EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Synthesis of Real-World Driving Cycles with Road Grade: A Semi-Markov Route Model and Acceleration-Increment Markov Chain

Beomho Lee, Deokjin Kim, Janghyeok Won, Taeho Park, Dohyun Park

World Electric Vehicle Journal · 07.11.2026 · 17 (11) · 585

DOI: 10.3390/wevj17110585 ↗

Kurz gefasst

Fachartikel in der Zeitschrift World Electric Vehicle Journal, online erschienen am 07.11.2026. Verfasst von Beomho Lee und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Evaluation Institute of Industrial Technology, Korea Institute of Energy Technology Evaluation and Planning.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Energy assessment of electrified vehicles requires driving cycles that include altitude change, whereas certification cycles and most synthesis methods describe flat-road speed profiles only. This study synthesizes time–speed–altitude driving cycles from real-world data (145.7 h, 10,491 km). Road grade was obtained from a vehicle-mounted barometric sensor corrected for its speed-dependent pressure drop, and its low-frequency drift was replaced by a digital elevation model (DEM). A distance-based semi-Markov route model of road types (urban/rural, highway and mountain) was combined with a 1 s Markov chain of type-specific acceleration and grade transitions; sign symmetrization and a grade-trend state removed the net-ascent bias. For 4934 synthesized cycles, the route model reduced the root mean square error (RMSE) of the steep-grade share by speed bin from 9.2 to 2.5 percentage points, and sampling the next acceleration instead of the next speed reduced the deviation of the mean positive vehicle-specific power (VSP) from +27% to +3.4%. With terminal decelerations sampled from measured stop approaches, the recoverable energy exceeded that of real-world windows ending with the same kind of stop approach by 3.3% (distance-weighted), and neglecting grade underestimated it by 18%. In day-level bootstrap and leave-one-day-out cross-validation, the traction power and energy deviations stayed within the day-to-day variability. A flat output reproduced the mean positive VSP, positive specific energy, and energy consumption of zero-grade real-world driving within 3%. Finally, a representative cycle and speed-group × VSP-severity cycles were selected.

Förderung

  • Korea Evaluation Institute of Industrial Technology
  • Korea Institute of Energy Technology Evaluation and Planning

Zitieren

APA

Lee, B., Kim, D., Won, J., Park, T., & Park, D. (2026). Synthesis of Real-World Driving Cycles with Road Grade: A Semi-Markov Route Model and Acceleration-Increment Markov Chain. World Electric Vehicle Journal, 17(11), 585. https://doi.org/10.3390/wevj17110585

Vancouver

Lee B, Kim D, Won J, Park T, Park D. Synthesis of Real-World Driving Cycles with Road Grade: A Semi-Markov Route Model and Acceleration-Increment Markov Chain. World Electric Vehicle Journal. 2026;17(11):585. doi:10.3390/wevj17110585

BibTeX
@article{Lee2026,
  author = {Lee, Beomho and Kim, Deokjin and Won, Janghyeok and Park, Taeho and Park, Dohyun},
  title = {Synthesis of Real-World Driving Cycles with Road Grade: A Semi-Markov Route Model and Acceleration-Increment Markov Chain},
  journal = {World Electric Vehicle Journal},
  year = {2026},
  volume = {17},
  number = {11},
  pages = {585},
  doi = {10.3390/wevj17110585}
}
Kostenlos starten →