Sizing and optimization process of hybrid electric propulsion system for heavy-duty vehicle based on Gaussian process modeling considering traction motor characteristics
- Title
- Sizing and optimization process of hybrid electric propulsion system for heavy-duty vehicle based on Gaussian process modeling considering traction motor characteristics
- Author
- 임명섭
- Keywords
- Component sizing; Equivalent circuit; Fuel economy optimization; Gaussian process modeling (GPM); Hybrid electric propulsion system
- Issue Date
- 2022-06
- Publisher
- Elsevier Ltd
- Citation
- Renewable and Sustainable Energy Reviews, v. 161, article no. 112286, Page. 1-12
- Abstract
- This paper suggests a sizing and optimization scheme of a hybrid electric propulsion system for a heavy-duty vehicle. The considered propulsion system consists of four in-wheel traction motors with a planetary gear, and the power source is configured with a battery and engine–generator set. To optimize fuel economy by powertrain sizing, the vehicle design process and vehicle simulation were constructed. Optimization was then performed using Gaussian process modeling (GPM). During the optimization, the variation of the gross weight of the propulsion system was considered. In addition, the change in the efficiency map of the traction motor was precisely reflected. The sampling points for GPM were determined from the Optimal Latin hypercube design. Subsequently, the fuel economy surrogate model was generated via the GPM. Optimization was then performed using the steepest gradient descent algorithm. Finally, the maximized fuel economy model was verified using a vehicle simulation.
- URI
- https://www.sciencedirect.com/science/article/pii/S1364032122002052?via%3Dihubhttps://repository.hanyang.ac.kr/handle/20.500.11754/177970
- ISSN
- 1364-0321;1879-0690
- DOI
- 10.1016/j.rser.2022.112286
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > AUTOMOTIVE ENGINEERING(미래자동차공학과) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML