DC Side Fault Analysis in an Off Shore Wind Farm based High Voltage DC Transmission Link

摘要

This paper proposes a simple event detection and classification technique for disturbances on the DC side of a voltage source converter based high voltage DC transmission link which carries power from an off shore based wind farm to main grid. The voltage and current samples near the DC terminals of the high voltage DC transmission link are initially extracted and are directly processed to generate the target feature sets which consist of energy, mean and variance respectively. Subsequently the target feature sets are passed through a simple decision tree based classifier for event classification. The proposed technique has been tested for a number of disturbances on the DC side of the HVDC link where it has been verified as a superior technique as compared with some of the existing event classification technique based on wavelet transform. The entire model has been simulated in MATLAB/ Simulink and the testing has been verified in MATLAB/ Editor. The proposed technique avoids the need of any signal processing technique to process the raw signal (dc current), thus inhabiting less computations and faster execution and hence is an efficient algorithm for real time applications.