Reinforced explorit on optimizing vehicle powertrains

Victor Parque Tenorio, Masakazu Kobayashi, Masatake Higashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

How to build optimal vehicular powertrains? We study this question and propose an algorithm inspired by a domain-general design process. The basic idea is to interplay co-biasingly between the local approximations of discrete design and the global refinements of continuous parameters. The proposed method was evaluated to design powertrains of four types of vehicles: Series Hybrid Electric Vehicle(SHEV), Parallel Hybrid Electric Vehicle(PHEV), Fuel Cell(FC) and Electric Vehicle(EV). Simulation results show noticeable improvements on mileage per gas emissions over different study cases. To our knowledge, this is the first study aiming at designing vehicle powertrains considering the holistic point of view.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages579-586
Number of pages8
Volume8227 LNCS
EditionPART 2
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event20th International Conference on Neural Information Processing, ICONIP 2013 - Daegu, Korea, Republic of
Duration: 2013 Nov 32013 Nov 7

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume8227 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other20th International Conference on Neural Information Processing, ICONIP 2013
CountryKorea, Republic of
CityDaegu
Period13/11/313/11/7

Fingerprint

Hybrid Electric Vehicle
Powertrains
Hybrid vehicles
Electric Vehicle
Local Approximation
Fuel Cell
Design Process
Refinement
Electric vehicles
Gas emissions
Series
Fuel cells
Simulation
Design
Knowledge
Gas

Keywords

  • Electric vehicle
  • Explorit
  • Fuel cell
  • Parallel hybrid electricvehicle
  • Reinforcement learning
  • Series hybrid electric vehicle

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Parque Tenorio, V., Kobayashi, M., & Higashi, M. (2013). Reinforced explorit on optimizing vehicle powertrains. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 2 ed., Vol. 8227 LNCS, pp. 579-586). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8227 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-42042-9_72

Reinforced explorit on optimizing vehicle powertrains. / Parque Tenorio, Victor; Kobayashi, Masakazu; Higashi, Masatake.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8227 LNCS PART 2. ed. 2013. p. 579-586 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8227 LNCS, No. PART 2).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Parque Tenorio, V, Kobayashi, M & Higashi, M 2013, Reinforced explorit on optimizing vehicle powertrains. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 edn, vol. 8227 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 8227 LNCS, pp. 579-586, 20th International Conference on Neural Information Processing, ICONIP 2013, Daegu, Korea, Republic of, 13/11/3. https://doi.org/10.1007/978-3-642-42042-9_72
Parque Tenorio V, Kobayashi M, Higashi M. Reinforced explorit on optimizing vehicle powertrains. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 2 ed. Vol. 8227 LNCS. 2013. p. 579-586. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). https://doi.org/10.1007/978-3-642-42042-9_72
Parque Tenorio, Victor ; Kobayashi, Masakazu ; Higashi, Masatake. / Reinforced explorit on optimizing vehicle powertrains. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8227 LNCS PART 2. ed. 2013. pp. 579-586 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).
@inproceedings{30a9bb11264348c981856e2de948171c,
title = "Reinforced explorit on optimizing vehicle powertrains",
abstract = "How to build optimal vehicular powertrains? We study this question and propose an algorithm inspired by a domain-general design process. The basic idea is to interplay co-biasingly between the local approximations of discrete design and the global refinements of continuous parameters. The proposed method was evaluated to design powertrains of four types of vehicles: Series Hybrid Electric Vehicle(SHEV), Parallel Hybrid Electric Vehicle(PHEV), Fuel Cell(FC) and Electric Vehicle(EV). Simulation results show noticeable improvements on mileage per gas emissions over different study cases. To our knowledge, this is the first study aiming at designing vehicle powertrains considering the holistic point of view.",
keywords = "Electric vehicle, Explorit, Fuel cell, Parallel hybrid electricvehicle, Reinforcement learning, Series hybrid electric vehicle",
author = "{Parque Tenorio}, Victor and Masakazu Kobayashi and Masatake Higashi",
year = "2013",
doi = "10.1007/978-3-642-42042-9_72",
language = "English",
isbn = "9783642420412",
volume = "8227 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 2",
pages = "579--586",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 2",

}

TY - GEN

T1 - Reinforced explorit on optimizing vehicle powertrains

AU - Parque Tenorio, Victor

AU - Kobayashi, Masakazu

AU - Higashi, Masatake

PY - 2013

Y1 - 2013

N2 - How to build optimal vehicular powertrains? We study this question and propose an algorithm inspired by a domain-general design process. The basic idea is to interplay co-biasingly between the local approximations of discrete design and the global refinements of continuous parameters. The proposed method was evaluated to design powertrains of four types of vehicles: Series Hybrid Electric Vehicle(SHEV), Parallel Hybrid Electric Vehicle(PHEV), Fuel Cell(FC) and Electric Vehicle(EV). Simulation results show noticeable improvements on mileage per gas emissions over different study cases. To our knowledge, this is the first study aiming at designing vehicle powertrains considering the holistic point of view.

AB - How to build optimal vehicular powertrains? We study this question and propose an algorithm inspired by a domain-general design process. The basic idea is to interplay co-biasingly between the local approximations of discrete design and the global refinements of continuous parameters. The proposed method was evaluated to design powertrains of four types of vehicles: Series Hybrid Electric Vehicle(SHEV), Parallel Hybrid Electric Vehicle(PHEV), Fuel Cell(FC) and Electric Vehicle(EV). Simulation results show noticeable improvements on mileage per gas emissions over different study cases. To our knowledge, this is the first study aiming at designing vehicle powertrains considering the holistic point of view.

KW - Electric vehicle

KW - Explorit

KW - Fuel cell

KW - Parallel hybrid electricvehicle

KW - Reinforcement learning

KW - Series hybrid electric vehicle

UR - http://www.scopus.com/inward/record.url?scp=84893388591&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84893388591&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-42042-9_72

DO - 10.1007/978-3-642-42042-9_72

M3 - Conference contribution

AN - SCOPUS:84893388591

SN - 9783642420412

VL - 8227 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 579

EP - 586

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -