Modeling for a complicated industrial object based on recurrent neural network
Wei Jianping, LI Hua-de, Sun Ming, Sun Shaoyuan · 2002
This paper discusses the architecture and algorithm of a class of dynamical neural network, the Elman recurrent neural network (RNN). Based on this network an approach for modeling a nonlinear time-varying industrial object, the direct current arc, is proposed. Compared with other modeling method for the object, the model based on RNN is proved to have better performance.