Fleet data-based energy consumption prediction framework (EPF)

An All-Electric Alpine Crossing: Time-Optimal Strategy Calculation via Fleet-Based Vehicle Data


Recently, individual electric mobility gains significance due to legislation and social discussion. Customers demand longer battery ranges. Advanced planning is a different and more sustainable approach. Potentially, they assist drivers in exploiting the installed range on long journeys. Earlier research of the authors showed that an optimal combination of speed, charging choice and amount potentially reduces overall traveling time on long trips. In this work, a dynamic programming algorithm controls this strategy set time-optimally on an all-electric route from Munich to Verona. For this, location-specific fleet-based data of over 600 000 km are used to improve the reliability of the strategy set in two ways. Firstly, the data provide more realistic location- and time-specific velocity bounds for speed control. Secondly, they provide fleet-sourced dynamics to a traceable analytical consumption model. These additional dynamics lead to 1.8 - 2.3 $ more energy demand in the strategy planning compared to a less accurate consumption map-based approach. Here, the incorporation of dynamics increases the optimizations’ reliability. Also, the time-dependent fleet-data allows finding an optimal departure time for the given route. In total, the incorporation of fleet information enhances the robustness of the optimization. This enables a more seamless experience of electric mobility on long trips.