The individuals in each species population are initially
evaluated in collaboration with a single randomly selected
individual from the other species population. Thereafter, the
GA proceeds as before, however, alternating between species
after each offspring is formed and evaluated with the elite
member from the other species; see algorithm outline in
Algorithm 1 and [77] for discussions on collaboration strategies.
In the SCGA, the models use identical parameters to
the single VAWT experiments, however, 16 input neurons are
now required. In addition, the model weights must be reinitialized
each time before training due to the temporal nature of
pairing with the elite member, and the GA runs for one generation
(using the model approximated fitnesses where real fitness
is unknown) before the individual with the highest approximated
fitness and a randomly selected unevaluated individual
are evaluated with the elite member from the other species;
see outline in Algorithm 2.