Optimization Algorithms for Power Flow and Energy Storage in Smart Grid

Abstract

In smart grid, the penetration of two-way flows of electricity and information makes it capable of integrating distributed renewable energy sources (RES) and a large number of demand side users much more effectively. However, as a crucial part of smart grid, the renewable energy is intermittent and very difficult to predict, which then results in the inconsistent power in the grid. As a strategy to deal with the intermittent RES and provide demand response benefits in smart grid, integrating energy storage in the system introduces new features to the optimal power flow (OPF) problem. Therefore, in this thesis, we study the optimal power flow problem with energy storage dynamics in purely resistive power networks. We first formulate it as a dynamic nonconvex OPF problem. Then we propose an efficient second order cone programming (SOCP) relaxation for this nonconvex problem. Next, by exploiting a recently-discovered zero duality gap property in the OPF problem, we apply the optimization decomposition techniques to the problem and propose efficient algorithms to obtain the global optimal solution using distributed message passing algorithms. The decomposition methods offer new interesting insights on the equilibrium load profile smoothing feature over space and time through the relationship between the optimal dual solution in the OPF and the energy storage dynamics. We investigate the energy storage dynamics under different problem settings and verify that the distributed algorithms can converge fast to the global optimal solution by numerical simulations in IEEE test systems.
Although these decomposition algorithms are very efficient, they are offline algorithms, where the future demand information is needed. However, this type of information is not available before the end of the whole period in reality. Thus, in this thesis, we also develop a simple online algorithm where the knowledge of the future demand information is not necessary. Leveraging the zero duality gap property in the OPF problem, we first characterize the optimal solution of the OPF problem and decomposed problems which are decoupled over time. Then, based on these results, we propose a distributed online algorithm for solving the OPF problem, where no precise knowledge of future demands are required in contrast to the existing approach. We use the competitive ratio to show the theoretical performance of the online algorithm. Through extensive simulations in the IEEE 14-bus system, we show that the algorithm can converge fast to the optimal solution. Moreover, the actual ratio between the online algorithm result and the offline optimal value is less than 1.2 by properly choosing the system parameters.

Date of Award	25 Jan 2016
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Chee Wei TAN (Supervisor)