.png){kind=icon}
.png){kind=icon}
Return
An efficient physics-informed continuous-time model for battery state estimation under power-only runtime observability
.png){kind=icon}
PRE
AI
DOI：10.1016/j.est.2026.123528.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Physics-informed method estimates battery state using only power measurements. • Charge conservation ensures physical consistency without runtime voltage sensing. • Time-to-empty prediction incorporates Monte Carlo uncertainty quantification. • Sub-millisecond latency (0.5 ms) and O(1) memory indicate edge feasibility. • Achieves 1.13% SOC RMSE across 7565 records via chemistry-adapted validation.