Control improvement of a C sugar cane crystallization using an auto tuning PID controller based on linearization of a neural network

Abstract : The industrial process of the sugar cane crystallization produces a residual that still contains a lot of soluble sucrose and the objective of the factory is to improve its extraction. Therefore, there are substantial losses justifying the search for the optimization of the process. Crystallization process studied on the industrial site is based on the “three massecuites process". The third step of this process constitutes the final stage of exhaustion of the sucrose dissolved in the mother liquor. During the process of the third step of crystallization (Ccrystallization), the phase that is studied and whose control is to be improved, is the growing phase (crystal growth phase). The study of this process on the industrial site is a problem in its own. A control scheme is proposed to improve the standard PID control law used in the factory. An auto-tuning PID controller based on instantaneous linearization of a neural network is then proposed.
Liste complète des métadonnées

http://hal.univ-reunion.fr/hal-01202301
Contributor : Réunion Univ <>
Submitted on : Saturday, September 19, 2015 - 4:46:35 PM
Last modification on : Thursday, March 28, 2019 - 11:24:13 AM

Identifiers

  • HAL Id : hal-01202301, version 1

Collections

Citation

Sébastien Beyou, Brigitte Grondin-Perez, Michel Benne, Cédric Damour, Jean-Pierre Chabriat. Control improvement of a C sugar cane crystallization using an auto tuning PID controller based on linearization of a neural network. World Academy of Science, Engineering and Technology 2009, Sep 2009, Le Caire, Egypt. International Journal of Computer, Electrical, Automation, Control and Information Engineering, 3 (6), pp.1646-1651, 2009. 〈hal-01202301〉

Share

Metrics

Record views

66