Feasibility Study: Autonomous State Estimation in Distribution Systems

Abstract

We propose an autonomous state estimation based on robotic concepts and advanced state estimation methods. An autonomous, intelligent monitoring infrastructure is proposed that reliably and automatically detects devices as they are plugged-in or -out; it identifies changes in system state and automatically updates the real-time model of the system. The real-time model is used for control, operation, and optimization of the system via application software that are not addressed in this paper. The proposed infrastructure uses modern intelligent electronic devices (IEDs) named universal monitoring protection and control units (UMPCUs) that are capable of handling three key data sets for the component that they are attached to: 1) connectivity, 2) device model, and 3) measurements. The connectivity data represent the connecting points where a device is connected to the power grid, the device model data provide the mathematical model of the device as an object, and the measurements provide the numeric values of physical quantities such as voltages and currents captured by the data acquisition system. This paper describes a feasibility study of the proposed infra structure on a scaled-down three-substation power system laboratory setup. Performance metrics are provided that quantify total time latencies on IEC61850 implementation.