State and Line Parameter Estimation in Three Phase Distribution System

Abstract

The necessity for safe and stable techniques for monitoring, control, and operation of a distribution system has become increasingly important in the last few decades. The load flow analysis and state estimation are the most crucial components in monitoring and operational studies. Moreover, load flowstudies and state estimation are distinct methodologies utilized in transmission/ distribution system. The load flowanalysis is a method that gives a deterministic study of the network under steady-state conditions. In contrast, state estimation provides the most accurate estimates of the real operating condition of the distribution network while accounting for the presence of noisy as well as bad measurements. The architecture and radial topology of the distribution network differ significantly from those of the transmission network. Because of unbalanced nature of the distribution system, the state estimation (SE) algorithm originally intended for the transmission system needs to be suitably modified for the distribution system. The distribution system is usually unbalanced in nature configuration and has short feeder sections with higher R/X ratio resulting in higher I2R losses. Unlike transmission network, the distribution system is characterized by its vast geographical spread and limited realtime measurement which can lead to significant observability challenge. Despite the recent implementation of advanced metering infrastructure (AMI), the installation of smart meters at all nodes is not economically feasible. State estimation technique usually assumes that the line parameters values of a distribution network are exactly known. Nevertheless, this assertion is inaccurate, as the operating values of the line parameter can vary over a period of time due to factors such as ambient temperature, and conductor degradation. Seasonal fluctuations in soil resistivity and temperature are critical environmental factors that substantially influence the values of line parameters. In addition, human errors such as misreading, inaccuracies in the database input, and various measurement methods can lead to variations in the network parameters.