New licentiate thesis shows how to include uncertainties in residential grid simulations

2021-04-19

On 9 April Umar Hanif Ramadhani at the Division of Civil Engineering and Built Environment presented his licentiate thesis Uncertainty and correlation modeling for load flow analysis of future electricity distribution systems: Probabilistic modeling of low voltage networks with residential photovoltaic generation and electric vehicle charging. The thesis shows how improved uncertainty and correlation modeling can improve probabilistic simulation of electricity distribution grids.

Residential buildings, and hence the electricity distribution grids that these buildings are connected to, will include many new technologies that increase the uncertainty in both local supply and demand of power to and from the grids. In particular, photovoltaic (PV) systems will supply power to the buildings and to the grid depending on the local weather conditions, and electric vehicles (EVs) will have to be charged, potentially with a high charging power, depending on the driving habits of the residents.

Umar shows in his thesis how probabilistic load-flow (PLF) simulations of residential distribution grids can be improved by including more realistic simulation models for EV and PV technologies that consider uncertainty and correlation in production and demand but also the spatial allocation in the grid. The results show that differences in allocation can have a considerable impact on grid performance at different penetration levels. An energy management system can improve the performance, but also increases the correlation between nodes in the grid. The thesis concludes that it is important to include both correlation and uncertainty modeling to properly asses new building-related technologies on the grid, and that it is important to study these different technologies in combination.

The thesis was reviewed and discussed at the seminar by Sarah Rönnberg, Associate Professor at the Division of Energy Science at Luleå University.  

Umar’s licentiate thesis can be read in its entirety here.