Agent-based simulation to assess the impact of electric vehicles on power networks: Swindon Borough Case Study

Abstract

Due to air quality concerns and stricter carbon targets, surface transport electrification is quickly spreading, posing questions on the impact it will have on national and local electrical networks. This paper proposes an agent-based model to assess the per-minute weekday and weekend impact of the uptake of Electric Vehicles (EVs) over the next decade on local electrical and charging infrastructures, aimed at local decision-makers and stakeholders for transport electrification forecasting purposes. This study compares two scenarios, the first assessing the case where no restrictions are imposed on the daily charging events, and the second scenario where the peak electrical demand period between 5pm and 8.30pm is constrained for charging. Swindon Borough is selected as case study since it has one of UK’s highest EV adoption rates and has ambitious aims for decarbonisation. The results show that, over time, scenario two consistently lessened the constraints imposed on the grid by lowering the weekday and weekend peak loads up to 7% and 20%, respectively, and through lowering the usage rate of the charging infrastructure by around 12%. This scenario postponed the 5pm to 8.30pm EV charging demand to later hours, resulting in delayed load waves in residential areas that, over time, took values of higher proportion of the daily peak, forecasted to match it by 2036. However, controlling the EV demand through this strategy became less effective over time, and so, constraining charging between 7am and 2.30pm is suggested for further control. To conclude, this scenario can be portrayed in reality by adding flexibility to the grid, through the use of time-of-use tariffs (TOUTs), hence, if well implemented, postponing the upgrade of the grid and the charging infrastructure, presenting savings to the network operator, charging network stakeholders and EV users. The paper thus highlights the advantages of using a model of a heterogeneous population with fine spatial and temporal detail to provide decision-support to key stakeholders in planning energy transitions.