Optimal energy storage control policies for the smart power grid

Electric energy storage devices are prime candidates for demand load management in the smart power grid. In this work, we address the optimal energy storage control problem from the side of the utility operator. The operator controller receives power demand requests with different power requirements and durations that are activated immediately. The controller has access to one energy storage device of finite capacity. The objective is to devise an energy storage control policy that minimizes long-term average grid operational cost. The cost is a convex function of instantaneous power demand that is satisfied from the grid, and it reflects the fact that each additional unit of power needed to serve demands is more expensive as the demand load increases. For the online dynamic control problem, we derive a threshold-based control policy that attempts to maintain balanced power consumption from the grid at all times, in the presence of continual generation and completion of demands. The policy adaptively performs charging or discharging of the storage device. The former increases power consumption from the grid and the latter satisfies part of the grid demand from the stored energy. We prove that the policy is asymptotically optimal as the storage capacity becomes large, and we numerically show that it performs very well even for finite capacity. The off-line problem over a finite time horizon that assumes a priori known power consumption to be satisfied at all times, is formulated and solved with Dynamic Programming. Finally, we show that the model, approach and structure of the optimal policy can be extended to also account for a renewable source that feeds the storage device.