Premium
Dynamic process modeling with recurrent neural networks
Author(s) -
You Yong,
Nikolaou Michael
Publication year - 1993
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.690391009
Subject(s) - recurrent neural network , feed forward , computer science , feedforward neural network , artificial neural network , control theory (sociology) , nonlinear system , computation , process (computing) , echo state network , process dynamics , system dynamics , time delay neural network , control engineering , artificial intelligence , algorithm , engineering , control (management) , physics , quantum mechanics , operating system
A method of nonhlinear static and dynamic process modeling via recurrent neural networks (RNNs) is studied. An RNN model is a set of coupled nonlinear ordinary differential equations in continuous time domain with nonlinear dynamic node characteristics as well as both feedforward and feedback connections. For such networks, each physical input to a system corresponds to exactly one input to the network. The system's dynamics are captured by the internal structure of the network. The structure of RNN models may be more natural and attractive than that of feedforward neural network models, but computation time for training is longer. Our simulation results show that RNNs can learn both steady‐state relationships and process dynamics of continuous and batch, single‐input/single‐output and multiinput/multioutput systems in a simple and direct manner. Training of RNNs shows only small degradation in the presence of noise in the training data. Thus, RNNs constitute a feasible alternative to layered feedforward back propagation neural networks in steady‐state and dynamic process modeling and model‐based control.