Article
Energy & Fuels
Muhammad Aurangzeb, Ai Xin, Sheeraz Iqbal, Muhammad Zeshan Afzal, Hossam Kotb, Kareem M. AboRas, Yazeed Yasin Ghadi, Bello-Pierre Ngoussandou
Summary: This research study investigates the use of a droop-ANN model to enhance power quality in vehicle-to-grid (V2G) systems. Simulation results demonstrate that the droop-ANN model significantly improves power quality across various battery states of charge and charging/discharging scenarios, highlighting its potential to enhance stability and reliability in V2G systems.
INTERNATIONAL JOURNAL OF ENERGY RESEARCH
(2023)
Article
Engineering, Electrical & Electronic
Mohammadali Kargar, Chen Zhang, Xingyong Song
Summary: This article studies the problem of autonomous hybrid electric vehicles following a leader, integrating the external dynamics and powertrain dynamics for optimization. A customized control strategy based on Approximate Dynamic Programming and neural networks is proposed, and the accuracy of the optimization solution is improved by applying the concept of reachable sets. Three case studies demonstrate that the examined integrated control strategy significantly improves fuel consumption compared to the separated optimization method.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Automation & Control Systems
Peng Mei, Hamid Reza Karimi, Hehui Xie, Fei Chen, Cong Huang, Shichun Yang
Summary: Considering the importance of energy management strategy for hybrid electric vehicles, this paper addresses the energy optimization control issue using reinforcement learning algorithms. It establishes a hybrid electric vehicle power system model and designs a hierarchical energy optimization control architecture based on networked information. Three learning-based energy optimization control strategies, namely Q-learning, deep Q network (DQN), and deep deterministic policy gradient (DDPG) algorithms, are introduced. The superiority of the DDPG algorithm over Q-learning and DQN algorithms in terms of robustness and faster convergence for better energy management purposes is illustrated through simulation.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Mohammadali Kargar, Tohid Sardarmehni, Xingyong Song
Summary: This article focuses on the control of powertrain energy management for an autonomous HEV and introduces a new control strategy based on flexible power demand. The power flexibility is incorporated into the Approximate Dynamic Programming (ADP) framework. An example is provided to demonstrate the feasibility of the proposed method.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Energy & Fuels
Dong Hu, Hui Xie, Kang Song, Yuanyuan Zhang, Long Yan
Summary: This study proposes an apprenticeship-reinforcement learning (A-RL) framework based on expert demonstration (ED) model embedding to improve efficient energy management strategies (EMS) for hybrid electric vehicles (HEV). The framework combines apprenticeship learning (AL) with deep reinforcement learning (DRL) and uses the ED model to guide the DRL. The results show significant improvement in training convergence rate and fuel economy.
Article
Thermodynamics
Chunyang Qi, Yiwen Zhu, Chuanxue Song, Guangfu Yan, Da Wang, Feng Xiao, Xu Zhang, Jingwei Cao, Shixin Song
Summary: This research introduces a novel reinforcement learning-based deep Q-learning algorithm for the energy management strategy of HEVs. The proposed method not only addresses the issue of sparse reward during training, but also achieves optimal power distribution. Additionally, the hierarchical structure of the algorithm enhances exploration of the vehicle environment, leading to improved training efficiency and reduced fuel consumption.
Article
Engineering, Aerospace
Ye Xie, Shaoming He, Al Savvaris, Antonios Tsourdos, Dan Zhang, Anhuan Xie
Summary: This paper proposes a new convexification method to simplify the energy management problem of a hybrid electric propulsion system and proves the equality between the original problem and the convexified problem. By variable change and equality relaxation, the original problem is convexified and the effectiveness of the convexified problem is verified through numerical examples. Compared with the benchmark optimization-dynamic programming, convex optimization shows important advantages in terms of optimization computation time and control value fluctuation.
AEROSPACE SCIENCE AND TECHNOLOGY
(2022)
Article
Thermodynamics
Shradhdha Sarvaiya, Sachin Ganesh, Bin Xu
Summary: Hybrid Electric Vehicles (HEV) bridge the gap between traditional internal combustion engine vehicles and electric motor-powered vehicles. Battery life evaluation and Energy Management Strategy (EMS) play crucial roles in prolonging battery lifespan. This research compares different control strategies for battery life optimization, emphasizing the impact of parameters like temperature and current on battery aging.
