Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine
To address the problem that a single energy management strategy cannot adapt to complex driving conditions, in this paper, a real-time energy management strategy for different driving conditions is proposed to improve fuel economy. First, in order to improve the accuracy and stability of the driving...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-06-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/15/12/4353 |
_version_ | 1797487755601641472 |
---|---|
author | Penghui Qiang Peng Wu Tao Pan Huaiquan Zang |
author_facet | Penghui Qiang Peng Wu Tao Pan Huaiquan Zang |
author_sort | Penghui Qiang |
collection | DOAJ |
description | To address the problem that a single energy management strategy cannot adapt to complex driving conditions, in this paper, a real-time energy management strategy for different driving conditions is proposed to improve fuel economy. First, in order to improve the accuracy and stability of the driving condition identifier, a feature fusion extreme learning machine (FFELM) is used for identification. Secondly, equivalent consumption minimization strategy (ECMS) offline optimization is conducted for different types of driving cycles, and the effect of driving cycle type and driving distance on the energy management strategy under the optimization result is analyzed. A real-time energy management strategy combining driving cycle type, driving distance, and optimal power allocation factor is proposed. To demonstrate the effectiveness of the proposed strategy, combined driving cycles were used for testing. The simulation results show that the proposed strategy can improve the equivalent fuel consumption by 10.21% compared to the conventional strategy CD-CS. The equivalent fuel economy can be improved by 2.5% compared to the single ECMS strategy with the less computational burden. Thus, it is demonstrated that the proposed strategy can be effectively adapted to different driving conditions and shows better real-time and economic performance. |
first_indexed | 2024-03-09T23:53:15Z |
format | Article |
id | doaj.art-1eb3672fa40c44b483e0f075fa412fab |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-03-09T23:53:15Z |
publishDate | 2022-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-1eb3672fa40c44b483e0f075fa412fab2023-11-23T16:29:46ZengMDPI AGEnergies1996-10732022-06-011512435310.3390/en15124353Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning MachinePenghui Qiang0Peng Wu1Tao Pan2Huaiquan Zang3Institute of Electrical Engineering, Yanshan University, Qinhuangdao 066004, ChinaInstitute of Electrical Engineering, Yanshan University, Qinhuangdao 066004, ChinaInstitute of Electrical Engineering, Yanshan University, Qinhuangdao 066004, ChinaInstitute of Electrical Engineering, Yanshan University, Qinhuangdao 066004, ChinaTo address the problem that a single energy management strategy cannot adapt to complex driving conditions, in this paper, a real-time energy management strategy for different driving conditions is proposed to improve fuel economy. First, in order to improve the accuracy and stability of the driving condition identifier, a feature fusion extreme learning machine (FFELM) is used for identification. Secondly, equivalent consumption minimization strategy (ECMS) offline optimization is conducted for different types of driving cycles, and the effect of driving cycle type and driving distance on the energy management strategy under the optimization result is analyzed. A real-time energy management strategy combining driving cycle type, driving distance, and optimal power allocation factor is proposed. To demonstrate the effectiveness of the proposed strategy, combined driving cycles were used for testing. The simulation results show that the proposed strategy can improve the equivalent fuel consumption by 10.21% compared to the conventional strategy CD-CS. The equivalent fuel economy can be improved by 2.5% compared to the single ECMS strategy with the less computational burden. Thus, it is demonstrated that the proposed strategy can be effectively adapted to different driving conditions and shows better real-time and economic performance.https://www.mdpi.com/1996-1073/15/12/4353energy management strategyfeature fusion extreme learning machinedriving condition identifierreal-time |
spellingShingle | Penghui Qiang Peng Wu Tao Pan Huaiquan Zang Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine Energies energy management strategy feature fusion extreme learning machine driving condition identifier real-time |
title | Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine |
title_full | Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine |
title_fullStr | Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine |
title_full_unstemmed | Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine |
title_short | Real-Time Energy Management Strategy Based on Driving Conditions Using a Feature Fusion Extreme Learning Machine |
title_sort | real time energy management strategy based on driving conditions using a feature fusion extreme learning machine |
topic | energy management strategy feature fusion extreme learning machine driving condition identifier real-time |
url | https://www.mdpi.com/1996-1073/15/12/4353 |
work_keys_str_mv | AT penghuiqiang realtimeenergymanagementstrategybasedondrivingconditionsusingafeaturefusionextremelearningmachine AT pengwu realtimeenergymanagementstrategybasedondrivingconditionsusingafeaturefusionextremelearningmachine AT taopan realtimeenergymanagementstrategybasedondrivingconditionsusingafeaturefusionextremelearningmachine AT huaiquanzang realtimeenergymanagementstrategybasedondrivingconditionsusingafeaturefusionextremelearningmachine |