Comparison of supervisory control strategies for series plug-in hybrid electric vehicle powertrains through dynamic programming

Rakesh M. Patil, Zoran Filipi, Hosam Kadry Fathy

Research output: Contribution to journalArticle

61 Citations (Scopus)

Abstract

This paper compares the optimal fuel and electricity costs associated with two supervisory control strategies for a series plug-in hybrid electric vehicle (PHEV) using dynamic programming. One strategy has no restrictions on engine fuel usage and the second is restricted to fuel usage only after the battery is depleted below a certain threshold. Both strategies are optimized using deterministic dynamic programming (DDP) to ensure a fair comparison. The DDP algorithm is implemented using a backward-looking powertrain model. Such an approach resolves the computational issues arising because of: 1) the interpolations required to obtain value function estimates and 2) the characterization of constraints through penalty functions. The primary conclusion is that there is no significant difference in the optimal performance of the two control strategies for the series PHEV except when gasoline is unreasonably cheap <1/gal. This result contrasts sharply with previous controller performance results for parallel and power-split PHEVs where the two strategies' performance is shown to differ for any gasoline price and driving distance. The reason for the contrast is the flexibility of engine operation in a series PHEV. The results are examined for different relative fuel and electricity prices and trip lengths.

Original languageEnglish (US)
Article number6516018
Pages (from-to)502-509
Number of pages8
JournalIEEE Transactions on Control Systems Technology
Volume22
Issue number2
DOIs
StatePublished - Mar 1 2014

Fingerprint

Plug-in hybrid vehicles
Powertrains
Dynamic programming
Gasoline
Electricity
Engines
Interpolation
Controllers
Costs

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

@article{2ea48f2620344b9eb0e771117883104a,
title = "Comparison of supervisory control strategies for series plug-in hybrid electric vehicle powertrains through dynamic programming",
abstract = "This paper compares the optimal fuel and electricity costs associated with two supervisory control strategies for a series plug-in hybrid electric vehicle (PHEV) using dynamic programming. One strategy has no restrictions on engine fuel usage and the second is restricted to fuel usage only after the battery is depleted below a certain threshold. Both strategies are optimized using deterministic dynamic programming (DDP) to ensure a fair comparison. The DDP algorithm is implemented using a backward-looking powertrain model. Such an approach resolves the computational issues arising because of: 1) the interpolations required to obtain value function estimates and 2) the characterization of constraints through penalty functions. The primary conclusion is that there is no significant difference in the optimal performance of the two control strategies for the series PHEV except when gasoline is unreasonably cheap <1/gal. This result contrasts sharply with previous controller performance results for parallel and power-split PHEVs where the two strategies' performance is shown to differ for any gasoline price and driving distance. The reason for the contrast is the flexibility of engine operation in a series PHEV. The results are examined for different relative fuel and electricity prices and trip lengths.",
author = "Patil, {Rakesh M.} and Zoran Filipi and Fathy, {Hosam Kadry}",
year = "2014",
month = "3",
day = "1",
doi = "10.1109/TCST.2013.2257778",
language = "English (US)",
volume = "22",
pages = "502--509",
journal = "IEEE Transactions on Control Systems Technology",
issn = "1063-6536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Comparison of supervisory control strategies for series plug-in hybrid electric vehicle powertrains through dynamic programming. / Patil, Rakesh M.; Filipi, Zoran; Fathy, Hosam Kadry.

In: IEEE Transactions on Control Systems Technology, Vol. 22, No. 2, 6516018, 01.03.2014, p. 502-509.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparison of supervisory control strategies for series plug-in hybrid electric vehicle powertrains through dynamic programming

AU - Patil, Rakesh M.

AU - Filipi, Zoran

AU - Fathy, Hosam Kadry

PY - 2014/3/1

Y1 - 2014/3/1

N2 - This paper compares the optimal fuel and electricity costs associated with two supervisory control strategies for a series plug-in hybrid electric vehicle (PHEV) using dynamic programming. One strategy has no restrictions on engine fuel usage and the second is restricted to fuel usage only after the battery is depleted below a certain threshold. Both strategies are optimized using deterministic dynamic programming (DDP) to ensure a fair comparison. The DDP algorithm is implemented using a backward-looking powertrain model. Such an approach resolves the computational issues arising because of: 1) the interpolations required to obtain value function estimates and 2) the characterization of constraints through penalty functions. The primary conclusion is that there is no significant difference in the optimal performance of the two control strategies for the series PHEV except when gasoline is unreasonably cheap <1/gal. This result contrasts sharply with previous controller performance results for parallel and power-split PHEVs where the two strategies' performance is shown to differ for any gasoline price and driving distance. The reason for the contrast is the flexibility of engine operation in a series PHEV. The results are examined for different relative fuel and electricity prices and trip lengths.

AB - This paper compares the optimal fuel and electricity costs associated with two supervisory control strategies for a series plug-in hybrid electric vehicle (PHEV) using dynamic programming. One strategy has no restrictions on engine fuel usage and the second is restricted to fuel usage only after the battery is depleted below a certain threshold. Both strategies are optimized using deterministic dynamic programming (DDP) to ensure a fair comparison. The DDP algorithm is implemented using a backward-looking powertrain model. Such an approach resolves the computational issues arising because of: 1) the interpolations required to obtain value function estimates and 2) the characterization of constraints through penalty functions. The primary conclusion is that there is no significant difference in the optimal performance of the two control strategies for the series PHEV except when gasoline is unreasonably cheap <1/gal. This result contrasts sharply with previous controller performance results for parallel and power-split PHEVs where the two strategies' performance is shown to differ for any gasoline price and driving distance. The reason for the contrast is the flexibility of engine operation in a series PHEV. The results are examined for different relative fuel and electricity prices and trip lengths.

UR - http://www.scopus.com/inward/record.url?scp=84897648886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897648886&partnerID=8YFLogxK

U2 - 10.1109/TCST.2013.2257778

DO - 10.1109/TCST.2013.2257778

M3 - Article

AN - SCOPUS:84897648886

VL - 22

SP - 502

EP - 509

JO - IEEE Transactions on Control Systems Technology

JF - IEEE Transactions on Control Systems Technology

SN - 1063-6536

IS - 2

M1 - 6516018

ER -