Scholarlink

<p>In this paper, a novel machine learning based real-time dynamic programming for battery thermal management system (ML-RTDP-BTMS) has been proposed for electric vehicles (EVs). The proposed machine learning based method uses a predictive machine learning technique to improve the performance of RTDP. The machine learning (ML) model helps to forecast battery parameters such as temperature and the state of charge (SoC), by previous data analysis. This makes RTDP more dynamic and approachable decisions in real time. The proposed combination fulfills those shortcomings of conventional techniques which makes it difficult to control the growing complexity and unpredictability of battery systems under dynamic operating conditions. A comparative analysis shows that the ML-RTDP-BTMS improves battery performance, reduces thermal stress, and increases total system efficiency significantly by least amount of energy consumption (7 kWh). Statistical study and simulations results validate the approach with lowest mean temperature deviation (2.5&#xa0;°C), demonstrating better performance in controlling battery temperature and extending battery life.</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

Machine Learning-Based Real-Time Dynamic Programming for Battery Thermal Management System (ML-RTDP-BTMS) for Electric Vehicles

  • 期刊论文

  • 发表日期：

    2026年2月20日

查看全文
  • Kalpana Chauhan,
  • Rajeev Kumar Chauhan

摘要

In this paper, a novel machine learning based real-time dynamic programming for battery thermal management system (ML-RTDP-BTMS) has been proposed for electric vehicles (EVs). The proposed machine learning based method uses a predictive machine learning technique to improve the performance of RTDP. The machine learning (ML) model helps to forecast battery parameters such as temperature and the state of charge (SoC), by previous data analysis. This makes RTDP more dynamic and approachable decisions in real time. The proposed combination fulfills those shortcomings of conventional techniques which makes it difficult to control the growing complexity and unpredictability of battery systems under dynamic operating conditions. A comparative analysis shows that the ML-RTDP-BTMS improves battery performance, reduces thermal stress, and increases total system efficiency significantly by least amount of energy consumption (7 kWh). Statistical study and simulations results validate the approach with lowest mean temperature deviation (2.5 °C), demonstrating better performance in controlling battery temperature and extending battery life.