This paper proposes a security algorithm based on
thewavelet-alienation-neural technique for detecting, classifying, and
locating faults on Thyristor-Controlled Series compensator (TCSC)
compensated lines. A fault index has been calculated using wavelet
transform and alienation coefficients with post-fault current signals
measured/ sampled for quarter cycle time at both near and far end buses
for fault detection and classification. The location of the fault is
predicted using an Artificial Neural Network (ANN) after the fault has
been diagnosed. Approximate coefficients (quarter cycle time) of both
voltage and current signals, from both buses, were provided as input to
ANN. Various case studies, such as variations in TCSC position, fault
location, sampling frequency, power flow path, incipient angle of fault,
TCSC control strategy, fault resistance, and load switching conditions,
have verified the robustness of the proposed safety system.
Optimization-Based Formulations for Short-Circuit Studies with Inverter-Interfaced Generation in PowerModelsProtection.jl
