Research on Multivariable Time Delay System Multi-step Predictive Dynamic Modeling

This paper introduces a dynamic modeling method of multi-step predictive neural network in the large time delay system, and discusses the algorithm and structure of neural network and dynamic response characteristics of the large time delay system in detail. Especially, based on single variable time delay system modeling framework, a new algorithm is presented to implement the neural network multi-step predictive modeling of multivariable large time delay system. According to the data of the waterworks coagulating sedimentation time delay system, the simulation of the multivariable multi-step predictive modeling shows that these results are effective.