Self-Adaptive Autoreclosing Scheme usingI Artificial Neural Network and Taguchi's Methodology in Extra High Voltage Transmission Systems

Conventional automatic reclosures blindly operate for permanent, semi-permanent or
transient faults on an overhead line without any discrimination after allowing some
estimated time delay. Reclosing onto a line with uncleared fault often results in, not
only loss of stability and synchronism but also damage to system equipments, as a
consequence. The thesis focuses on methods to discriminate a temporary fault from a
permanent one, and accurately determine fault extinctiontime in an extra high voltage
(EHV) transmission line in a bid to develop a self-adaptive automatic reclosing
scheme. The fault identification prior to reclosing is based on optimized artificial
neural network associated with three training algorithms, namely, Standard Error
Back-Propagation, Levenberg Marquardt and Resilient Back-Propagation algorithms.
In addition, Taguchi's methodology is employed in optimizing the parameters of each
algorithm used for training, and in deciding the number of hidden neurons of the
neural network. To get data for training the neural networks, a range of faults are
simulated on two case studies -single machine -infinite bus model (connected via
EHVtransmission line) and a benchmark IEEE 9-bus electric system. The spectra of
the fault voltage data are analyzed using Fast Fourier Transform, and it has been
found out that the DC, the fundamental and the first four harmonic components can
sufficiently and uniquely represent the condition of each fault. In each case study, the
neural network is fed with the normalized energies of the DC, the fundamental and
the first four harmonics of the faulted voltages, effectively trained with a set of
training data, and verified with a dedicated testing data obtained from fault voltage
signals generated on IEEE 14-bus electric system model. The results show the
efficacy of the developed adaptive automatic reclosing scheme. This effectively
means it is possible to avoid reclosing before any fault on a transmission line (be it
temporary or permanent) is totally cleared.