Deep Learning for Battery State Prediction in Electric Vehicles: From State of Charge to State of Health

The Transport and Traffic Engineering Laboratory (LICIT-ECO7) is a Joint Research Unit which is placed under the supervision of University Gustave Eiffel and ENTPE (Post-graduate School of Transport and Civil Engineering).

LICIT-ECO7 is also part of the University de Lyon and Lyon Urban School.

The laboratory is located on both the ENTPE Campus and the University Gustave Eiffel Bron Campus in Lyon.

LICIT-ECO7 is part of the JRT GEST (Joint Research Team on Energy Management and Storage for Transport).

LICIT-ECO7’s main research theme is the Dynamic Modelling, Monitoring and Control of Mobility Networks as well as New Vehicles and New Components Usage, Ageing and Control.

Materials
• Historical battery data and Electric Vehicle usage datasets
• Python programming language and libraries such as PyTorch
• Tools for data preprocessing and visualization, such as Pandas and Matplotlib

In the context of electric vehicles (EVs), accurately forecasting battery states is essential for both optimal vehicle performance and an enhanced user experience. Two key battery states include the State of Charge (SOC), which reflects the amount of energy still available for use, and the State of Health (SOH), which indicates the battery's long-term degradation (Luo et al., 2022).

The SOC represents the battery's current energy level as a percentage of its total capacity, which directly impacts the vehicle's remaining driving range and immediate driving decisions. Conversely, the SOH evaluates the battery's condition relative to its original performance, offering valuable insights into its degradation over time and aiding in planning for maintenance or replacement.

Accurate SOC predictions empower users to calculate the remaining range and strategically plan recharging stops, promoting energy-efficient driving and facilitating short-term decisions. Additionally, when linked to electric vehicle usage patterns, SOC estimations could contribute significantly to adapting traffic management strategies, encouraging energy-efficient driving behaviors, and regulating traffic flow based on vehicles' real-time energy states.

This internship aims to explore advanced deep learning techniques for predicting the State of Charge (SOC) of EV batteries using historical usage and operational data (Mayemba et al., 2024; Tian et al., 2023).

The focus includes:

• Utilizing sequential deep learning algorithms to process time-series data.
• Integrating recent attention mechanism architectures to enhance prediction accuracy and scalability.

After achieving accurate SOC predictions, this methodology could be expanded to estimate the SOH of electric vehicle batteries, utilizing long-term data to identify battery degradation patterns and assess lifespan.

Furthermore, the project offers the possibility of progression through a PhD opportunity, enabling comprehensive research and advanced exploration within this domain.

The ideal candidate should preferably have:

• Experience with deep learning and familiarity with sequence modeling and
  time-series analysis.
• Proficiency in Python programming and experience with deep learning
  frameworks, such as PyTorch.
• Interest in energy management, electric vehicles, and urban traffic systems.