Battery State of Charge and State of Health Estimation Using a New Hybrid Deep Neural Network Approach
The increasing adoption of Battery Electric Vehicles (BEVs) is driving advancements in battery management systems (BMS) to address challenges like cost and range anxiety, both tied to battery performance. This paper investigates various state of charge (SOC) and state of health (SOH) estimation meth...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10738796/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The increasing adoption of Battery Electric Vehicles (BEVs) is driving advancements in battery management systems (BMS) to address challenges like cost and range anxiety, both tied to battery performance. This paper investigates various state of charge (SOC) and state of health (SOH) estimation methods, presenting a novel hybrid neural network that combines Gate Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) models. Our proposed method demonstrates significant improvements in SOH and SOC estimation accuracy, with minimal training data required. Key contributions include (1) a hybrid GRU-LSTM model improving SOC/SOH accuracy, (2) self-optimization capabilities, (3) effective handling of temperature variations without OCV-SOC lookup tables, and (4) its application to various lithium battery types. Experimental results show the method achieves an RMSE of 2% and MAE of 1.7% for SOC, and an RMSE of 0.65% and MAE of 0.85% for SOH across a temperature range of −10°C to 40°C. These results indicate a reliable and cost-effective approach for BEV battery management, contributing to the wider adoption of sustainable transportation. |
|---|---|
| ISSN: | 2169-3536 |