Fault detection for offshore wind farm connected to onshore grid via voltage source converter‐high voltage direct current

The fault study in offshore wind farm connected to onshore grid is carried out using empirical mode decomposition (EMD) which is a powerful technique for analysing linear, non‐linear, stationary and non‐stationary signals. The efficacy of the proposed approach is carried out by virtue of comparative assessment with other well‐established signal processing techniques: wavelet transform, Stockwell transform, hyperbolic Stockwell transform in the literature. The intrinsic mode functions (IMFs) are obtained using EMD for signals (normal and fault) retrieved at different sections. Hilbert transform is applied to each IMF in order to evaluate the magnitude and phase angle information used for analysing the signal. Both qualitative and quantitative analyses are carried out to demonstrate the effective detection of fault occurrence. The adopted approach accurately discriminates the occurrence of fault in AC/DC network section. The validation of the detection algorithm has been demonstrated through real‐time digital simulator studies. In addition, faulty sections are characterised adopting classification strategy using support vector machines.