Robust Voltage Control using Demand-Side Flexibility with Time and Spatial Dependency

The installation of renewable energy sources (RES) causes voltage deviation at distribution grids (DGs). To address it, the coordinated DG operation of with demand-side flexibility with considering uncertainty of RES output has been investigated. Centralized control for on load tap changers (OLTCs) has also attracted attention whose tap positions can be adjusted after the real RES output is revealed. However, the previous studies on robust optimization (RO) do not consider the centralized controlled OLTCs. Additionally, they assumed that the RES outputs among different places and time steps are independent, and this assumption causes over-conservativeness, leading to high DG operation costs. We propose a novel RO model for voltage control with centralized controlled OLTCs and demand-side flexibility. We also constructed uncertainty sets of RES output considering the time and spatial dependency between neighboring time steps and places to address the issue of over-conservativeness. A solving algorithm for the proposed RO model is designed on the basis of the mountain climbing algorithm and modified column-and-constraint generation. A numerical experiment was conducted with a modified IEEE 34 bus feeder model, and its result indicates that the proposed RO model can achieve voltage stabilization with a lower DG operation cost than the conventional model.