Protection of Power Distribution Systems with Distributed Generation: Islanding Detection and Protection Coordination

Abstract

Microgrids can be operated either grid-connected to reduce system losses and for peak shaving or islanded to increase reliability and provide backup power during utility outage. Such dual configuration capability poses operational and safety issues, and also presents challenges in the design of the protection system since fault current magnitudes will vary depending on the microgrid operating mode. This work investigates two key protection issues involved with the installation of distributed generation in power distribution systems. The first part of this thesis presents a new passive islanding detection method for grid-connected inverterbased distributed generation (DG) systems is proposed. A statistical signal processing algorithm known as Estimation of Signal Parameters via Rotational Invariance Techniques (ESPRIT) is used to extract new features from measurements of the voltage and frequency at the point of common coupling (PCC) as islanding indicators. The new features are defined based on a damped-sinusoid model for power system voltage and frequency waveforms, and include modal initial amplitudes, oscillation frequencies, damping factors and initial phases. A set of training cases generated on the IEEE 34-bus system was used to train a na¨ıve-Bayes classifier that discriminates islanding and non-islanding events. Cross-validation was used to evaluate the performance of the proposed islanding detection method. The results show that by using the new features extracted from ESPRIT, the classifier is capable of discriminating islanding and non-islanding events with an accuracy close to 100%. In the second part of this thesis, a microgrid protection scheme that relies on optimally sizing fault current limiters (FCL) and optimally setting directional overcurrent relays is proposed. The protection scheme is optimally designed taking into account both modes of operation (grid-connected and islanded). The problem has been formulated as a constrained non-linear programming (NLP) problem and is solved using the genetic algorithm (GA) with the static penalty constraint-handling technique. The proposed approach is tested on a typical radial distribution system and on the IEEE 30 bus looped power distribution system equipped with synchronous-based distributed generation (DG).

Date of Award	2011
Original language	American English
Supervisor	Hatem Zein El din (Supervisor)