Power-distribution grids consist of assets such as transformers, cables, and switches, of which the proper utilization is essential for the provision of a secure and reliable power supply to end customers. Distribution-system operators (DSOs) are responsible for the operation and maintenance of these assets. Due to the increased use of renewable sources such as wind and solar, grid assets are prone to operation conditions outside safe boundaries, such as overloading, large voltage unbalance, and a rise in voltage. At present, distribution grids are poorly monitored by DSOs, and the above-mentioned problems may thereby go unnoticed until the failure of a critical asset occurs. The deployment of smart meters in distribution grids has enabled measurements of grid variables such as power, current, and voltage. However, their measurements are used only for billing purposes, and not for monitoring and improving the operating condition of distribution grids. In this paper, a state-estimation algorithm is proposed that utilizes smart-meter data for offline analysis, and estimates the loading of grid assets and power losses. Single- and three-phase state-estimation algorithms are compared through simulation studies on a real-life low-voltage distribution grid using measured smart-meter data. The three-phase state-estimation algorithm based on the nonlinear weighted least-squares method was found to be more accurate in estimating cable loading and line power losses. The proposed method is useful for DSOs to analyze power flows in their distribution grids and take necessary actions such as grid upgrades or the rerouting of power flows.
The automatic phase-selective photovoltaic grid-connection system used to control three-phase imbalance of low voltage power grid
