Model identification of time-delay nonlinear system with FIR neural network

The FIR neural network model and its temporal backpropagation algorithm are introduced in this paper. Due to its time-delay dynamic characteristics, it is fit well to be applied to model identification of the time-delay nonlinear system. Model identification has been completed successfully on actual datasets of 6 stands tandem hot mill with FIR neural network, and the results show its good characteristics.