Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning

This paper deals with the HEV real-time energy management problem using deep reinforcement learning with connected technologies such as Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I). In the HEV energy management problem, it is important to run the engine efficiently in order to minimi...

Full description

Bibliographic Details
Main Authors:	Shota Inuzuka, Bo Zhang, Tielong Shen
Format:	Article
Language:	English
Published:	MDPI AG 2021-08-01
Series:	Energies
Subjects:	HEV energy management connected technology deep reinforcement learning
Online Access:	https://www.mdpi.com/1996-1073/14/17/5270

id	doaj-fc5eacaa6cf04fc59cc82be8403c1c3c
record_format	Article
spelling	doaj-fc5eacaa6cf04fc59cc82be8403c1c3c2021-09-09T13:42:44ZengMDPI AGEnergies1996-10732021-08-01145270527010.3390/en14175270Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement LearningShota Inuzuka0Bo Zhang1Tielong Shen2Faculty of Science and Technology, Sophia University, Tokyo 102-8554, JapanFaculty of Science and Technology, Sophia University, Tokyo 102-8554, JapanFaculty of Science and Technology, Sophia University, Tokyo 102-8554, JapanThis paper deals with the HEV real-time energy management problem using deep reinforcement learning with connected technologies such as Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I). In the HEV energy management problem, it is important to run the engine efficiently in order to minimize its total energy cost. This research proposes a policy model that takes into account road congestion and aims to learn the optimal system mode selection and power distribution when considering the far future by policy-based reinforcement learning. In the simulation, a traffic environment is generated in a virtual space by IPG CarMaker and a HEV model is prepared in MATLAB/Simulink to calculate the energy cost while driving on the road environment. The simulation validation shows the versatility of the proposed method for the test data, and in addition, it shows that considering road congestion reduces the total cost and improves the learning speed. Furthermore, we compare the proposed method with model predictive control (MPC) under the same conditions and show that the proposed method obtains more global optimal solutions.https://www.mdpi.com/1996-1073/14/17/5270HEV energy managementconnected technologydeep reinforcement learning
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Shota Inuzuka Bo Zhang Tielong Shen
spellingShingle	Shota Inuzuka Bo Zhang Tielong Shen Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning Energies HEV energy management connected technology deep reinforcement learning
author_facet	Shota Inuzuka Bo Zhang Tielong Shen
author_sort	Shota Inuzuka
title	Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning
title_short	Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning
title_full	Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning
title_fullStr	Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning
title_full_unstemmed	Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning
title_sort	real-time hev energy management strategy considering road congestion based on deep reinforcement learning
publisher	MDPI AG
series	Energies
issn	1996-1073
publishDate	2021-08-01
description	This paper deals with the HEV real-time energy management problem using deep reinforcement learning with connected technologies such as Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I). In the HEV energy management problem, it is important to run the engine efficiently in order to minimize its total energy cost. This research proposes a policy model that takes into account road congestion and aims to learn the optimal system mode selection and power distribution when considering the far future by policy-based reinforcement learning. In the simulation, a traffic environment is generated in a virtual space by IPG CarMaker and a HEV model is prepared in MATLAB/Simulink to calculate the energy cost while driving on the road environment. The simulation validation shows the versatility of the proposed method for the test data, and in addition, it shows that considering road congestion reduces the total cost and improves the learning speed. Furthermore, we compare the proposed method with model predictive control (MPC) under the same conditions and show that the proposed method obtains more global optimal solutions.
topic	HEV energy management connected technology deep reinforcement learning
url	https://www.mdpi.com/1996-1073/14/17/5270
work_keys_str_mv	AT shotainuzuka realtimehevenergymanagementstrategyconsideringroadcongestionbasedondeepreinforcementlearning AT bozhang realtimehevenergymanagementstrategyconsideringroadcongestionbasedondeepreinforcementlearning AT tielongshen realtimehevenergymanagementstrategyconsideringroadcongestionbasedondeepreinforcementlearning
_version_	1717760585059270656

Real-Time HEV Energy Management Strategy Considering Road Congestion Based on Deep Reinforcement Learning

Similar Items