Researchers have developed two novel approaches to improve battery health diagnostics and design. The first, RoSIP-Batt, uses a physics-guided Transformer network to jointly predict State of Health (SOH) and Remaining Useful Life (RUL) from charging profiles, achieving significant error reductions on benchmark datasets. The second framework employs physics-informed learning with virtual sensing to infer hard-to-measure battery design parameters from standard BMS measurements, enabling more informed battery design and reducing prediction errors. AI
IMPACT These advancements could lead to more reliable and longer-lasting batteries, crucial for the widespread adoption of electric vehicles and grid-scale energy storage.
RANK_REASON Two academic papers published on arXiv detailing new AI/ML approaches for battery diagnostics and design.
- arXiv
- digital twin
- lithium-ion batteries
- machine learning
- physics-informed learning
- solid-state diffusion coefficient
- vehicle-to-grid
- virtual sensing
- Huazhong University of Science and Technology
- MIT-Stanford
- Nasa
- RoSIP-Batt
- Rotary Position Embedding
- Transformer++
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
