Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes

In this paper, a novel control algorithm is proposed using Model Predictive Control for improving energy consumption of fully electric vehicles (FEVs) and Deep Reinforcement Learning (DRL) for understanding the driving environment in real-time, respectively. The dynamics of the FEVs and the brushles...

Full description

Bibliographic Details
Main Author: Kiwon Yeom
Format: Article
Language:English
Published: Elsevier 2022-11-01
Series:Energy Reports
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352484722022028
_version_ 1797952177314988032
author Kiwon Yeom
author_facet Kiwon Yeom
author_sort Kiwon Yeom
collection DOAJ
description In this paper, a novel control algorithm is proposed using Model Predictive Control for improving energy consumption of fully electric vehicles (FEVs) and Deep Reinforcement Learning (DRL) for understanding the driving environment in real-time, respectively. The dynamics of the FEVs and the brushless DC motor (BLDC) based powertrain system is utilized for the DRL process, and a Model Predictive Control (MPC) based optimal speed strategy for minimizing battery power consumption is developed by updating the approximation model through the repeated experience. Using the high fidelity car simulator (CarSim), the proposed algorithm is evaluated under freeway cruising scenarios with road slopes as well as preceding car following. The simulation results demonstrate in overall potential energy saving of 3.2%.
first_indexed 2024-04-10T22:42:14Z
format Article
id doaj.art-de9a28635f284ca2a3664b118bb430d4
institution Directory Open Access Journal
issn 2352-4847
language English
last_indexed 2024-04-10T22:42:14Z
publishDate 2022-11-01
publisher Elsevier
record_format Article
series Energy Reports
spelling doaj.art-de9a28635f284ca2a3664b118bb430d42023-01-16T04:08:42ZengElsevierEnergy Reports2352-48472022-11-018599604Learning model predictive control for efficient energy management of electric vehicles under car following and road slopesKiwon Yeom0Department of Human Intelligence and Robot Engineering, Sangmyung Unviersity, Cheonan-si, South KoreaIn this paper, a novel control algorithm is proposed using Model Predictive Control for improving energy consumption of fully electric vehicles (FEVs) and Deep Reinforcement Learning (DRL) for understanding the driving environment in real-time, respectively. The dynamics of the FEVs and the brushless DC motor (BLDC) based powertrain system is utilized for the DRL process, and a Model Predictive Control (MPC) based optimal speed strategy for minimizing battery power consumption is developed by updating the approximation model through the repeated experience. Using the high fidelity car simulator (CarSim), the proposed algorithm is evaluated under freeway cruising scenarios with road slopes as well as preceding car following. The simulation results demonstrate in overall potential energy saving of 3.2%.http://www.sciencedirect.com/science/article/pii/S2352484722022028Eco-drivingElectric vehiclesModel Predictive ControlDeep Reinforcement LearningOptimal speed controlEnergy consumption
spellingShingle Kiwon Yeom
Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
Energy Reports
Eco-driving
Electric vehicles
Model Predictive Control
Deep Reinforcement Learning
Optimal speed control
Energy consumption
title Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
title_full Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
title_fullStr Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
title_full_unstemmed Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
title_short Learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
title_sort learning model predictive control for efficient energy management of electric vehicles under car following and road slopes
topic Eco-driving
Electric vehicles
Model Predictive Control
Deep Reinforcement Learning
Optimal speed control
Energy consumption
url http://www.sciencedirect.com/science/article/pii/S2352484722022028
work_keys_str_mv AT kiwonyeom learningmodelpredictivecontrolforefficientenergymanagementofelectricvehiclesundercarfollowingandroadslopes