Hybrid genetic neural networks for chaotic system control

By incorporating the temporal difference prediction technique with the genetic algorithm, a novel hybrid genetic neural network (known as HyGANN) for controlling nonlinear chaotic system based on the scheme of small perturbations was presented . The HyGANN trained by reinforcement learning algorithm could generate small perturbation time series signals to suppress the chaotic states. The computer simulations on Henon map chaotic system have shown that the behavior of period 1 and period 2 can be controlled effectively and the high period orbit can be directed towards desired periodic trajectory.