulating energy and retains its accuracy with an arm inductor-less operation. Furthermore, the model can be treated as time-invariant under square-wave modulation and analyzed in the small-signal domain to obtain system transfer functions, allowing small-signal analysis. Based on the developed model, predictive control is designed to achieve fast dynamic control with reduced computational burden. In addition, a feedforward current control is proposed to suppress the steady-state current harmonics of the dc current caused by the enlarged circulating energy due to the low-inertia feature. A medium-voltage dc (MVDC) simulation case study and experimental results on downscaled low-inertia SST hardware are provided to validate the developed model and the predictive controller performance.