Health monitoring algorithms for space application batteries

Abstract

Prototype battery health monitoring algorithms (support vector machine, dynamic neural network, confidence prediction neural network, and usage pattern analysis) were developed and tested on the battery data (voltage, current, temperature, etc.) collected from several 4-amp hour lithium ion (Li-ion) battery cells supplied by United Lithium Systems. The battery data was collected under different operating conditions (storage and charge/discharge cycling under room and 50degC temperatures. The results show that the battery health monitoring algorithms is feasible for determining the health state of a Li-ion cell yielding remaining useful life information to the user.