Software

BMS Estimation ESTIMATING BATTERY STATES AND LIFETIME AT THE EDGE IS A CHALLENGING TASK, PARTICULARLY WITH CHEMISTRIES LIKE LFP. Accurate estimation is critical for each cell, as the overall performance of the pack depends on its weakest cell. Historically, state estimation requirements for automotive, mobility, and energy storage applications have been around 3% for NMC and 5% for LFP for State of Charge SoC, and under 5% for State of Health SoH. However, recent expectations demand accuracy better than 2% for both SoC and SoH across chemistries like NMC and LFP. Meeting these stringent requirements has driven the adoption of advanced physical models and data-driven approaches, leveraging the latest developments in AI to achieve unprecedented precision.

Graphic sourced from 1

Sources: 1 Accelerating eMobility with Automotive Grade AI for Batteries - EE Times, Exploring next-generation AI battery management systems with Infineon and Eatron technologies

Extracted from the PDF; charts and images may read imperfectly.