SandboxAQ, in collaboration with the US Army’s DEVCOM C5ISR Center, has achieved a significant breakthrough in battery health analytics.
The team has completed the first full-scale calendar-aging campaign, combining deep-cycle operational data with rigorous storage testing across various lithium-ion chemistries and formats, including cylindrical and high-energy pouch cells.
This effort produced a vast dataset reflecting diverse storage durations and temperatures, aligned with the Army’s demanding shelf-life and performance standards. Cells underwent alternating store-then-cycle protocols to simulate real-world deployment and redeployment, generating millions of hours of dQ/dV and impedance data.
The result is a detailed understanding of how temperature, state of health (SOH), and operational profiles affect capacity loss, charge retention, and remaining useful life (RUL). This data powers SandboxAQ’s latest Large Quantitative Models (LQMs) – AI-driven tools that predict capacity fade from a single Reference Performance Test (RPT).
Internal validation shows the models achieving less than 1% error over 18-month horizons, slashing traditional shelf-life qualification timelines by over 90%. The models also enable rapid RUL screening, helping to avoid premature disposal and identify risks before deployment.
Looking ahead, integration into field chargers will allow real-time battery health monitoring and replacement decisions, enhancing mission readiness and reducing logistical strain.
“The Army’s tactical edge depends on batteries that perform flawlessly after years on the shelf and months in the field,” said Dr Ty Sours, senior research scientist at SandboxAQ. “We can now deliver actionable battery health and logistics insights in days – not years – while slashing sustainment cost and risk.”
