Scholarlink

<p>In the effort to minimize carbon emissions, hybrid energy storage systems (HESS) have become crucial, especially in electric vehicle applications. These systems combine batteries and supercapacitors to effectively manage energy and power demands. However, coordinating power between the two storage elements is challenging due to their distinct energy and power density characteristics. This paper presents the design and implementation of a hybrid machine learning-based controller that integrates fuzzy and reinforcement learning for efficient power coordination within a passive HESS topology. The study begins with an evaluation of various energy storage technologies, identifying the battery–supercapacitor combination as a cost-effective, lightweight, and high-performance solution. Although passive topologies offer simplicity and efficiency, they traditionally lack flexibility. To address this challenge, the proposed controller integrates fuzzy logic with reinforcement learning to optimize power distribution. This effectively reduces battery stress by lowering its load contribution from 41.23 to 31.09% during transient conditions. The overall system efficiency significantly improved, rising from 88.586% with fuzzy-only control to 97.34% with the integrated reinforcement learning and fuzzy logic approach. This new method outperforms the baseline passive system efficiency, which stands at 77.69%. Hardware-in-the-loop validation conducted on the dSPACE RTI platform confirmed the controller's robustness and real-time effectiveness. The results demonstrate the effectiveness of the proposed fuzzy and reinforcement learning-based approach. It enhances the flexibility, efficiency, and reliability of HESS, contributing meaningfully to the advancement of sustainable electric mobility.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Designing an Intelligent Power Coordinate Controller for Hybrid Energy Storage Devices Using Machine Learning Algorithms in Electric Vehicle Application

  • 期刊论文

  • 发表日期：

    2026年2月19日

查看全文
  • Kurakula Anudeep,
  • Alivelu M. Parimi,
  • Sandip S. Deshmukh,
  • Parikshit Sahatiya

摘要

In the effort to minimize carbon emissions, hybrid energy storage systems (HESS) have become crucial, especially in electric vehicle applications. These systems combine batteries and supercapacitors to effectively manage energy and power demands. However, coordinating power between the two storage elements is challenging due to their distinct energy and power density characteristics. This paper presents the design and implementation of a hybrid machine learning-based controller that integrates fuzzy and reinforcement learning for efficient power coordination within a passive HESS topology. The study begins with an evaluation of various energy storage technologies, identifying the battery–supercapacitor combination as a cost-effective, lightweight, and high-performance solution. Although passive topologies offer simplicity and efficiency, they traditionally lack flexibility. To address this challenge, the proposed controller integrates fuzzy logic with reinforcement learning to optimize power distribution. This effectively reduces battery stress by lowering its load contribution from 41.23 to 31.09% during transient conditions. The overall system efficiency significantly improved, rising from 88.586% with fuzzy-only control to 97.34% with the integrated reinforcement learning and fuzzy logic approach. This new method outperforms the baseline passive system efficiency, which stands at 77.69%. Hardware-in-the-loop validation conducted on the dSPACE RTI platform confirmed the controller's robustness and real-time effectiveness. The results demonstrate the effectiveness of the proposed fuzzy and reinforcement learning-based approach. It enhances the flexibility, efficiency, and reliability of HESS, contributing meaningfully to the advancement of sustainable electric mobility.