FPGA-based implementation of lithium-ion battery SOH estimator using particle filter

Lithium-ion battery has already become the most widely applied energy storage system in various industrial scenarios. The state of health (SOH) estimation and prediction are vital functions in the advanced battery management system (BMS). However, those SOH estimators are always with high computing complexity, while the BMSs are always implemented based on the embedded processor. The contradiction between limited computing ability and complex computing process directly restricts the implementation of various SOH estimators. In this term, this paper introduces the field programmable gate array (FPGA) as the controller of the advanced BMS. A lithium-ion battery SOH estimator fused with on-line measurable degradation features and battery degradation empirical model is developed based on particle filter (PF) algorithm. The computing process of the PF algorithm is paralleled to make the SOH estimator model much more suitable for the FPGA processor. The experimental results illustrate that the proposed SOH is with high accuracy and robustness and also performed great computing ability with low power consumption.