Voltage Control in Active Distribution Networks Under Uncertainty in the System Model: A Robust Optimization Approach

Within the context of ancillary services for active distribution networks (ADNs), application of intelligent control techniques is required in order to achieve specific operation objectives. Despite their differences, most control mechanisms proposed in the literature rely on the assumption that the distribution network operator has an accurate and up-to-date model of the network topology and a complete knowledge of the line parameters, i.e., a correct network admittance matrix Y. However, this assumption does not always hold in reality due to both an incomplete knowledge of the grid asset and/or a physical change of the line parameters. In this paper, we consider the problem of optimal voltage control in ADNs under uncertain, but bounded, line parameters with no assumptions on the parameter's uncertainty distribution. In particular, availability of a monitoring infrastructure is assumed and the goal is to control the active and reactive power injections of a number of distributed generators connected to the network buses in coordination with the transformers on-load tap changers. The optimal control problem is formulated as a mixed-integer linear problem by means of sensitivity coefficients and a robust optimization framework is used in order to account for the uncertainties in the network admittance matrix. In order to estimate the benefits of the proposed method, the evaluation of the algorithm is carried out by using both the IEEE 13- and the IEEE 34-nodes test feeder.