Voltage and power balance strategy without communication for a modular solid state transformer based on adaptive droop control.

Abstract

Solid State Transformers (SST) are attracting considerable attention due to their great application potential in future smart grids. It is an essential technology capable of promoting the modernization of the electric power distribution system and it is considered a key element for interfacing future microgrid systems to medium voltage utility grids, allowing plug-and-play integration with multiple renewable energy sources, storage devices and DC power systems. Its main advantages in relation to conventional transformers are substantial reduction of volume and weight, fault isolation capability, voltage regulation, harmonic filtering, reactive power compensation and power factor correction. A three-stage modular cascaded topology has been considered as an adequate candidate for the SST implementation, consisting of multiple power modules with input series and output parallel connection. The modular structure presents many advantages, e.g., redundancy, flexibility, lower current harmonic content and voltage stress on the power switches, however component tolerances and mismatches between modules can lead to DC link voltage imbalance and unequal power sharing that can damage the solid state transformer. This paper proposes a decentralized strategy based on adaptive droop control capable of promoting voltage and power balance among modules of a modular cascaded SST, without relying on a communication network. The behavior of the proposed strategy is assessed through a MATLAB/Simulink simulation model of an 100 kVA SST and shows that power and voltage balance are attained through inner power distribution of the SST modules, being transparent to elements connected to the transformer input and output ports. Besides that, real-time simulation results are presented to validate the proposed control strategies. The performance of embedded algorithms is evaluated by the implementation of the SST in a real-time simulation hardware, using a Digital Signal Processor (DSP) and high level programming.