Article
Energy & Fuels
Zhen Zhang, Tiezhu Zhang, Jichao Hong, Hongxin Zhang, Jian Yang
Summary: This article proposes a master-slave hybrid electric vehicle (MSHEV) with multiple energy sources, which transitions between different working modes and has lower power consumption and energy loss compared to electric vehicles (EVs) under actual driving conditions. By optimizing the battery state of charge and constructing a Response Surface Model-based approximate model and Multi-Island Genetic Algorithm-based optimization model, the energy management of the optimized MSHEV is enhanced, indicating significant importance and reference value in the optimization of energy management of hybrid electric vehicles.
Article
Engineering, Electrical & Electronic
Mehrdad Ehsani, Krishna Veer Singh, Hari Om Bansal, Ramin Tafazzoli Mehrjardi
Summary: Electric and hybrid electric vehicles are promising solutions for reducing pollution and saving fossil fuels, requiring optimization of components, systems, and controls in their design. This overview discusses the challenges and future technologies in the field, serving as a reference for researchers working on EV/HEV.
PROCEEDINGS OF THE IEEE
(2021)
Review
Energy & Fuels
Zhen Song, Yue Pan, Huicui Chen, Tong Zhang
Summary: Fuel cell hybrid electric vehicles, powered by both fuel cell and lithium-ion battery, have been recognized as a potential alternative to internal combustion vehicles due to their high energy density and conversion efficiency. Energy management and thermal management are critical for stable power output, and strategies integrating temperature effects are crucial for improving efficiency and service life.
Article
Thermodynamics
Chunyang Qi, Chuanxue Song, Feng Xiao, Shixin Song
Summary: This paper investigates the generalization capability of energy management strategies for hybrid electric vehicles and proposes a multi-agent reinforcement learning algorithm. By analyzing typical features and using an auxiliary agent, the generalization performance of energy management strategies is improved.
Review
Green & Sustainable Science & Technology
Teng Liu, Wenhao Tan, Xiaolin Tang, Jinwei Zhang, Yang Xing, Dongpu Cao
Summary: This paper summarizes driving cycle-driven energy management strategies for HEVs and emphasizes the importance of driving cycles in the field. It reviews relevant literature, studies different types of EMSs, and showcases the status quo of driving cycle databases. Finally, it discusses the future prospects of energy management technologies related to driving cycles.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Article
Engineering, Electrical & Electronic
Ningkang Yang, Lijin Han, Rui Liu, Zhengchao Wei, Hui Liu, Changle Xiang
Summary: This article proposes a multiobjective energy management strategy based on multiagent reinforcement learning for a hybrid electric vehicle. The strategy takes into consideration fuel economy improvement, battery state of charge maintenance, battery degradation reduction, and constraint on ultracapacitor state of charge. The proposed strategy combines game theory and reinforcement learning to achieve a Nash equilibrium among multiple objectives. Simulation results show that the proposed strategy outperforms single-agent reinforcement learning and dynamic programming in optimizing multiple objectives.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2023)
Review
Energy & Fuels
Xueqin Lu, Siwei Li, XiangHuan He, Chengzhi Xie, Songjie He, Yuzhe Xu, Jian Fang, Min Zhang, Xingwu Yang
Summary: This paper comprehensively summarizes the advantages and applicable scenarios of hybrid electric vehicle energy management strategy based on model predictive control. The specific application and actual performance of this strategy in different systems are analyzed. The research aims to provide guidance and design ideas for researchers and promote the development of energy management strategies.
JOURNAL OF ENERGY STORAGE
(2022)
Article
Thermodynamics
Ningkang Yang, Lijin Han, Changle Xiang, Hui Liu, Xunmin Li
Summary: This paper proposes a real-time indirect reinforcement learning based strategy to reduce fuel consumption. By introducing a high-order Markov Chain for an accurate environment model, the learning process is more precise and effective.
