Intelligent Battery Health-Aware Energy Management Strategy for Hybrid Electric Bus: A Deep Reinforcement Learning Method

Ruchen Huang; Hongwen He

doi:10.46855/energy-proceedings-9997

Intelligent Battery Health-Aware Energy Management Strategy for Hybrid Electric Bus: A Deep Reinforcement Learning Method

Ruchen Huang
, Hongwen He^*

^*Corresponding author for this work

School of Mechanical Engineering

Beijing Institute of Technology

Research output: Contribution to journal › Conference article › peer-review

2 Citations (Scopus)

Abstract

This paper proposes an intelligent battery health-aware energy management strategy (EMS) for the hybrid electric bus (HEB) with a deep reinforcement learning (DRL) method. Firstly, an EMS based on twin delayed deep deterministic policy gradient (TD3) algorithm considering battery health is innovatively designed to minimize the total operating cost of the HEB. Secondly, the superiority of the proposed EMS over the state-of-the-art deep deterministic policy gradient (DDPG) based strategy is validated. Simulation results show that the proposed EMS accelerates the convergence by 24.00% and reduces the total operating cost by 9.58% compared with the EMS based on DDPG.

Original language	English
Journal	Energy Proceedings
Volume	25
DOIs	https://doi.org/10.46855/energy-proceedings-9997
Publication status	Published - 2022
Event	Applied Energy Symposium, MIT A+B 2022 - Cambridge, United States Duration: 5 Jul 2022 → 8 Jul 2022

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

battery health
deep reinforcement learning
energy management
hybrid electric bus
twin delayed deep deterministic policy gradient (TD3)

Access to Document

10.46855/energy-proceedings-9997

Cite this

@article{1238387db1904b55bf6f18a91c045db1,

title = "Intelligent Battery Health-Aware Energy Management Strategy for Hybrid Electric Bus: A Deep Reinforcement Learning Method",

abstract = "This paper proposes an intelligent battery health-aware energy management strategy (EMS) for the hybrid electric bus (HEB) with a deep reinforcement learning (DRL) method. Firstly, an EMS based on twin delayed deep deterministic policy gradient (TD3) algorithm considering battery health is innovatively designed to minimize the total operating cost of the HEB. Secondly, the superiority of the proposed EMS over the state-of-the-art deep deterministic policy gradient (DDPG) based strategy is validated. Simulation results show that the proposed EMS accelerates the convergence by 24.00\% and reduces the total operating cost by 9.58\% compared with the EMS based on DDPG.",

keywords = "battery health, deep reinforcement learning, energy management, hybrid electric bus, twin delayed deep deterministic policy gradient (TD3)",

author = "Ruchen Huang and Hongwen He",

year = "2022",

doi = "10.46855/energy-proceedings-9997",

language = "English",

volume = "25",

journal = "Energy Proceedings",

issn = "2004-2965",

publisher = "Scanditale AB",

}

TY - JOUR

T1 - Intelligent Battery Health-Aware Energy Management Strategy for Hybrid Electric Bus

T2 - Applied Energy Symposium, MIT A+B 2022

AU - Huang, Ruchen

AU - He, Hongwen

PY - 2022

Y1 - 2022

N2 - This paper proposes an intelligent battery health-aware energy management strategy (EMS) for the hybrid electric bus (HEB) with a deep reinforcement learning (DRL) method. Firstly, an EMS based on twin delayed deep deterministic policy gradient (TD3) algorithm considering battery health is innovatively designed to minimize the total operating cost of the HEB. Secondly, the superiority of the proposed EMS over the state-of-the-art deep deterministic policy gradient (DDPG) based strategy is validated. Simulation results show that the proposed EMS accelerates the convergence by 24.00% and reduces the total operating cost by 9.58% compared with the EMS based on DDPG.

AB - This paper proposes an intelligent battery health-aware energy management strategy (EMS) for the hybrid electric bus (HEB) with a deep reinforcement learning (DRL) method. Firstly, an EMS based on twin delayed deep deterministic policy gradient (TD3) algorithm considering battery health is innovatively designed to minimize the total operating cost of the HEB. Secondly, the superiority of the proposed EMS over the state-of-the-art deep deterministic policy gradient (DDPG) based strategy is validated. Simulation results show that the proposed EMS accelerates the convergence by 24.00% and reduces the total operating cost by 9.58% compared with the EMS based on DDPG.

KW - battery health

KW - deep reinforcement learning

KW - energy management

KW - hybrid electric bus

KW - twin delayed deep deterministic policy gradient (TD3)

UR - https://www.scopus.com/pages/publications/85188873071

U2 - 10.46855/energy-proceedings-9997

DO - 10.46855/energy-proceedings-9997

M3 - Conference article

AN - SCOPUS:85188873071

SN - 2004-2965

VL - 25

JO - Energy Proceedings

JF - Energy Proceedings

Y2 - 5 July 2022 through 8 July 2022

ER -

Intelligent Battery Health-Aware Energy Management Strategy for Hybrid Electric Bus: A Deep Reinforcement Learning Method

Abstract

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this