Lithium-Ion Cell Sorting and Cell Performance Modeling for Spacecraft Battery

The Earth orbiting spacecrafts use lithium-ion batteries as an energy storage device. Each battery is made up of several series-and parallel-connected cells. The constraints concerning the cell parameters, obtained by testing, drive the positioning of the cells in the battery. Positioning by cell sorting to meet the required battery performance can be approached through the genetic algorithm (GA). A GA improved cell sorting algorithm (GICSA) is proposed in this work. It sorts the cell parameters randomly with GA to match them within a given set of tight constraints. It sorts different cell types and numbers of cells for different battery configurations. It gives the practically best possible number of matched cells for the battery. Even the closely matched cells in the battery exhibit slightly different performance and experience different temperatures during their use. Due to prohibitive costs, the spacecraft application does not have an option of repair or replacement once it is in space. Hence, there is a need to understand the behavior of cells mounted within a battery. A nonlinear regression model is developed to estimate cell performance in the battery (CPB) used for spacecraft applications. The CPB model is validated with experimental results and is within a root mean square error of 2% for the voltage.