A Closed-Loop Hybrid Stochastic-Robust Framework for Health-Aware PV–BESS Sizing

Health-aware sizing of photovoltaic (PV) and battery energy storage systems (BESS) in behind-the-meter (BTM) facilities must reconcile expected-cost efficiency with worst-case reliability. We propose a closed-loop hybrid stochastic-robust optimization (SRO) framework that fuses asynchronous PV, electricity-price, carbon-intensity, and load data via Decoupled–Recombined Sampling into typical and critical scenario sets. The framework co-optimizes economic and carbon objectives over typical scenarios while enforcing strict feasibility across critical ones, validated by a high-fidelity electrochemical simulation. On a representative case, it incurs only a 0.9% realized cost premium over stochastic programming (vs. 10.1% for robust optimization) with zero grid violations and lower cycle-induced aging.