Vehicle-to-grid for car sharing -- A simulation study for 2030

TL;DR

This study analyzes vehicle-to-grid (V2G) opportunities for a national-scale car sharing fleet in Switzerland using an end-to-end, data-driven agent-based pipeline. It combines a synthetic population, a data-driven mode-choice model trained on MOBIS data, and a scalable V2G optimization via ADMM to quantify power flexibility (12–50 MW depending on scenario) and peak shaving under 2030 scenarios. Six scenario designs fuse user growth with business roadmaps and station expansion, revealing that adding stations can boost demand and V2G flexibility, while larger fleets enhance scheduling feasibility and resilience. A case study with real electricity prices demonstrates a sweet spot for mutual gains between fleet operators and DSOs, highlighting policy and market design needs to unlock fleet-based ancillary services.

Abstract

The proliferation of car sharing services in recent years presents a promising avenue for advancing sustainable transportation. Beyond merely reducing car ownership rates, these systems can play a pivotal role in bolstering grid stability through the provision of ancillary services via vehicle-to-grid (V2G) technologies - a facet that has received limited attention in previous research. In this study, we analyze the potential of V2G in car sharing by designing future scenarios for a national-scale service in Switzerland. We propose an agent-based simulation pipeline that considers population changes as well as different business strategies of the car sharing service, and we demonstrate its successful application for simulating scenarios for 2030. To imitate car sharing user behavior, we develop a data-driven mode choice model. Our analysis reveals important differences in the examined scenarios, such as higher vehicle utilization rates for a reduced fleet size as well as in a scenario featuring new car sharing stations. These disparities translate into variations in the power flexibility of the fleet available for ancillary services, ranging from 12 to 50 MW, depending on the scenario and the time of the day. Furthermore, we conduct a case study involving a subset of the car sharing fleet, incorporating real-world electricity pricing data. The case study substantiates the existence of a sweet spot involving monetary gains for both power grid operators and fleet owners. Our findings provide guidelines to decision makers and underscore the pressing need for regulatory enhancements concerning power trading within the realm of car sharing.

Figures (18)

• Figure 1: Overview of synthetic car sharing data generation pipeline. A synthetic population with mobility profiles is generated from projected census data. On the other hand, a mode choice model is trained on labeled tracking data, using features such as trip distance, day times, and car availability as input. This model is applied to the synthetic data, yielding a set of synthetic car sharing trips.
• Figure 2: Mode share among trips of MOBIS users with a car sharing subscription
• Figure 3: Exemplary distribution of new stations in the simulated car sharing service. For visualization purposes, the population locations are displayed as a distribution (blue) via Kernel Density Estimation. The new stations generated with our algorithm (yellow) follow the distribution of former stations (orange) and cover additional regions with high population density.
• Figure 4: Confusion matrix of mode choice model on test set.
• Figure 5: Importance of feature in the XGBoost model. The distance is most important as expected, but also socio-demographic features play an important role.