Application of neural network for time-varying large delay system

The compensative control method is commonly applied in the delay system, but it needs the accurate mathematics model of objective. It is impossible to get the accurate model by traditional ways when the parameters are time-varying. Neural network is used for the identification of the large time-varying delay system in this paper. The abilities of the nonlinear map, self-learning, and parallel processing of the neural network ensure the accuracy of the system model identification. This method is used for parameter identification in the boiler backwater control system, and the results are satisfying.