Load-adaptive real-time energy management strategy for battery/ultracapacitor hybrid energy storage system using dynamic programming optimization

Energy management is crucial in battery/ultracapacitor hybrid energy storage systems in electric vehicles. Rule based control is one typical strategy in real-time management, but its adaptability in dynamic load is quite poor. This paper aims to develop a practical energy management strategy with ne...

Full description

Saved in:
Bibliographic Details
Published in:Journal of power sources Vol. 438; p. 227024
Main Authors: Liu, Chang, Wang, Yujie, Wang, Li, Chen, Zonghai
Format: Journal Article
Language:English
Published: Elsevier B.V 31.10.2019
Subjects:
Energy management strategy
Hybrid energy storage system
Load-adaptive
Optimal analysis
Rule based control
Hybrid energy storage system
Optimal analysis
Energy management strategy
Rule based control
Load-adaptive
ISSN:0378-7753, 1873-2755
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Energy management is crucial in battery/ultracapacitor hybrid energy storage systems in electric vehicles. Rule based control is one typical strategy in real-time management, but its adaptability in dynamic load is quite poor. This paper aims to develop a practical energy management strategy with near-optimal performance in both energy-saving and battery life extending. Firstly, dynamic programming (DP) analysis is used to find out the optimal control mode. Three-segment control rules are then extracted from the DP results. A functional relationship is established between the power splitting parameters and load statistics. Finally, a load-adaptive rule based control strategy is proposed based on that. Two composite load cycles are tested for verification. Results show that compared with the ordinary rule based control strategy, the proposed strategy has the stronger capability of battery protecting and energy-saving under unknown load patterns, where the battery Ah throughput and total energy loss are reduced by 3.4%–15.7% and 3.0%–15.1%, respectively. The results are quite close to DP results, showing that the proposed strategy can achieve near-optimal energy management in real time with low computational cost. [Display omitted] •A practical load-adaptive real-time energy management strategy is designed.•Control rules are extracted from the optimization results of 4 load cycles.•Real-time power splitting factors are decided by functions of load statistics.•Near optimal results are given under unknown cycles by the proposed strategy.
ISSN:0378-7753
1873-2755
DOI:10.1016/j.jpowsour.2019.227024