DQL energy management: An online-updated algorithm and its application in fix-line hybrid electric vehicle

Zou, Runnan; Fan, Likang; Dong, Yanrui; Zheng, Siyu; Hu, Chenxing

doi:10.1016/j.energy.2021.120174

Cited by 38 publications

(12 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers are attempting to create new technologies and techniques that may be an alternative for the existing conventional fossil-based vehicle system because of their significant potential positive environmental impacts as well as the current research emphasis focusing mostly on EVs. Refs [77,85,88,108,134,152,163] investigate models based on EMSs for EV applications.…”

Section: Bibliometric Evaluation Of Journals Publishers and Countriesmentioning

confidence: 99%

“…"Energy Management Systems", "Electric Vehicles", "Secondary Batteries", "Energy Efficiency", "Hybrid Energy Storage Systems", and "Plug-in Hybrid Vehicles" are six areas of study that are gaining high interest. Different researchers have performed various types of literature evaluations and techniques in the fields of economic advantages of EV applications, EMSs, optimization and control for cost reduction, the flexibility of system operations, and the reducing of carbon emissions [77,85,88,108,134,141,144,152,163,194,195].…”

Section: Document Authorship Analysismentioning

confidence: 99%

“…In ref [142,149,152,153,155], the authors explored various issues related to optimal EV design and power distribution. However, the power flow and EV design need further improvement.…”

Section: Optimal Ev Design and Power Distribution Challengesmentioning

confidence: 99%

See 2 more Smart Citations

Optimized Energy Management Schemes for Electric Vehicle Applications: A Bibliometric Analysis towards Future Trends

et al. 2021

View full text Add to dashboard Cite

Concerns over growing greenhouse gas (GHG) emissions and fuel prices have prompted researchers to look into alternative energy sources, notably in the transportation sector, accounting for more than 70% of carbon emissions. An increasing amount of research on electric vehicles (EVs) and their energy management schemes (EMSs) has been undertaken extensively in recent years to address these concerns. This article aims to offer a bibliometric analysis and investigation of optimized EMSs for EV applications. Hundreds (100) of the most relevant and highly influential manuscripts on EMSs for EV applications are explored and examined utilizing the Scopus database under predetermined parameters to identify the most impacting articles in this specific field of research. This bibliometric analysis provides a survey on EMSs related to EV applications focusing on the different battery storages, models, algorithms, frameworks, optimizations, converters, controllers, and power transmission systems. According to the findings, more articles were published in 2020, with a total of 22, as compared to other years. The authors with the highest number of manuscripts come from four nations, including China, the United States, France, and the United Kingdom, and five research institutions, with these nations and institutions accounting for the publication of 72 papers. According to the comprehensive review, the current technologies are more or less capable of performing effectively; nevertheless, dependability and intelligent systems are still lacking. Therefore, this study highlights the existing difficulties and challenges related to EMSs for EV applications and some brief ideas, discussions, and potential suggestions for future research. This bibliometric research could be helpful to EV engineers and to automobile industries in terms of the development of cost-effective, longer-lasting, hydrogen-compatible electrical interfaces and well-performing EMSs for sustainable EV operations.

show abstract

Section: Bibliometric Evaluation Of Journals Publishers and Countriesmentioning

confidence: 99%

Section: Document Authorship Analysismentioning

confidence: 99%

See 1 more Smart Citation

Optimized Energy Management Schemes for Electric Vehicle Applications: A Bibliometric Analysis towards Future Trends

et al. 2021

View full text Add to dashboard Cite

show abstract

“…On the basis of velocity prediction results, the required power between engine and motor are distributed by reinforcement learning strategy in real time in References 33 and 34, which reduced the HEV energy consumption cost. Zou et al 35 takes the advantages of deep learning and reinforcement learning and proposes a deep Q‐learning strategy, while combined with model predictive control, an online update strategy is designed to improve the convergence speed of the deep Q‐learning strategy and enhance the HEV fuel economy. However, the energy management strategy based on intelligent algorithms not only needs to rely on controllers with strong arithmetic capability, but also requires a large amount of raw data for neural network training to improve the velocity prediction accuracy, which limits the application to the above control methods.…”

