Online Detection of Disturbances in Power System using Least Square Estimation

Abstract

A stable, uninterrupted and reliable power supply is the main objective of the power system control system. With increase in energy demand the power system has become more exposed to fluctuations and disturbances. Thus, it has become very necessary to find out ways for online detection of any abrupt change in the power system which consumes lesser amount of time so that its aftereffect can be minimized. With use of Phasor Measurement Units (PMU), which can send data at a very faster rate in a synchronized manner, we can study the load pattern, voltage profile and other parameters to detect any abrupt change in the power system.
The methods that were proposed earlier for disturbance detection are time consuming and are scanning each and every bus continuously. The main motive of the work is to develop a disturbance detection technique for power system using the synchronized data of PMU which is less time consuming. We have also tried to develop a technique in which the total number of buses that are observed or monitored continuously is reduced. Such critical buses can be monitored continuously so that any disturbance in the nearby buses can be reflected in these buses and necessary steps can be taken as soon as possible.
In this thesis we have used the method of least square estimation to detect the disturbances occurring in power system online. The least square estimation of voltage magnitude of buses is done. Then the statistical method is used for detecting the large disturbances in the system. Any disturbance at a particular bus is reflected in the nearby buses in the form of change in voltage magnitude or change in angle. This concept is used to reduce the number of buses that is to be monitored continuously, forming a set of critical measurements.