Carbon-Aware Optimization of Battery and Thermal Storage for Residential 24/7 Carbon-Free Electricity Under Dynamic Grid Conditions

This study develops a probabilistic carbon-aware optimization framework for residential 24/7 carbon-free electricity (CFE) under uncertain load, PV generation and dynamic grid carbon intensity. Historical half-hourly monitoring data are represented through KDE-Copula scenario generation, while grid carbon factors are described by time-series decomposition and ARMA-based residual scenarios. A two-stage mixed-integer linear program jointly sizes and dispatches battery energy storage (BESS) and thermal energy storage (TES) by minimizing annualized lifecycle CO2 emissions. The results show that coordinated BESS–TES operation improves PV utilization and avoids carbon-intensive grid imports, especially in winter. In the examined capacity range, increasing BESS capacity from 0 to 100 kWh reduces annual lifecycle emissions from approximately 8600 to 6000 kg CO2 yr−1, corresponding to about a 30% reduction, whereas the marginal benefit of additional TES is constrained by DHW demand and its embodied emissions. Electrical storage is therefore the principal carbon-shifting resource, while a moderately sized TES complements it by moving heat-pump operation toward low-carbon and PV-rich periods. The framework provides a practical basis for carbon-aware design of residential electrification systems.