Optimal identification method of nonlinear system based on GA-GHNNs P
Lin Xiao-fei, Weng Mu-yun · 2009
Gaussian-Hopfield neural network algorithm (GHNNs) is the most commonly used method of solving the identification problem of nonlinear systems, but learning rule (LMS rule) is easy to fall into local optimum. Genetic algorithm (GA) has globally optimal ability and can solve the locally optimal problem well. This paper puts forward GA-GHNNs algorithm and uses GA algorithm to solve the optimum parameters of GHNNs network. And finally, it carries out simulation experiments to prove the validity of the algorithm. Simulation results also show that this method has the ability to distinguish nonlinear systems.