RBFネットワークとGAによるHammerstein型非線形システムの同定  [in Japanese] IDENTIFICATION OF HAMMERSTEIN TYPE NONLINEAR SYSTEMS USING RBF NETWORKS AND GA  [in Japanese]

This paper proposes an identification method of Hammerstein type nonlinear systems by using radial basis function (RBF) networks and genetic algorithm (GA). An unknown nonlinear static part to be estimated is approximately represented by an RBF network. The weighting parameters of the RBF network and the system parameters of the linear dynamic part are estimated by the linear leastsquares method. The adjusting parameters for the RBF network structure, i. e. the number, centers and widths of the RBF are properly determined by using the GA, in which the Akaike information criterion (AIC) is utilized as the fitness value function. Simulation results are shown to examine the effectiveness of the proposed method.

This paper proposes an identification method of Hammerstein type nonlinear systems by using radial basis function (RBF) networks and genetic algorithm (GA). An unknown nonlinear static part to be estimated is approximately represented by an RBF network. The weighting parameters of the RBF network and the system parameters of the linear dynamic part are estimated by the linear leastsquares method. The adjusting parameters for the RBF network structure, i. e. the number, centers and widths of the RBF are properly determined by using the GA, in which the Akaike information criterion (AIC) is utilized as the fitness value function. Simulation results are shown to examine the effectiveness of the proposed method.