Battery energy storage systems (BESS) are critical infrastructure for grid flexibility and renewable integration, but their operational economics depend heavily on understanding how individual battery subsystems age differently over time. A new research framework addresses a persistent gap: most optimization models treat battery packs as uniform assets, ignoring the reality that strings and modules degrade heterogeneously due to variations in thermal conditions, charge cycles, and manufacturing tolerances.
The study evaluated four operational scenarios for an energy arbitrage application, ranging from simple models that ignore aging to sophisticated frameworks that account for subsystem-level degradation and its cost implications. Results demonstrate that the fully informed scenario—combining precise string-level modeling with aging costs embedded in the optimization objective—achieved 21% higher revenue per unit of state-of-health loss compared to baseline approaches.
More critically, ignoring subsystem heterogeneity can produce dispatch schedules that are technically infeasible or economically suboptimal. When aged subsystems are treated identically to fresh ones, the optimizer may generate power schedules that exceed safe operating limits for degraded strings, requiring costly real-time adjustments or curtailment.
The framework accounts for capacity losses, operational mismatches, and missed revenue opportunities—a comprehensive view often absent in traditional BESS optimization. For operators running systems over extended periods (5–10+ years), modeling aging variability transitions from technical refinement to business necessity.
These findings have immediate implications for BESS developers, aggregators, and grid operators. As battery fleet ages and competition intensifies, the ability to extract maximum value from heterogeneous assets will drive competitive advantage. Integrating degradation awareness into control algorithms enables smarter dispatch decisions that balance short-term arbitrage returns against long-term capacity preservation—a trade-off that simple models cannot optimize.
