Multi-step-ahead prediction using dynamic recurrent neural networks

• Published in: Neural networks Vol. 13; no. 7; pp. 765 - 786
• Main Authors: Parlos, A.G., Rais, O.T., Atiya, A.F.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.09.2000; Elsevier Science
• Subjects: Algorithms; Applied sciences; Complex dynamics; Dynamic learning; Dynamic recurrent networks; Electric, optical and optoelectronic circuits; Electronics; Exact sciences and technology; IIR networks; Multi-step-ahead prediction; Neural networks; Neural Networks (Computer); Nonlinear Dynamics; Nonlinear predictors; Complex dynamics; Multi-step-ahead prediction; Nonlinear predictors; Dynamic recurrent networks; IIR networks; Dynamic learning; Infinite impulse response filter; Multistep method; Non linear filter; Descent method; Network architecture; Open loop; Autoregressive model; Neural network; Learning algorithm; Empirical model; Gradient method; Dynamic method
• ISSN: 0893-6080; 1879-2782
• DOI: 10.1016/S0893-6080(00)00048-4

A method for the development of empirical predictive models for complex processes is presented. The models are capable of performing accurate multi-step-ahead (MS) predictions, while maintaining acceptable single-step-ahead (SS) prediction accuracy. Such predictors find applications in model predictive controllers and in fault diagnosis systems. The proposed method makes use of dynamic recurrent neural networks in the form of a nonlinear infinite impulse response (IIR) filter. A learning algorithm is presented, which is based on a dynamic gradient descent approach. The effectiveness of the method for accurate MS prediction is tested on an artificial problem and on a complex, open-loop unstable process. Comparative results are presented with polynomial Nonlinear AutoRegressive with eXogeneous (NARX) predictors, and with recurrent networks trained using teacher forcing. Validation studies indicate that excellent generalization is obtained for the range of operational dynamics studied. The research demonstrates that the proposed network architecture and the associated learning algorithm are quite effective in modeling the dynamics of complex processes and performing accurate MS predictions.