In April 2016 Manchester eScholar was replaced by the University of Manchester’s new Research Information Management System, Pure. In the autumn the University’s research outputs will be available to search and browse via a new Research Portal. Until then the University’s full publication record can be accessed via a temporary portal and the old eScholar content is available to search and browse via this archive.

Related resources

Search for item elsewhere

University researcher(s)

Academic department(s)

Online Prediction of the Post-Disturbance Frequency Behaviour of a Power System

Wall, Peter Richard

[Thesis]. Manchester, UK: The University of Manchester; 2013.

Access to files

FULL-TEXT.PDF (pdf)

Abstract

The radical changes that are currently occurring in the nature of power systems means that in the future it may no longer be possible to guarantee security of supply using offline security assessment and planning. The increased uncertainty, particularly the reduction and variation in system inertia that will be faced in the future must be overcome through the use of adaptive online solutions for ensuring system security. The introduction of synchronised measurement technology means that the wide area real time measurements that are necessary to implement these online actions are now available.The objective of the research presented in this thesis was to create methods for predicting the post-disturbance frequency behaviour of a power system with the intent of contributing to the development of real time adaptive corrective control for future power systems. Such a prediction method would generate an online prediction based on wide area measurements of frequency and active power that are recorded within the period of approximately one second after a disturbance to the active power balance of the system. Predictions would allow frequency control to respond more quickly and efficiently as it would no longer be necessary to wait for the system frequency behaviour to violate pre-determined thresholds.The research presented in this thesis includes the creation of an online method for the simultaneous detection of the time at which a disturbance occurred in a power system, or area of a power system, and the estimation of the inertia of that system, or area. An existing prediction method based on approximate models has been redesigned to eliminate its dependence on offline information. Furthermore, the thesis presents the novel application of pattern classification theory to frequency prediction and a five class example of pattern classification is implemented.