Optical-sensing-oriented remaining useful life estimation of lithium-ion batteries using ALA-tuned VMD and a BiTCN-AM model

In Proc. SPIE, 2026.09

Lithium-ion battery remaining-useful-life estimation with ALA-optimized VMD decomposition and a bidirectional temporal convolutional attention model, validated across the NASA and CALCE datasets.

PDFVol. 14327 GitHubCODE WebsitePaper CiteBibTeX

Affiliations and publisher

SPIE

Lithium-ion Battery Remaining Useful Life EstimationALA-VMD · BiTCN-AM · NASA / CALCE

0:00 / 0:00

ALA-VMD-BiTCN-AM battery RUL prediction framework — Method · ALA optimization, VMD decomposition, and BiTCN-AM prediction

Standard VMD and ALA-VMD battery capacity decomposition comparison — Adaptive Decomposition · Standard VMD vs. ALA-VMD

NASA B0005 battery capacity and RUL prediction results — NASA B0005 · Capacity trajectory and RUL prediction

CALCE CS2_38 battery capacity and RUL prediction results — CALCE CS2_38 · Cross-dataset validation

0.01137NASA RMSE0 cycleNASA RUL absolute error0.00490CALCE RMSE99.93%CALCE R²