Title: Layered control strategies for hybrid electric vehicles based on optimal control

Authors: Domenico Bianchi, Luciano Rolando, Lorenzo Serrao, Simona Onori, Giorgio Rizzoni, Nazar Al-Khayat, Tung-Ming Hsieh, Pengju Kang

Addresses: Department of Electrical and Information Engineering, Center of Excellence DEWS, University of L'Aquila, Via Campo di Pile – Zona industriale di Pile, 67100 L'Aquila, Italy. ' Dipartimento di Energetica, Politecnico di Torino, C.so Duca degli Abruzzi, 24, 10129 Torino, Italy. ' Center for Automotive Research, The Ohio State University, 930 Kinnear Rd., Columbus, OH 43210, USA. ' Center for Automotive Research, The Ohio State University, 930 Kinnear Rd., Columbus, OH 43210, USA. ' Department of Mechanical and Aerospace Engineering; Center for Automotive Research, The Ohio State University, 930 Kinnear Rd., Columbus, OH 43210, USA. ' Cummins Inc., 500 Jackson Street, Columbus, IN 47201, USA. ' Cummins Inc., 500 Jackson Street, Columbus, IN 47201, USA. ' Cummins Inc., 500 Jackson Street, Columbus, IN 47201, USA

Abstract: Dynamic programming is known to provide the optimal solution to the energy management problem. However, it is not implementable online because it requires complete a-priori knowledge of the driving cycle and high computational requirements. This article presents a methodology to extract an implementable rule-based strategy from the dynamic programming results and thus build a near-optimal controller. The case study discussed in this paper focused on mode switching in a series/parallel hybrid vehicle, in which a clutch may be used to change the powertrain topology. Because of the complexity of the system, the controller is divided in two layers: the supervisory controller, which decides the powertrain configuration, and the energy management, which decides the power split. The process of deriving the rules from the optimal solution is described in detail. Then, the performance of the resulting rule-based strategy is studied and compared with the solution given by dynamic programming, which functions as a benchmark. Then another comparison is performed with respect to the equivalent consumption minimisation strategy (ECMS) which, if optimally tuned, can achieve optimal performance as close to DP as possible with the advantage of being implementable.

Keywords: hybrid electric vehicles; HEVs; energy management; rule-based control; dynamic programming; layered control strategy; rule-based control; heuristics; optimal control; supervisory control.

DOI: 10.1504/IJEHV.2011.042147

International Journal of Electric and Hybrid Vehicles, 2011 Vol.3 No.2, pp.191 - 217

Published online: 28 Aug 2011 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article