Intelligent Digital Twins for Battery Management with Explainable AI

Intelligent Digital Twins for Battery Management with Explainable AI

Dr. D. Vijaya Lakshmi1, Ganji Sai Deepak Reddy2, Thara Kapuganti3

1Professor & HOD, Mahatma Gandhi Institute of Technology

2,3UG Student, Mahatma Gandhi Institute of Technology

Abstract- This project proposes a novel approach in the battery state prediction of electric vehicles using Explainable Data-Driven Digital Twins with advanced machine learning algorithms such as deep Neural Networks (DNN), Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNN), Support Vector Regression (SVR), Support Vector Machines (SVM), Feedforward Neural Networks (FNN), Radial Basis Function networks (RBF), Random Forests (RF), and Extreme Gradient Boosting (XGBoost). Therefore, the aim would be to better predict important parameters of the battery, such as SOC and SOH, by running these under different operation conditions. These are achieved in the digital twin model by means of a diversification of a set of algorithms to achieve very high accuracy as well as robustness, followed by the employment of XAI techniques to better explain the predicted values. Preliminary results suggest that this integrated approach significantly outperforms traditional methods in terms of accuracy and reliability and significantly contributes to the advancement of intelligent, adaptive battery management systems crucial for future electric mobility.

Keywords: Electric vehicles, battery prediction, Digital Twins, machine learning, DNN, LSTM, XGBoost, SOC, SOH, Explainable AI, adaptive battery management