Lie Generator Networks turn 60-second pulse data into battery diagnostics

Electrochemical impedance spectroscopy (EIS) is the gold standard for lithium-ion battery diagnostics, revealing mechanism-specific degradation invisible to standard voltage and current measurements. But EIS requires expensive dedicated hardware and minutes-long acquisitions, making it impractical for field deployment. Now, a new paper from Shafayeth Jamil and Rehan Kapadia introduces Lie Generator Networks (LGN), an AI framework that extracts the same diagnostic and prognostic information from just 60 seconds of post-pulse voltage relaxation—data that battery management systems already collect. LGN is a structure-preserving identification framework that learns the generator matrix of relaxation dynamics with stability guaranteed by its architecture, yielding time constants precise enough to resolve electrochemical variation that conventional curve fitting cannot detect from identical data.

Across five datasets totaling over 850 cells from four institutions and multiple chemistries, LGN tracks degradation with near-perfect rank correlation (|ρ_s| = 0.999), enables cross-validated reconstruction of full Nyquist spectra at 2% median error across 227 cells, predicts which capacity-matched cells fail first from three early diagnostics, and recovers Arrhenius activation energies without any physics priors, retraining, or cell-specific tuning. Critically, LGN requires no training data, no impedance hardware, and no chemistry-specific calibration, converting any existing relaxation pulse into an impedance-grade diagnostic. This opens the door to real-time health monitoring, rapid second-life grading, production-line quality control, and physics-informed prognosis from minutes of measurement.