Nørgaard, Jacob Bitsch

Nørgaard, Jacob Bitsch

PROJECT TITLE: Multi-time Scale Modelling of Power Electronic Converters in Power System Applications    

PhD period: 2017.09.01 – 2021.08.31.
Section: Power Electronic Systems
Research Programmes: Efficient and Reliable Power Electronics 
Supervisor: Frede Blaabjerg
Co-Supervisors: Tomislav Dragicevic and Yongheng Yang
Contact Information

Collaborator: To be announced later.
Funding: Villum Fonden.


The increasing penetration of renewable energy sources has facilitated an expansion in the number of power electronic converters in power systems. Studying the effects of this is crucial for understanding the stability and reliability of the overall system. In this aspect, the possibility to have a unified model which can account for multiple stressors is indispensable.

The Center of Reliable Power Electronics (CORPE) at Aalborg University has proposed a multi-time-scale mission profile which accounts for temperature as a stressor at the component level. However, to fully assess a multi-time scale power electronic converter based power system with multiple stressors, further research is needed.

Mutual interactions of multiple converters at multiple time scales with respect to reliability and power grid stability is not trivial. This statement becomes even more accurate when the models are extended to include multiple physics in time scales ranging from microseconds to years. There are many factors that influence the overall system performance and reliability, ranging from switching transients to intricate failure modes, converter control strategies, and system resonances.

The trade-off between model accuracy and simulation speed is a well-known problem in engineering. Highly accurate Finite Element Multi-Physics based models can be constructed in simulation software which will precisely predict the performance of a certain component. The problem arises when the system complexity becomes so large that the simulation cannot be conducted in a feasible time frame. There are different methods to work around this issue, however, they all require a relaxation of the problem formulation. Either in the form of the discretization time step or in the model complexity. The choice of how to solve this is determined by a precise selection of what information should be extracted from the model. The cross-dependency of parameters only further complicates this process.

This leads to the need of a multi-time scale modelling approach.


Publications in journals and conference papers may be found at VBN.