Thermal Stress Based Power Routing of Smart Transformer with CHB and DAB Converters

Vivek Raveendran, Markus Andresen, Giampaolo Buticchi, Marco Liserre

Research output: Journal PublicationArticlepeer-review

24 Citations (Scopus)


The smart transformer (ST) is a potential solution for an upgrade of the electric distribution grid, which enables to provide services to the grid and dc connectivity. However, the power electronics within the system are challenged by high reliability requirements. One possible solution to increase the reliability is to employ prognosis to predict the failures and avoid down times of the system. Traditional maintenance scheduling is based on the remaining useful lifetime (RUL) of the individual components or the forecasted failure probability. For a further increase of the time to the next maintenance, it is desired to have similar wear out of all components, which need to be maintained or exchanged. In this article, it is proposed to route the power internally in a modular power converter consisting of a cascaded H-bridge connected to dual active bridges in order to influence the RUL of its building blocks. Therefore, a thermal stress based wear-out control is designed for addressing the processed power dependent failures of the devices in the building blocks of the ST. Compared to the conventional power routing methods, the impact of the proposed system-level control considering electrical and thermal parameter variations is demonstrated using Monte Carlo analysis.

Original languageEnglish
Article number8798717
Pages (from-to)4205-4215
Number of pages11
JournalIEEE Transactions on Power Electronics
Issue number4
Publication statusPublished - Apr 2020


  • Cascaded H-bridge
  • dual active bridge
  • power routing
  • reliability
  • smart transformer
  • thermal stress

ASJC Scopus subject areas

  • Electrical and Electronic Engineering


Dive into the research topics of 'Thermal Stress Based Power Routing of Smart Transformer with CHB and DAB Converters'. Together they form a unique fingerprint.

Cite this