Function approximation reinforcement learning of energy management with the fuzzy REINFORCE for fuel cell hybrid electric vehicles

作     者：Liang Guo Zhongliang Li Rachid Outbib Fei Gao 

作者机构：Aix-Marseille UniversityLIS UMR CNRS 7020MarseilleFrance Universite de Franche-ComteUTBMCNRSinstitut FEMTO-STF-90000 BelfortFrance UTBMCNRSinstitut FEMTO-STF-90000 BelfortFrance 

出 版 物：《Energy and AI》 (能源与人工智能（英文）)

年 卷 期：2023年第13卷第3期

页      面：76-87页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 0808[工学-电气工程] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work has been supported by the ANR DEAL(contract ANR-20-CE05-0016-01) This work has also been partially funded by Region Sud Provence-Alpes-Cote d’Azur via project AMULTI(2021_02918) 

主　　题：Energy management strategy Fuel cell hybrid electric vehicle Reinforcement learning Fuzzy inference system Fuzzy policy gradient Hardware-in-loop 

摘      要：In the paper,a novel self-learning energy management strategy(EMS)is proposed for fuel cell hybrid electric vehicles(FCHEV)to achieve the hydrogen saving and maintain the battery *** the EMS,it is proposed to approximate the EMS policy function with fuzzy inference system(FIS)and learn the policy parameters through policy gradient reinforcement learning(PGRL).Thus,a so-called Fuzzy REINFORCE algorithm is first proposed and studied for EMS problem in the *** REINFORCE is a model-free method that the EMS agent can learn itself through interactions with environment,which makes it independent of model accuracy,prior knowledge,and expert ***,to stabilize the training process,a fuzzy baseline function is adopted to approximate the value function based on FIS without affecting the policy gradient ***-over,the drawbacks of traditional reinforcement learning such as high computation burden,long convergence time,can also be *** effectiveness of the proposed methods were verified by Hardware-in-Loop *** adaptability of the proposed method to the changes of driving conditions and system states is also verified.