In a compact unmanned-aerial-vehicle (UAV) battery pack, the interior cells become hotter than the perimeter cells and therefore age faster. A series string is limited by its weakest cell, and this aging non-uniformity is fixed once the hardware is built. The module layout and the active-balancing circuit are usually selected separately, so this effect remains outside the design loop. In this work, we treat the layout, the balancing-topology family, and the conductor sizing as joint pre-fabrication variables, enumerate the small admissible design space, and score every candidate by a coupled electro-thermal-balancing-aging simulation, so that all designs are ranked under identical assumptions. On a 24-cell NMC 21700 string, the algebraic connectivity π 2 predicts the balancing speed well ( π π = β 0 . 8 5 ), but it has no clear correlation with the long-term fade uniformity ( π π = + 0 . 0 6 ). The uniformity is determined by the layout, through the fraction of cells on the cooled perimeter. A single thermal index of the layout collapses the fade spread of all layouts, of a wide interior-cooling sweep, and of the 12Sβ48S pack sizes onto one affine law ( π 2 = 0 . 8 9 ). Therefore, the high-connectivity cell-to-pack circuits are the heaviest, but they are no more uniform, and they are dominated. A light topology on the most-cooled layout produces about 28% smaller fade spread at a lower mass, and both gains are kept under closed-loop aging feedback. The first-cell life is about 11% longer in the open-loop case, but this value is an upper bound that the feedback reduces toward parity.
No takes yet. Share an insight, caveat, or question.
Ghaeminezhad et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: