Battery is one of the most importance components in the field of energy system. In other hand LiFePO4 batteries have higher density of energy and more life cycle than nickel battery, but LiFePO4 need state of charge (SOC) prediction to solve the disadvantage of this battery. The objective of this study is to make method for calculate SOC that can compute it with high accuracy and fast computation time. Decision tree is one of logical computing based on supervised learning that can make precision of prediction. This study concentrated to develop decision tree for SOC prediction. Decision trees train with the real data from battery testing system. The average error from training data is 0.1789 with 0.047353 seconds time computation, and the average error from testing data is 1.9034. The conclusion from this study is decision tree can use to calculate SOC and it can be applied because of its accuracy and fast.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Decision tree for state of charge (SOC) prediction of LiFePO4 battery


    Beteiligte:
    Nizam, Muh. (Autor:in) / Mujianto, Agus (Autor:in) / Waloyo, Hery Tri (Autor:in) / Purwanto, Agus (Autor:in) / Inayati (Autor:in)


    Erscheinungsdatum :

    01.11.2015


    Format / Umfang :

    331113 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    State of charge (SoC) estimation of LiFePO4 battery module using support vector regression

    Haq, Irsyad Nashirul / Saputra, Riza Hadi / Edison, Frans et al. | IEEE | 2015



    Life Prediction Method of LiFePO4 Traction Battery Based on Interval Charging Capacity

    Lin, Chunjing / Wang, Fang / Bai, Guangli et al. | SAE Technical Papers | 2021