Physics-Based Use Case Analysis: Route Planning and Dynamic Load Prediction for Battery-Electric Buses

Abstract

The electrification of transportation and goods delivery networks, including ground, marine, and air, is becoming a reality around the world. These networks are logistically complex, and with this transition there is an added knowledge gap in the understanding of the energy storage system, such as lithium-ion battery packs. In Seattle, the local transit agency King County Metro is a national leader in transitioning to a fully electrified bus fleet and is the case study for this work. As transit agencies, such as King County Metro, adopt electrified vehicles, there are many new parameters that must be considered in route planning, fleet maintenance, and charging logistics. The literature shows that a longitudinal vehicle dynamics model can be used to estimate energy and load requirements along a route [1,2]. This work utilizes these known vehicle models, vehicle specifications, and battery information in a software tool. Software inputs include geographical information system (GIS) data, ridership data, and vehicle acceleration profiles. Model results allow the user to rank routes using metrics designed to consider route efficiency, vehicle performance, and known battery degradation events, such as regenerative braking and high-power discharge. This work also introduces how the battery state of charge, overpotential, current window, and deviations from equilibrium influence the route rankings. These results provide both systems-level and module-level insights beyond conventional vehicle and logistics models, allowing the user to make decisions that are predictive rather than reactive and can improve the lifetime of the batteries on-board.[1] Gallet, M., Massier, T. & Hamacher, T. Estimation of the energy demand of electric busesbased on real-world data for large-scale public transport networks. Appl. Energy 230,344–356 (2018).[2] Asamer, J., Graser, A., Heilmann, B. & Ruthmair, M. Sensitivity analysis for energydemand estimation of electric vehicles. Transp. Res. Part D Transp. Environ. 46, 182–199(2016).