|
Abstract
|
This paper proposes a novel approach for coherent
generators online clustering in a large power system following a
wide area disturbance. An interconnected power system may
become unstable due to severe contingency when it is operated
close to the stability boundaries. Hence, the bulk power system
controlled islanding is the last resort to prevent catastrophic
cascading outages and wide area blackout. Meanwhile, the
aggregation of the coherent generators is the most important step
in large power grids intentional defensive splitting to guarantee
the dynamic stability of the created islands and reduce the
computational burden of the huge initial search space. The
proposed method of this paper determines the coherent machines
based on the electrical parameters of the synchronous generators
instead of the dynamical parameters such as rotor angle or speed
curves, speed participation factors, etc. The stator and excitation
windings flux, excitation voltage and current have been proposed
as coherency indices. The proposed coherency based aggregation
has been carried out on New England 39-bus test system. The
time-domain simulation results demonstrate the effectiveness and
capability of the proposed method to identify the coherent
machines following a severe contingency.
|