Electric vehicle capacity forecasting model with application to load levelling

Abstract

There are many uncertainties associated with forecasting electric vehicle charging and discharging capacity due to the stochastic nature of human behavior surrounding usage and intermittent travel patterns. This uncertainty if unmanaged has the potential to radically change traditional load profiles. Therefore optimal capacity forecasting methods are important for large-scale electric vehicle integration in future power systems. This paper develops a capacity forecasting model considering eight particular uncertainties under three categories to overcome this issue. The model is then applied to a UK summer scenario in 2020. The results of this analysis demonstrate that the proposed model is accurate for charge and discharge prediction and a feasible basis for steady-state analysis required for large-scale electric vehicle integration.