GPS Data-Based Plug-in Hybrid Electric Vehicle Simulation
|Faculty/Professorship:||Information Systems and Energy Efficient Systems ; Fakultät Wirtschaftsinformatik und Angewandte Informatik: Abschlussarbeiten|
|Publisher Information:||Bamberg : Otto-Friedrich-Universität|
|Year of publication:||2019|
|Pages:||XXXI, 124 ; Illustrationen, Diagramme|
Dissertation, Otto-Friedrich-Universität Bamberg, 2019
|Licence:||Creative Commons - CC BY-NC-ND - Attribution - NonCommercial - NoDerivatives 4.0 International|
The automotive sector, while being an example of a highly innovative industry driven by strong competitive pressure and constant technological progress, has never had to deal with truly disruptive changes regarding its products, processes, or value network structure. In this regard, the rise of electric mobility constitutes an unprecedented market change as it implies an extensive redefinition of the product architecture of cars, not only involving new technologies but also new market entrants from highly innovative industries, the anticipation of new business models, and a dependency on the electrical grid as an additional, essential infrastructure component.
In this context, decisions regarding both the capacity of batteries and the charging network play a major role as they determine the electric range of the vehicles as well as overall system costs. At the same time, the transition from combustion-based transportation to electric transportation has a considerable impact on the power grid that also depends on the trade-off between battery capacities and the density and power ratings of chargers.
In order to assess such important aspects as electric reachability, grid impact, and battery versus infrastructure trade-offs, the mobility behavior of individuals plays an essential role. Literature suggests that GPS driving data analysis constitutes a means of choice to assess the impact of battery capacities and charging opportunities on electric range and on power grid demand. Still, a great share of publications does either use synthetic mobility profiles (“driving cycles”) or self-reported data and thus does not utilize the wealth of information that is available in actual movement data. Moreover, literature research indicates that prior work that considers the entirety of car drivers as a coherent whole without describing different types of drivers in greater detail, rarely takes high electric range and variations in the availability of both private and public charging infrastructure facilities into account. Thus, such studies focus on average effects, which reduces the precision and utility of their assessments.
In this work, the high granularity of real-world GPS time series from 1,000 conventional vehicles is utilized to reflect the natural mobility behavior of drivers and to compare meaningful driver segments. Potential charging locations are automatically identified, and the electric energy consumption and charging behavior of plug-in hybrid electric vehicles is closely approximated. This enables the identification of appropriate vehicle and infrastructure parameters for electric mobility target groups and the assessment of their impact in terms of the electrification of mileage and energy demand. The consideration of household level solar systems and of a load shifting method as parts of a possible future charging infrastructure complements the work.
Results suggest that large but realistic battery capacities have the potential to dissipate concerns about the need for an all-encompassing charging infrastructure. Dense charge points are only needed for vehicles with short electric range or for small groups of fast long-range drivers. Both solar charging and load shifting considerably help alleviate stress on the power grid.
Decision makers may use the results and the methodology underlying this work to identify vehicle and infrastructure requirements of distinct segments and to estimate the grid impact of vehicle charging. Consequently, insights about benefits and obstacles of electric mobility adoption may facilitate better decisions in both vehicle development and infrastructure planning.
|GND Keywords:||Elektrofahrzeug ; Benutzerverhalten ; Datenanalyse ; GPS|
|DDC Classification:||004 Computer science |
300 Social sciences, sociology & anthropology
|RVK Classification:||ST 620|
|Release Date:||12. December 2019|
originated at the
University of Bamberg
University of Bamberg