الفهرس | Only 14 pages are availabe for public view |
Abstract FACTS is a new technology using power electronics for controlling the parameters and structures of power systems for improving the power transfer capability of the system. Thyristor-Controlled Series Capacitor (TCSC) is a senes FACTS device which allows rapid and continuous changes of the transmission line impedance. However, this in turn introduces problems in conventional distance protection. The Static Synchronous Compensator (ST ATCOM) is introduced as a powerful FACTS tool for reactive power compensation. The measured impedance by distance relay at the relaying point in the presence of a STATCOM on the transmission line depends on the controlling parameters of STATCOM and on its installation location. The conventional distance relay characteristics are greatly SUbjected to maloperation in the form of over-reaching or under-reaching the fault point. This thesis proposes an approach based on Artificial Neural Networks (ANN) using the Total Least Square-Estimation of Signal Parameters via Rotational lnvariance Technique (TLS-ESPRlT) for fault type classification and faulted phase selection to be used in the protection of series compensated (TCSC) transmission lines and also for the protection of a transmission line employing STATCOM. The required features for the proposed algorithm are extracted from transient currents and voltages waveforms measured at the substation using TLS-ESPRlT. Since these transient waveforms are considered as a summation of damped sinusoids, TLS-ESPRlT is used to estimate different signal parameters mainly damping factors and frequencies of different modes contained in the signal. Those features can then be employed for fault type classification and faulted phase selection.Two different learning algorithms are used for training the neural network: Back propagation (BP) and Particle Swarm Optimization techniques (PSO). System simulation and results which are presented and analyzed in this thesis indicate the feasibility of using neural networks with TLS-ESPRIT in the protection of series compensated (TCSC) transmission lines and for the transmission lines which using STATCOM. |