The existing studies have shown that regarding the driving style as transient information to allocate vehicle power can significantly improve the fuel economy of hybrid electric vehicles (HEV s). To make full use of this information in a more efficient algorithm, a multi-objective energy management strategy (EMS) considering driving style recognition based on Proximal Policy Optimization (PPO) algorithm for the plug-in hybrid electric bus (PHEB) is proposed in this paper. Firstly, the optimized fuzzy C-means algorithm is used to identify driving conditions. Then the fuzzy rules combining the identification re-sults for driving style recognition are constructed. After that, the recognition results are used as the state of the deep reinforcement learning (DRL) algorithm and an energy management framework is established to minimize fuel consumption, maintain the battery SOC and ensure vehicle power performance. Simulation results have shown that the proposed EMS can significantly improve vehicle fuel economy under the combined driving cycles and real driving cycles.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep Reinforcement Learning based Multi-Objective Energy Management Strategy for a Plug-in Hybrid Electric Bus Considering Driving Style Recognition


    Contributors:
    Zhang, Chunmei (author) / Cui, Wei (author) / Cui, Naxin (author)


    Publication date :

    2022-10-28


    Size :

    3433130 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Multi-objective real-time optimization energy management strategy for plug-in hybrid electric vehicle

    Du, Siyu / Yang, Yiyong / Liu, Congzhi et al. | SAGE Publications | 2019



    Urban Driving with Multi-Objective Deep Reinforcement Learning

    Li, Changjian / Czarnecki, Krzysztof | ArXiv | 2018

    Free access

    Advanced Energy Management Strategies for Plug-In Hybrid Electric Vehicles via Deep Reinforcement Learning

    Mousa, Amr / Weiss, Gerhard Benedikt | British Library Conference Proceedings | 2022