Article
Thermodynamics
Shumin Ruan, Yue Ma, Ningkang Yang, Changle Xiang, Xunming Li
Summary: This paper proposes a framework of an energy-saving controller for hybrid electric vehicles (HEVs) based on explicit model predictive control (MPC), achieving real-time control by precomputing explicit solutions and coupling car-following control and energy management strategy together. Numerical simulations and hardware-in-the-loop experiments demonstrate the effectiveness and real-time performance of the proposed controller.
Article
Thermodynamics
Lijin Han, Ke Yang, Tian Ma, Ningkang Yang, Hui Liu, Lingxiong Guo
Summary: This paper proposes a real-time energy management strategy for hybrid electric vehicles based on reinforcement learning. By using a recursive Markov chain and eligibility trace algorithm, it aims to improve fuel economy and minimize battery degradation, enhancing the adaptability of EMS to various driving conditions.
Article
Thermodynamics
Zhengchao Wei, Yue Ma, Ningkang Yang, Shumin Ruan, Changle Xiang
Summary: This paper proposes a reinforcement learning-based power management strategy for hybrid electric power systems, integrating the economic rotational speed of the turboshaft engine and a safety constraints-based variable action space approach. The strategy effectively controls the discharging/charging power and state of charge of the battery while optimizing the action space to prevent violation of safety constraints. The results show a fuel consumption reduction of 4.29% under air-land driving conditions, and the hardware-in-the-loop experiment demonstrates real-time performance of the strategy.
Article
Thermodynamics
Shumin Ruan, Yue Ma, Ningkang Yang, Qi Yan, Changle Xiang
Summary: This paper proposes a novel multiobjective optimization controller based on the Nash bargaining game to improve energy efficiency, traffic safety, and driving comfort for hybrid electric vehicles. By treating longitudinal dynamic control and energy management strategy as independent players, the proposed controller outperforms other methods in terms of control performance optimality and robustness, as demonstrated by simulation results.
Article
Thermodynamics
Lingxiong Guo, Hui Liu, Lijin Han, Ningkang Yang, Rui Liu, Changle Xiang
Summary: This paper proposes a model predictive control-based predictive energy management strategy for dual-mode HEV. By predicting the future vehicle speed and designing an improved sequence quadratic programming algorithm, the computational efficiency and optimality are effectively improved. Meanwhile, a dynamic process coordination control algorithm is developed to address the torque coordination problem and mode shift dynamic process balance in energy management. Experimental results demonstrate the desirable performance of the proposed strategy in terms of fuel saving, real-time capability, and robustness.
Article
Engineering, Electrical & Electronic
Xiuqi Chen, Wei Wei, Qingdong Yan, Ningkang Yang, Jingqiu Huang
Summary: The stability of brake control is crucial for the safety of HDVs at high speeds. However, electro-hydraulic braking systems often suffer from significant delay, making it challenging to predict and control braking performance. To solve the torque tracking control problem with time delay, a deep inference and control method is proposed. The proposed method uses a data-driven model to identify the theoretical delay time under different rotating speeds and employs a deep Q-network learning approach to find the optimal control strategy considering the delay. Comparative simulations and PIL tests demonstrate the effectiveness and efficiency of the proposed control framework.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Ningkang Yang, Lijin Han, Changle Xiang, Hui Liu, Tian Ma, Shumin Ruan
Summary: This paper proposes a reinforcement learning-based real-time energy management strategy for hybrid electric vehicles. The proposed method achieves higher fuel efficiency and real-time optimization through a recursive Markov Chain model and heuristic search.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Proceedings Paper
Engineering, Mechanical
Lijin Han, Ke Yang, Xin Zhang, Ningkang Yang, Hui Liu, Jiaxin Liu
Summary: This paper presents an energy management strategy for hybrid electric vehicles using double Q-learning to reduce fuel consumption. The strategy aims to optimize engine performance by distributing mechanical energy and electrical energy during the vehicle's driving process, thereby achieving improved fuel economy and reduced emissions.
INTERNATIONAL CONFERENCE ON MECHANICAL DESIGN AND SIMULATION (MDS 2022)
(2022)