Planning of energy storage systems in power networks: Improving reliability of two-stage robust optimization algorithms

The usage of renewable energy sources such as wind and solar in power networks has increased in recent years, introducing uncertainty and variability into the system. Energy storage plays a critical role in mitigating these problems. Thus, placement, sizing, and operation of energy storage units constitute important problems and there are many related studies. In this thesis, the problem of placement of energy storage systems in power networks is studied by modeling it as a two-stage stochastic robust optimization. A hybrid method is used, combining deterministic equivalent for the stochastic part and column and constraint generation for the robust part, to solve the problem. Two new algorithms are proposed for the solution of the second-stage max-min optimization. The proposed methods are applied to the 6-bus, IEEE 14-bus, and IEEE 30-bus systems, and numerical analyses are performed to compare the proposed methods with other methods from the literature.