Physics-informed neural network improves battery health prediction

Researchers have developed PiDDM, a novel framework that integrates physics-informed neural networks with degradation kinetics to improve lithium-ion battery state-of-health predictions. By incorporating empirical Arrhenius degradation kinetics related to solid electrolyte interphase growth and lithium inventory loss, PiDDM encourages physically consistent capacity fade. Evaluations on a public dataset demonstrated that PiDDM achieved lower prediction errors compared to a multilayer perceptron and a baseline physics-informed neural network, particularly in long-term extrapolation scenarios where it avoided nonphysical capacity regeneration. AI

IMPACT This research offers a more accurate and physically consistent approach to predicting battery health, potentially improving the reliability of energy storage systems.

RANK_REASON The cluster contains an academic paper detailing a new modeling framework for battery health prediction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Arrhenius degradation kinetics
• arXiv
• lithium-ion battery
• multilayer perceptron
• Physics-Informed Neural Network
• PiDDM
• solid electrolyte interphase

• paper
• model release
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →