Artificial Intelligence based network strength estimation for LCC HVDC

The dynamic performance and stability of line-commutated converter (LCC) for High-Voltage, Direct Current (HVDC) transmission are dependent upon grid impedance seen by the converter. Its performance can benefit from additional information given by online measurement of grid impedance. With the accelerating transformation of the power systems towards a renewable generation-based one, it is important to include a measure of the amount of nearby voltage controlling devices to the concept of fault level. Another challenge faced by LCC converters is a sudden change in the grid impedance, like the ones experienced after a trip of an important AC network equipment. This paper addresses these challenges by using a more comprehensive concept of network strength which defines fault level accounting nearby voltage controlling devices. The paper also proposes a method to quickly estimate the resulting fault level after a sudden change in grid impedance. The task of network strength estimation is handled by using the Artificial Neural Networks (ANN). Results confirm that the method not only provides a reliable steady-state estimation but also a fast & accurate estimate of the resulting fault level after fault clearing.