A novel application-aware retired lithium-ion batteries regrouping approach to enable improved second life

Amidst increasing environmental concerns, the transport sector is undergoing a substantial transformation leading to a larger market share of electric vehicles and, in turn, a growing number of retired lithium-ion batteries. The reuse of these batteries in second-life applications can improve circularity by reducing environmental impacts and costs. Nevertheless, the performance of Second Life Batteries (SLB) is compromised by cell-to-cell variations, particularly in parallel-connected blocks, caused by the early-stage of the screening and grouping methodologies at end-of-life. This paper contributes to address SLB performance disparities by proposing an innovative application-aware cell (re-)grouping methodology capable of optimising of series–parallel modules creation using Particle Swarm Optimisation (PSO) and Genetic Algorithms (GA). The a posteriori grouping demonstrate superior performance over conventional application-agnostic screening and clustering methods by significantly reducing modules current (−34.4%) and thermal (−20.2%) gradients, and enhancing their Remaining Useful Life (RUL) (+22.7%) and capacity (+1.0%). The retired batches’ exploitation is also improved, with a lower scrap rate and higher number of high-performance modules, potentially positively impacting the cost and reducing the environmental footprint of SLB solutions. The study not only underscores the framework’s potential in optimising SLB lifecycle performance but also paves the way for a shift in the design paradigm towards a whole lifecycle perspective, welcoming future research to validate and refine these methodologies. • An innovative retired lithium-ion cell (re-)grouping methodology is proposed. • Second-life battery modules are optimised via metaheuristics and digital twins. • Modules’ current and thermal gradients are reduced, with RUL and capacity enhanced. • Scrap rate of retired cells is lowered, increasing high-performance module count. • Circular economy of batteries could benefit from a posteriori grouping methods.