Bi-level decision matrix based fault location method for multi-branch offshore wind farm transmission lines

Abstract

Multi-branch complex structure and harsh maintenance environment directly lead to the high-cost and time-consuming of fault detection in deep-sea offshore wind farms (DOWF). This paper proposes a traveling wave fault location method based on a bi-level decision matrix (BDM) aiming at multi-branch deep-sea offshore wind farm transmission lines. First, the first intrinsic mode function (IMF) signal is extracted by the variational mode decomposition (VMD) from initial traveling wave signals, which is further decomposed into S matrix by Stockwell-Transform (ST). Then based on the energy similarity between E matrix obtained through S matrix at all records, the area decision matrix is built to identify fault area. Second, S matrix recorded at both ends of fault area is extracted and analyzed by kurtosis to determine the traveling wave arrival time. Based on the fault area inherent topology and the traveling wave arrival time, the section decision matrix is built to identify fault section. Finally, the fault location is determined by the corresponding calculation process based on the identification results of BDM. Referring to the actual multi-branch topology of offshore wind farm (OWF), case studies are conducted under various fault conditions. The results show that the proposed method does not need to install recorders at all terminals, has high fault location accuracy, and is immune to noise.