State-of-charge stream processing for electric vehicles

Abstract

This paper processes SoC (State of Charge) streams to develop a battery consumption model of EVs (Electric Vehicles) on a target road. A stream consists of a series of SoC records, each of which is made up of temporal and spatial stamps, altitude, SoC reading, velocity, and the like. For a specific road, a probe EV has collected SoC records while driving multiple times. The driving distance is calculated by adding up the line segment connecting two consecutive sensor points. For streams starting and ending at the same two end points, we plot the absolute SoC change first to figure out the SoC change pattern. Next, the consumed amount is traced to find the similarity between different streams. Finally, each stream is normalized to develop a common SoC change model out of multiple drives.