Quantifying the Charging Flexibility of Electric Vehicles; An Improved Agent-Based Approach with Realistic Travel Patterns

Abstract

Existing modelling research that attempts to quantify how flexible charging sessions of electric vehicles are, have been constrained by either charging data or inadequate mobility data. This resulted in significant underestimations of the charging flexibility. In this article an agent-based model is developed that is able to quantify the charging flexibility more realistically. A new charging flexibility metric is defined that takes future trips and the state-of-charge of the vehicles into account. The developed approach leverages detailed activity patterns from the ALBATROSS-model by simulating vehicle utilization from a household perspective in different neighborhood types. The results show that the over 80% of the evening peak charging demand from electric vehicles can be mitigated when utilizing the charging flexibility. It also shows that about half of the charging demand can be extended by more than 40 h. These results demonstrate the great potential of electric vehicle to balance the grid and enable high degrees of renewable energy production.