Central reactive power control for smart low-voltage distribution grids

Abstract

Due to the high penetration of renewable distributed energy resources (DERs) into low-voltage distribution grids, particularly residential photovoltaic (PV) systems, a renewed interest in voltagecontrol strategies arises in the recent years. Although voltage-control strategies based on reactive power produced by smart PV inverters are a well developed technique, the use of central control strategies for LV distributed grids remains a challenging procedure because of the characteristics of these grids and the limitations of telecommunication networks, including the low measurement redundancy of low-voltage (LV) distribution system compared to power transmission systems. However, with the inclusion of Information and Communication Technology (ICT) components into the LV distribution grid, measurement data from smart meters are available for the control of these networks. This dissertation undertakes a central voltage-control strategy for smart LV distribution networks, using a novel optimal power flow (OPF) methodology in conjunction with the information collected from smart meters for the power flow calculation. The proposed strategy can simultaneously mitigate the PV reactive power fluctuations, as well as minimize the voltage rise and power losses. Moreover, a novel distribution power system state estimation method (DSE) based on the information provided by smart meters is developed to support central voltage-control strategies in real-time in case of telecommunication failure. In order to analyze the developed techniques, a more realistic scenario of smart LV distribution networks is designed, including high-resolved home power demand models and a novel approach for the power flow formulation of LV distribution systems. Comparison studies to validate the developed methodologies are presented for IEEE 30-bus, 57-bus system and a 30 smart home LV distribution system, showing the robustness of proposed methods compared with traditional approaches. The results are promising as voltage control is achieved fast and accurately, the reactive power is smoothed in reference to the typical optimization techniques and local control strategies. Furthermore, using the information provided by the smart meters in combination with the developed power flow formulation, the proposed central control system is able to mitigate the effects of temporal communication link failure from smart meters to utility company. To illustrate the applicability of the proposed method, a real-time digital simulator for smart distribution power grids with a central voltage-control system are used. The results show that the proposed method is effective to estimate the missing values in the case of communication problems between smart meters and the central control system, even for ill-conditioned systems.