Machine-Learning Based Passive Cell Balancing for Enhanced Battery Management in Electrified Transportation under Varying Thermal Conditions

سال انتشار: 1406
نوع سند: مقاله ژورنالی
زبان: انگلیسی
مشاهده: 17

فایل این مقاله در 14 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

JR_JREE-14-1_005

تاریخ نمایه سازی: 22 آبان 1405

چکیده مقاله:

Cell balancing plays a crucial role in the effective operation of a Battery Management System (BMS), as it helps prolong both battery runtime and overall lifespan. With the increasing demand for high-capacity and high-performance battery packs, attention has shifted toward optimizing various balancing techniques. Among them, passive balancing remains the most widely used method due to its simplicity and cost-effectiveness. However, because this method dissipates excess energy as heat through resistors, careful thermal management is essential. Ensuring that the BMS circuit board remains within a safe temperature range is key to maintaining system reliability and performance. This study explores the application of machine learning to optimize passive cell balancing for BMSs in electric vehicles. MATLAB simulations were performed on a three-cell battery pack with state-of-charge (SOC) values of ۲۰%, ۲۲%, and ۲۵%, using balancing resistors (۲۰ Ω, ۳۰ Ω, and ۴۰ Ω) under varying ambient temperatures (۲۰°C, ۲۵°C, and ۳۰°C). The results demonstrate that higher temperatures increase heat loss, potentially leading to cell failure, while higher resistor values extend balancing time but reduce power loss. To enhance the balancing process, a Feedforward Neural Network (FNN) model was developed to predict the optimal balancing resistors based on voltage, current, and temperature data obtained from the simulations. The FNN model achieved its best performance at ۲۰°C, with a Mean Absolute Error (MAE) of ۰.۱۳۶۶۸ and a Root Mean Squared Error (RMSE) of ۰.۲۲۱۰۸, demonstrating its effectiveness in improving passive balancing accuracy. This approach demonstrates the potential of machine learning to enhance the efficiency and reliability of passive cell balancing in BMSs, particularly under varying thermal conditions, thereby contributing to improved battery management in electric vehicles.

نویسندگان

- -

Department of Electrical Engineering, Jamia Millia Islamia, New Delhi, India.

- -

Department of Electrical Engineering, Jamia Millia Islamia, New Delhi, India.

- -

Department of Electrical Engineering, Aligarh Muslim University, Aligarh, India.

مراجع و منابع این مقاله:

لیست زیر مراجع و منابع استفاده شده در این مقاله را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود مقاله لینک شده اند :