Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections
To tackle the energy-saving optimization issue of plug-in hybrid electric trucks traversing multiple traffic light intersections continuously, this paper presents a double-layer energy management strategy that utilizes the dynamic programming–twin delayed deep deterministic policy gradient (DP-TD3)...
Saved in:
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-11-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/17/23/6022 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1846124315677818880 |
---|---|
author | Xin Liu Guojing Shi Changbo Yang Enyong Xu Yanmei Meng |
author_facet | Xin Liu Guojing Shi Changbo Yang Enyong Xu Yanmei Meng |
author_sort | Xin Liu |
collection | DOAJ |
description | To tackle the energy-saving optimization issue of plug-in hybrid electric trucks traversing multiple traffic light intersections continuously, this paper presents a double-layer energy management strategy that utilizes the dynamic programming–twin delayed deep deterministic policy gradient (DP-TD3) algorithm to synergistically optimize the speed planning and energy management of plug-in hybrid electric trucks, thereby enhancing the vehicle’s passability through traffic light intersections and fuel economy. In the upper layer, the dynamic programming (DP) algorithm is employed to create a speed-planning model. This model effectively converts the nonlinear constraints related to the position, phase, and timing information of each traffic signal on the road into time-varying constraints, thereby improving computational efficiency. In the lower layer, an energy management model is constructed using the twin delayed deep deterministic policy gradient (TD3) algorithm to achieve optimal allocation of demanded power through the interaction of the TD3 agent with the truck environment. The model’s validity is confirmed through testing on a hardware-in-the-loop test machine, followed by simulation experiments. The results demonstrate that the DP-TD3 method proposed in this paper effectively enhances fuel economy, achieving an average fuel saving of 14.61% compared to the dynamic programming–charge depletion/charge sustenance (DP-CD/CS) method. |
format | Article |
id | doaj-art-c1485d0f71a241e8b4581b6d6d9be9bb |
institution | Kabale University |
issn | 1996-1073 |
language | English |
publishDate | 2024-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj-art-c1485d0f71a241e8b4581b6d6d9be9bb2024-12-13T16:25:47ZengMDPI AGEnergies1996-10732024-11-011723602210.3390/en17236022Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light IntersectionsXin Liu0Guojing Shi1Changbo Yang2Enyong Xu3Yanmei Meng4School of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaDongfeng Liuzhou Motor Co., Ltd., Liuzhou 545005, ChinaDongfeng Liuzhou Motor Co., Ltd., Liuzhou 545005, ChinaSchool of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaTo tackle the energy-saving optimization issue of plug-in hybrid electric trucks traversing multiple traffic light intersections continuously, this paper presents a double-layer energy management strategy that utilizes the dynamic programming–twin delayed deep deterministic policy gradient (DP-TD3) algorithm to synergistically optimize the speed planning and energy management of plug-in hybrid electric trucks, thereby enhancing the vehicle’s passability through traffic light intersections and fuel economy. In the upper layer, the dynamic programming (DP) algorithm is employed to create a speed-planning model. This model effectively converts the nonlinear constraints related to the position, phase, and timing information of each traffic signal on the road into time-varying constraints, thereby improving computational efficiency. In the lower layer, an energy management model is constructed using the twin delayed deep deterministic policy gradient (TD3) algorithm to achieve optimal allocation of demanded power through the interaction of the TD3 agent with the truck environment. The model’s validity is confirmed through testing on a hardware-in-the-loop test machine, followed by simulation experiments. The results demonstrate that the DP-TD3 method proposed in this paper effectively enhances fuel economy, achieving an average fuel saving of 14.61% compared to the dynamic programming–charge depletion/charge sustenance (DP-CD/CS) method.https://www.mdpi.com/1996-1073/17/23/6022plug-in hybrid electric truckeco-drivingenergy management strategytraffic light |
spellingShingle | Xin Liu Guojing Shi Changbo Yang Enyong Xu Yanmei Meng Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections Energies plug-in hybrid electric truck eco-driving energy management strategy traffic light |
title | Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections |
title_full | Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections |
title_fullStr | Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections |
title_full_unstemmed | Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections |
title_short | Co-Optimization of Speed Planning and Energy Management for Plug-In Hybrid Electric Trucks Passing Through Traffic Light Intersections |
title_sort | co optimization of speed planning and energy management for plug in hybrid electric trucks passing through traffic light intersections |
topic | plug-in hybrid electric truck eco-driving energy management strategy traffic light |
url | https://www.mdpi.com/1996-1073/17/23/6022 |
work_keys_str_mv | AT xinliu cooptimizationofspeedplanningandenergymanagementforpluginhybridelectrictruckspassingthroughtrafficlightintersections AT guojingshi cooptimizationofspeedplanningandenergymanagementforpluginhybridelectrictruckspassingthroughtrafficlightintersections AT changboyang cooptimizationofspeedplanningandenergymanagementforpluginhybridelectrictruckspassingthroughtrafficlightintersections AT enyongxu cooptimizationofspeedplanningandenergymanagementforpluginhybridelectrictruckspassingthroughtrafficlightintersections AT yanmeimeng cooptimizationofspeedplanningandenergymanagementforpluginhybridelectrictruckspassingthroughtrafficlightintersections |