Generation and energy storage planning decomposing complexities in modeling networks, binary variables and uncertainties

As the loads and uncertainties of power systems keep increasing, and energy storage (ES) becomes more affordable, it is more important to have proper generation and ES planning with high modeling accuracy. In consideration of complexities from power flows, binary variables and uncertainties, a novel planning model with three decision steps is proposed to avoid exponentially increasing computational difficulties caused by the direct superposition of different complexities. To lower the requirement on network modeling by utilizing properties of specific power systems, second-order conic power flow models are used first to acquire information about possible operating conditions. After that, tailored linear power flow models are established specifically to achieve accurate network modeling at low complexities, which enables more precise uncertainty modeling. Besides, based on the complementary features of generators and ES in terms of functions and installation flexibilities, complexities from binary variables and uncertainty modeling are decomposed to further improve the accuracy of uncertainty modeling and obtain proper planning decisions. Comprehensive case studies verify the effectiveness of the proposed model in achieving accurate modeling of both networks and uncertainties at the same time of maintaining computational tractability.