Section: Introductionmentioning

confidence: 99%

Energy optimization for intelligent hybrid electric vehicles based on hybrid system approach in a car‐following process

Zhou

Wang

Ding

2022

Optim Control Appl Methods

View full text Add to dashboard Cite

Considering the hybrid dynamical characteristics of multimode hybrid electric vehicle during a car-following process, a novel energy management strategy based on hybrid system theory is proposed in this article. Firstly, nonlinear powertrain model and longitudinal dynamics of a single-shaft parallel hybrid electric vehicle is built, and hybrid characteristics combined with continuous dynamics and discrete events during operating mode transition are described. Secondly, the nonlinear powertrain model and characteristics are approximated by the piecewise affine method, the mixed-logic dynamic modeling method is introduced and system variables are defined. Then the hybrid dynamical model for multimode hybrid electric vehicle in a car-following process is established with the hybrid system description language. Thirdly, according to the principle of receding horizon control, a hybrid model predictive controller is designed and applied for optimizing the multiobjective energy management problem in the car-following process. Finally, the proposed control strategy is compared with a rule-based control strategy, and the simulation results show that the proposed control strategy achieves fuel efficiency improvement while ensure the dynamic performance, driving safety and ride comfort during the car-following process, and the designed controller meets the requirement of deep fusion of car-following and energy management.

show abstract

“…The deep reinforcement learning method is also applied to EMS for Eco-driving [21], Du et al proposed a new framework for achieving experience sampling more reasonable [22]. Zou et al carried out a combination between model predictive control and deep network to accelerate the learning speed [23]. However, these two explorationexploitation policies are not completely biased, they are still having some difficulty in balancing the explorationexploitation process [24], especially on how to define the value of the deciding factor in epsilon-greedy policy to achieve an optimal control strategy.…”

Section: Introductionmentioning

confidence: 99%

Impact of Exploration-to-exploitation Ratio on Energy Saving Potential of Plug-in Hybrid Vehicles Controlled by Reinforcement Learning

Shuai,

Zhou,

Williams

et al. 2021

Preprint

View full text Add to dashboard Cite

Electric vehicles, including hybrids, will domain the road transport in future cities. Reinforcement learning has shown its capacity in online optimization of energy management strategy for hybrid vehicles. Exploration of new control settings and exploitation of the existing control policy are two key procedures in reinforcement learning but there is a lack of study on how the exploration-to-exploitation (E2E) ratio affects the energy efficiency improvement for hybrid vehicles. This paper introduces two decay functions, 'Reciprocal functionbased decay' (RBD) and 'Step-based decay' (SBD), to generate E2E ratio trajectories for reinforcement learning algorithm which is conventionally based on Exponential decay (EXD) function. By monitoring the improving rate of vehicle energy efficiency in the learning process, the vehicle controlled by Q-learning algorithm based on the SBD function has shown the best compared with the vehicles based on the RBD function and the EXD function. The improving rate can be more than 6.21%. In the HiL testing, the SBD can save 1.52% energy compared to the EXD in real-time control under a predefined driving cycle.

show abstract

DQL energy management: An online-updated algorithm and its application in fix-line hybrid electric vehicle

Cited by 38 publications

References 38 publications

Optimized Energy Management Schemes for Electric Vehicle Applications: A Bibliometric Analysis towards Future Trends

Optimized Energy Management Schemes for Electric Vehicle Applications: A Bibliometric Analysis towards Future Trends

Energy optimization for intelligent hybrid electric vehicles based on hybrid system approach in a car‐following process

Impact of Exploration-to-exploitation Ratio on Energy Saving Potential of Plug-in Hybrid Vehicles Controlled by Reinforcement Learning

Contact Info

Product

Resources

About