http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : Yunduan Cui (검색결과 3 건)
- 무료
- 기관 내 무료
- 유료
-
Dongfang Zhang,Yunduan Cui,Yao Xiao,Shengxiang Fu,Suk Won Cha,Namwook Kim,Hongyan Mao,Chunhua Zheng 한국정밀공학회 2024 International Journal of Precision Engineering and Vol.11 No.1
With the rapid development of artificial intelligence, deep reinforcement learning (DRL)-based energy management strategies (EMSs) have become an important research direction for hybrid electric vehicles recently, which still face some problems such as fragile convergence characteristics, slower convergence speed, and unsatisfactory optimization effects. In this research, a novel DRL algorithm, i.e. an improved soft actor-critic (ISAC) algorithm is applied to the EMS of a fuel cell hybrid vehicle (FCHV), in which the priority experience replay (PER) and emphasizing recent experience (ERE) methods are adopted to improve the convergence performance of the algorithm and to enhance the FCHV fuel economy. In addition, the fuel cell durability is also considered in the proposed EMS based on a nonlinear fuel cell degradation model while considering the fuel economy. Results indicate that the FCHV fuel consumption of the proposed EMS is decreased by 7.87%, 2.79%, and 2.44% compared to that of the deep deterministic policy gradient (DDPG)-based, the twin delayed deep deterministic policy gradient (TD3)-based, and the SAC-based EMSs respectively while the fuel consumption gap to the dynamic programming-based EMS is narrowed to 2.37% by the proposed EMS. Moreover, the proposed EMS presents the best training performance considering both the convergence speed and stability, and the convergence speed of the proposed EMS is increased by an average of 47.89% compared to that of the other DRL-based EMSs. Furthermore, the fuel cell durability is improved by more than 95% using the proposed EMS compared to that of the EMS without considering the fuel cell degradation.
-
Wei Li,Jiaye Ye,Yunduan Cui,김남욱,차석원,Chunhua Zheng 한국정밀공학회 2022 International Journal of Precision Engineering and Vol.9 No.3
A speedy reinforcement learning (RL)-based energy management strategy (EMS) is proposed for fuel cell hybrid vehicles (FCHVs) in this research, which approaches near-optimal results with a fast convergence rate based on a pre-initialization framework and meanwhile possesses the ability to extend the fuel cell system (FCS) lifetime. In the pre-initialization framework, well-designed power distribution-related rules are used to pre-initialize the Q-table of the RL algorithm to expedite its optimization process. Driving cycles are modeled as Markov processes and the FCS power difference between adjacent moments is used to evaluate the impact on the FCS lifetime in this research. The proposed RL-based EMS is trained on three driving cycles and validated on another driving cycle. Simulation results demonstrate that the average fuel consumption difference between the proposed EMS and the EMS based on dynamic programming is 5.59% on the training driving cycles and the validation driving cycle. Additionally, the power fluctuation on the FCS is reduced by at least 13% using the proposed EMS compared to the conventional RL-based EMS which does not consider the FCS lifetime. This is significantly beneficial for improving the FCS lifetime. Furthermore, compared to the conventional RL-based EMS, the convergence speed of the proposed EMS is increased by 69% with the pre-initialization framework, which presents the potential for realtime applications.
-
Dezhou Xu,Chunhua Zheng,Yunduan Cui,Shengxiang Fu,Namwook Kim,Suk Won Cha 한국정밀공학회 2023 International Journal of Precision Engineering and Vol.10 No.1
Hybrid vehicles (HVs) that equip at least two different energy sources have been proven to be one of effective and promising solutions to mitigate the issues of energy crisis and environmental pollution. For HVs, one of the core supervisory control problems is the power distribution among multiple power sources, and for this problem, energy management strategies (EMSs) have been studied to save energy and extend the service life of HVs. In recent years, with the rapid development of artificial intelligence and computer technologies, learning algorithms have been gradually applied to the EMS field and shortly become a novel research hotspot. Although there are some brief reviews on the learning-based (LB) EMSs for HVs in recent years, a state-of-the-art and thorough review related to the applications of learning algorithms in HV EMSs still lacks. In this paper, learning algorithms applied in HV EMSs are categorized and reviewed in terms of the reinforcement learning algorithms and deep reinforcement learning algorithms. Apart from presenting the recent progress of learning algorithms applied in HV EMSs, advantages and disadvantages of different learning algorithms and LB EMSs are also discussed. Finally, a brief outlook related to the further applications of learning algorithms in HV EMSs, such as the integration towards autonomous driving and intelligent transportation system, is presented.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
