A novel modelling method of Hammerstein systems with a time-delay block

The Hammerstein model is an effective way to model the nonlinear systems, which contains a nonlinear static block followed by a linear dynamic block. However, when facing with the strong nonlinear time-delay effects, the nonlinear memoryless block can not work very well. Hence, a time-delay neural network is used as a quasi-dynamic nonlinear block in the established nonlinear Hammerstein model which now is called the Hammerstein-QD model. The radial basis function (RBF) neural network is adopted as the time-delay nonlinear function of the Hammerstein-QD model. Then the recursive least square algorithm is investigated to implement the parameter estimation. The MATLAB simulation results validate the effectiveness of the proposed time-delay system modelling method and the estimation method.