Abstract-A two-level real-time voltage control scheme is proposed to keep voltages within specified limits in distribution grids using Distributed Generation Units (DGUs). It combines a local and a centralized control of their reactive powers. The local control provides fast response after a disturbance, reducing its impact and enhancing voltage quality. The centralized control uses measurements collected throughout the network to bring the voltages inside tighter limits and balance the various DGU contributions. To this purpose, it adjusts in a coordinated way their reactive power set-points, taking into account the local controls. This discrete control solves at each time step a multitime-step constrained optimization inspired of Model Predictive Control. The method effectiveness is demonstrated on a 75-bus test system hosting 22 DGUs.

This paper presents a real-time, centralized control system acting on the active and reactive powers of distributed generators when the network experiences voltage and/or thermal limits violation. The control resorts on multi-step receding-horizon optimization. The objective is to minimize the deviations of Dispersed Generation Units (DGU) active and reactive powers from reference values. The reactive power corrections have priority over the active ones. Furthermore, the formulation is such that DGU powers are restored to their desired schedule as soon as operating conditions allow doing so. Three modes of operation of the proposed controller are presented, involving dispatchable units as well as DGUs operated to track maximum power output. The effectiveness of the proposed control is illustrated through detailed simulations of a 75-bus, 11-kV system hosting 22 DGUs.

Abstract-A real-time control scheme is proposed to correct undesired voltages in distribution networks by adjusting the reactive powers produced by Dispersed Generation Units (DGUs). This is performed at two levels: locally, in DGU controllers, and centrally by a supervisory controller acting on some of the DGUs only. The centralized, discrete controller solves a multitime-step constrained optimization inspired of Model Predictive Control. Unacceptable voltage deviations are partially but rapidly corrected by the local controls. Further but slower corrections are applied by the upper level, taking advantage of the coordination between DGUs. This hybrid control is of practical interest where the deployment of the upper level control is not feasible or affordable over all DGUs. The proposed method effectiveness is demonstrated on a 75-bus test system.

No abstract