Abstract In this paper, an integrated neurofuzzy-JITL model is proposed for batch processes. The neurofuzzy is employed to build a global model with its excellent extrapolating ability while Just-in-Time Learning (JITL) is used to build the local ARX model due to its good local dynamic modeling ability. In addition, Simulated Annealing (SA) algorithm is adopted to obtain the optimal weights of two models. As a result, the integrated model has better global generalization ability and higher accuracy. Lastly, the effectiveness of the presented integrated neurofuzzy-JITL model is verified by example.
Integrated Neurofuzzy-JITL Model and Its Application in Batch Processes
2014-01-01
11 pages
Article/Chapter (Book)
Electronic Resource
English
Integrated Neurofuzzy-JITL Model and Its Application in Batch Processes
British Library Conference Proceedings | 2014
|An LMI Approach to Iterative Learning Control Based on JITL for Batch Processes
Springer Verlag | 2017
|Anticipatory Neurofuzzy Control
NTRS | 1994
|Anticipatory Neurofuzzy Control
Online Contents | 1994
Hierarchical neurofuzzy model for real-time flood forecasting
British Library Online Contents | 2013
|