Stochastic Multi-Objective Optimization for Grid-Scale Energy Storage Sizing and Siting

The integration of variable renewable energy sources reduces greenhouse gas emissions but introduces significant operational challenges to modern power systems. Grid-scale energy storage systems provide the essential flexibility required to mitigate these short-term uncertainties. Furthermore, these technologies enable a more economical system operation and actively assist in reducing overall greenhouse gas emissions. This paper proposes a stochastic multi-objective formulation for the optimal sizing and siting of grid-scale energy storage systems. The model simultaneously minimizes the total expected system costs and the expected thermal emissions. The Augmented $\varepsilon $ -Constraint (AUGMECON) method is implemented to generate the exact Pareto frontier. The proposed stochastic framework captures load and renewable generation uncertainties using representative daily scenarios. The mathematical formulation fully integrates the strict unit commitment constraints of the conventional thermal fleet. Case studies are conducted on the modified IEEE 24-bus Reliability Test System. The results demonstrate how the trade-off between financial savings and environmental priorities fundamentally alters the optimal spatial allocation and capacity requirements of the energy storage systems.