No abstract

No abstract

The development of multiterminal VSC-HVDC grids poses a new challenge: a fast and selective protection algorithm for such grids must be implemented. Therefore, this paper addresses a novel protection algorithm, which proves to be accurate, fast and selective for various combinations of fault distance, resistances and power flow scenarios in a multiterminal VSC-HVDC environment. It relies on the wavelet transform and principal component analysis to retrieve information from the local DC current measurement, generating a highly informative input for a genetic fuzzy system. Based on previously simulated fault scenarios, the genetic fuzzy system is able to decide if the input signal refers to an internal or external fault, generating the trip signal when needed. Furthermore, the method has the advantage of using local measurements to avoid communication delay. Finally, the algorithm was implemented in hardware and tested in a hardware-in-the-loop simulation, which emphasizes the practical use of the proposed protection scheme.

No abstract

No abstract