Abstract
Motivated by the global environmental concern, renewable generation from in-termittent sources (such as wind and solar) grows quickly and will constitute a large portion of the total electricity supply. However, the higher penetration rate of the renewable generation introduces more uncertainty and variability to the grid. To mitigate and reduce the variability of the random renewable generation, energy storage system is widely used and studied. Energy storage has been recognized as an environmentally friendly candidate that can provide both stability and flexibility for the operation of the next-generation power sys-tem. As a result, it is becoming increasingly important for the operators to jointly optimize the renewable generation system and the energy storage sys-tem.This thesis focuses on the two topics concerning the operation of the system with both renewable generation and energy storage: (i) How to find a com-putable optimal policy for the system if such policy exists. (ii) If no computable optimal policy exists, can we find some priority rules or approximation meth-ods to assist the operator?Regarding the first topic, we study two problem in this thesis. We consider the customer owns intermittent renewable generation and faces (possibly ran-dom) electricity prices and different types of AC/DC load in Chapter 2; and the customer owns renewable generation and faces (possibly random) fluctuating electricity prices and demand charge in Chapter 3. In both problems, we provide a complete characterization on an optimal threshold policy, and implement the characterized optimal policy in realistic settings with random renewable gener-ation and electricity prices.For the second topic, we study the joint scheduling of EV charging and stor-age operation in Chapter 4 and characterize an important priority rule for an optimal scheduling policy. Furthermore, we study the segregated linear deci-sion rules for distributionally robust control with linear dynamics and quadratic cost and apply the proposed method to approximately optimize a system with both the renewable generation and the energy storage in Chapter 5.