Back to Search Start Over

The "acceleration cliff": An investigation of the possible error source of the VSP distributions generated by Wiedemann car-following model.

Authors :
Lu, Hongyu
Song, Guohua
Yu, Lei
Source :
Transportation Research Part D: Transport & Environment. Dec2018, Vol. 65, p161-177. 17p.
Publication Year :
2018

Abstract

Highlights • Field and simulated car-following trajectories were compared in the Wiedemann Cartesian systems. • The existence of the regime structure led to the "acceleration cliffs". • The "acceleration cliffs" were found to be responsible for the unrealistic VSP distributions. • The "acceleration cliffs" could be eliminated by implementing continuous acceleration models. Abstract Errors have been found in the speed-specific vehicle specific power (VSP) distributions generated by the Wiedemann car-following model. The impact of the existence of the regime structure on the distributions was investigated in this paper. The analysis was conducted based on massive field car-following trajectories and numerical simulation of the Wiedemann model. The field and simulated speed-specific VSP distributions and acceleration distributions were compared. The field and simulated accelerations were analyzed with the help of the Wiedemann Cartesian systems. The unrealistic "acceleration cliffs" were found along with the regime borders. It was indicated that the "acceleration cliff" was a source of the generated emission errors. The existence of the discrete regimes was responsible for the "acceleration cliff." It was indicated that the "acceleration cliff" could not be eradicated by the parameter calibration, unless the logics of the regimes were modified. A data-driven model was designed based on the Wiedemann plane. The "acceleration cliffs" were successfully eliminated by implementing continuous accelerations. It was indicated that the Wiedemann model had the potential to generate realistic VSP distributions. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
13619209
Volume :
65
Database :
Academic Search Index
Journal :
Transportation Research Part D: Transport & Environment
Publication Type :
Academic Journal
Accession number :
133216343
Full Text :
https://doi.org/10.1016/j.trd.2018.08.004