0
Research Papers

Real-World Robustness for Hybrid Vehicle Optimal Energy Management Strategies Incorporating Drivability Metrics

[+] Author and Article Information
Daniel F. Opila

Department of Mechanical Engineering,
Carnegie Mellon University,
Pittsburgh, PA 15213
e-mail: dopila@andrew.cmu.edu

Xiaoyong Wang

Research and Advanced Engineering,
Ford Motor Company,
Dearborn, MI 48120
e-mail: xwang67@ford.com

Ryan McGee

Research and Advanced Engineering,
Ford Motor Company,
Dearborn, MI 48120
e-mail: rmcgee3@ford.com

R. Brent Gillespie

Department of Mechanical Engineering,
University of Michigan,
Ann Arbor, MI 48109
e-mail: brentg@umich.edu

Jeffrey A. Cook

Electrical Engineering and
Computer Science Department,
University of Michigan,
Ann Arbor, MI 48109
e-mail: jeffcook@umich.edu

J. W. Grizzle

Electrical Engineering and
Computer Science Department,
Department of Mechanical Engineering,
University of Michigan,
Ann Arbor, MI 48109
e-mail: grizzle@umich.edu

As explained in Ref. [6], two additional states are required to represent the stochastic drive cycle and two more to track drivability metrics.

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT, AND CONTROL. Manuscript received July 12, 2013; final manuscript received May 3, 2014; published online August 8, 2014. Assoc. Editor: Shankar Coimbatore Subramanian.

J. Dyn. Sys., Meas., Control 136(6), 061011 (Aug 08, 2014) (10 pages) Paper No: DS-13-1271; doi: 10.1115/1.4027680 History: Received July 12, 2013; Revised May 03, 2014

Hybrid vehicle fuel economy and drive quality are coupled through the “energy management” controller that regulates power flow among the various energy sources and sinks. This paper studies energy management controllers designed using shortest path stochastic dynamic programming (SP-SDP), a stochastic optimal control design method which can respect constraints on drivetrain activity while minimizing fuel consumption for an assumed distribution of driver power demand. The performance of SP-SDP controllers is evaluated through simulation on large numbers of real-world drive cycles and compared to a baseline industrial controller provided by a major auto manufacturer. On real-world driving data, the SP-SDP-based controllers yield 10% better fuel economy than the baseline industrial controller, for the same engine and gear activity. The SP-SDP controllers are further evaluated for robustness to the drive cycle statistics used in their design. Simplified drivability metrics introduced in previous work are validated on large real-world data sets.

FIGURES IN THIS ARTICLE
<>
Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Vehicle configuration

Grahic Jump Location
Fig. 2

High level baseline controller architecture

Grahic Jump Location
Fig. 3

Statistics of real-world driving. Figure 3(a) is the cumulative distribution function of trip distance for the source data and two subsets. Mean distances for the sets are: Source data set–11.7 mi., Ensemble 1–12.7 mi., Ensemble 2–9.9 mi. Figure 3(b) is the cumulative distribution function of vehicle velocity for the source data, two subsets, and two government test cycles.

Grahic Jump Location
Fig. 4

Statistical fuel economy. The SP-SDP controller family designed on Ensemble 1 is simulated on Ensemble 1 and Ensemble 2, and each cycle is corrected for SOC. The mean, standard deviation, and 10th and 90th percentile are calculated. The mean, standard deviation (error bars), and 10th and 90th percentile are also calculated for the baseline controller.

Grahic Jump Location
Fig. 5

Fuel economy and drivability metrics on the FTP and NEDC cycles for five controller options. Controller families are designed with statistics from FTP, NEDC, Ensemble 1, and Ensemble 2. All fuel economy figures are normalized to the baseline controller performance on FTP, shown as a large green dot in Fig. 5(a).

Grahic Jump Location
Fig. 6

Fuel economy on Ensemble 1 for controllers designed using statistics from FTP, NEDC, Ensemble 1, and Ensemble 2

Grahic Jump Location
Fig. 7

Drivability metric reduction. The seven complex engine and transmission metrics are divided into two categories, mean dwell times and short-duration dwell times. These metrics are then reduced to the two simplified metrics.

Grahic Jump Location
Fig. 8

Comparison of simple and complex engine activity metrics on the FTP cycle. Engine on durations less than some number of seconds are compared to engine events. Data are shown for cutoffs of 3, 5, 10, and 30 s. SP-SDP controllers are shown with a linear least squares fit. The baseline controller is shown as a large green circle for the same four cutoff criteria.

Grahic Jump Location
Fig. 9

Comparison of simple and complex gear shift metrics on the FTP cycle. Gear dwell times less or equal to 1 s between shifts or clutch disengagements are compared to gear events. The SP-SDP controllers are shown with a linear least squares fit and the baseline controller is shown as a large green circle.

Grahic Jump Location
Fig. 10

Comparison of simple and detailed drivability metrics for the Ensemble 1 cycles. Detailed engine activity metrics are compared to the simplified engine events metric. The 100 cycles are treated as a single trip. Each marker represents the same drive cycle, the concatenated Ensemble 1.

Grahic Jump Location
Fig. 11

Fuel economy on Ensemble 1 treated as a concatenated cycle for controllers designed using statistics from FTP, NEDC, Ensemble 1, and Ensemble 2

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In