Action integrals estimate the conductor temperature rise in
the coils, commonly referred to as ∫ I 2(t)dt. There are two
action integrals: the total action for the current time step k
based on the action from the previous time step k−1 (2), and
an estimate of the additional action that would accumulate if a
fault were to occur and the current were to decay exponentially
from the present state (3). In this context, t is the time
between each time step, and τ is the L/R time constant of
the circuit under consideration.
Since it does not make sense to compute action for postdisruption
currents, this algorithm only uses the currents for the
currently executing time step as its input current vector.