Probabilistic assessment of new time-domain distance relay algorithms

• Time-domain distance relay algorithms with polygonal characteristic are developed.
• Time-domain distance relay algorithms are compared to a phasor-domain algorithm.
• Methodology for assessing security and dependability of protective relay operations.

In this paper we propose novel distance relay algorithms and a new test methodology for assessing the security and dependability of protective relay operations. The proposed algorithms with a polygonal operating characteristic are based on a time-domain phase comparator and compared to an algorithm based on discrete Fourier transformation (DFT), which is considered as a classical solution in this area.Algorithms are usually tested on the worst case scenarios which purpose is to check if the algorithm would maloperate under some conditions and this kind of tests is necessary considering the enormous economic damage in case of power system blackouts. However, this approach is not enough for comparison of different algorithms if each of them passed those tests. Every comparison based on our choice of the test cases is in some way subjective and is not enough for assessment of security, dependability and average tripping time. In order to avoid a subjective selection of the simulation parameters we propose a probabilistic approach where the real frequency of some network conditions is considered using the probability density functions of the main variables measured in a real power system. The security, dependability and the relay tripping time are evaluated through a few thousand fault simulations and comparisons of the algorithms are presented in the paper. The proposed testing model can be adjusted to any transmission network if appropriate data is available.