Solid State Transformers (SSTs): A Potential Game Changer for Future Power Distribution Grids

Abstract

The power distribution system is going through a brisk transformation with a higher penetration of renewable energy sources, battery energy storage and DC loads like electric vehicles and data-centres. This has necessitated increasing power electronics based control of power flow within the distribution grid. Solid-state transformer (SST) is an emerging power electronic technology, also referred to as power electronic transformers, energy routers, etc., which is a cascaded combination of power converter cells with medium-frequency isolation. SSTs are used to interface a medium-voltage (MV) network (e.g. the utility distribution grid) with a low-voltage (LV) network (e.g. a microgrid). It has been shown in recent literature that for MVAC-LVDC applications (e.g. where solar energy or battery integration is required), SSTs have proven to be a competitive technology, in terms of efficiency, power-density and cost, compared to the on-load tap-changer (OLTC) transformers. This tutorial will discuss various aspects of control, reliable operation, practical implementation and optimal design of SSTs. It will start off with a brief introduction of the SST technology and its potential in future smart-grid paradigm, as well as provide a succinct overview of the various SST topologies. Secondly, the discussion will shift to a particular topology called Cascaded Modular SST (CMSST), where modelling and control aspects will be discussed in detail. One of the key challenges in adapting power electronic transformers is its reliability. Therefore, special attention is given to fault diagnosis and fault tolerant control schemes for SSTs. Thirdly, the practical implementation aspect of the SST based on industrial design and product requirements will be discussed in some detail. The critical aspects of design layouts, design margins, costing analysis on long run, component choices will be dealt with. Some critical components such as high frequency transformer mandate custom design and development due to insulation requirements. The need for development of custom tools for design is also discussed here based on the design of a 6.6 kV SST’s hardware implementation. Subsequently, a multi-objective optimal design strategy for the grid-connected SST technology will be discussed. Here, a hybrid (analytical+numerical) model based local optimization strategy, followed by a global optimization strategy using limited number of optimal datasets fed to low data-hungry machine-learning algorithms will be explained which aim to reduce the overall computational burden. Apart from design optimization results for a realistic 22 kV, 1 MVA SST, scaled-down experimental measurements and subsequent validation for a 1.5 kV, 15 kVA SST’s design showing the merits of the discussed strategy. Finally, the tutorial will conclude with the main takeaway message from this tutorial and the potential areas/ applications where SST technology has the capability to dominate in the near future.