A Feature-Turned XGBoost Model For Real Time SOC Prediction

  • Unique Paper ID: 175794
  • Volume: 11
  • Issue: 11
  • PageNo: 4679-4681
  • Abstract:
  • The evolution of cloud-based lithium-ion battery management systems has revolutionized state-of-charge (SOC) estimation. Traditional estimation methods, such as Extended Kalman Filter (EKF), struggle with accuracy and computational efficiency. This paper presents a comparative analysis of deep-learning-based SOC estimation algorithms, including Feedforward Neural Networks (FNN), Gated Recurrent Units (GRU), and Long Short-Term Memory (LSTM). The extension introduces XGBoost for feature optimization, reducing RMSE and MAE errors. The integration of cloud computing enhances computational capabilities, allowing real-time estimation. Results indicate that EKF and XGBoost outperform conventional techniques, providing faster and more precise SOC predictions, ensuring efficient battery management and prolonging battery life for electric vehicles.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 11
  • PageNo: 4679-4681

A Feature-Turned XGBoost Model For Real Time SOC Prediction

Related Articles