Constructive learning neural network applied to identification and control of a fuel-ethanol fermentation process

被引:37
作者
da Cruz Meleiro, Luiz Augusto [1 ]
Von Zuben, Fernando Jose [2 ]
Maciel Filho, Rubens [3 ]
机构
[1] Univ Fed Rural Rio de Janeiro, Dept Food Engn, BR-23890971 Seropedica, RJ, Brazil
[2] FEEC UNICAMP, State Univ Campinas, Sch Elect & Comp Engn, BR-13083970 Campinas, SP, Brazil
[3] FEQ UNICAMP, State Univ Campinas, Sch Chem Engn, BR-13081970 Campinas, SP, Brazil
关键词
Model predictive control; Constructive neural networks; Fermentation process; Bioprocess identification; Dynamic simulation; MODEL-PREDICTIVE CONTROL; ALGORITHMS; SIMULATION;
D O I
10.1016/j.engappai.2008.06.001
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In the present work, a constructive learning algorithm was employed to design a near-optimal one-hidden layer neural network structure that best approximates the dynamic behavior of a bioprocess. The method determines not only a proper number of hidden neurons but also the particular shape of the activation function for each node. Here, the projection pursuit technique was applied in association with the optimization of the solvability condition, giving rise to a more efficient and accurate computational learning algorithm. As each activation function of a hidden neuron is defined according to the peculiarities of each approximation problem, better rates of convergence are achieved, guiding to parsimonious neural network architectures. The proposed constructive learning algorithm was successfully applied to identify a MIMO bioprocess, providing a multivariable model that was able to describe the complex process dynamics, even in long-range horizon predictions. The resulting identification model was considered as part of a model-based predictive control strategy, producing high-quality performance in closed-loop experiments. (C) 2008 Elsevier Ltd. All rights reserved.
引用
收藏
页码:201 / 215
页数:15
相关论文
共 41 条
[1]  
ANDRIETTA SR, 1994, ADVANCES IN BIOPROCESS ENGINEERING, P47
[2]  
ANDRIETTA SR, 1994, THESIS STATE U CAMPI
[3]  
Astrom K. J., 1997, COMPUTER CONTROLLED
[4]   Kinetics of ethanol fermentation with high biomass concentration considering the effect of temperature [J].
Atala, DIP ;
Costa, AC ;
Maciel, R ;
Maugeri, F .
APPLIED BIOCHEMISTRY AND BIOTECHNOLOGY, 2001, 91-3 (1-9) :353-365
[5]   ON A CLASS OF EFFICIENT LEARNING ALGORITHMS FOR NEURAL NETWORKS [J].
BARMANN, F ;
BIEGLERKONIG, F .
NEURAL NETWORKS, 1992, 5 (01) :139-144
[6]   1ST-ORDER AND 2ND-ORDER METHODS FOR LEARNING - BETWEEN STEEPEST DESCENT AND NEWTON METHOD [J].
BATTITI, R .
NEURAL COMPUTATION, 1992, 4 (02) :141-166
[7]  
CAMACHO EF, 1999, MODEL BASED PREDICTI
[8]  
Clarke D., 1994, Advances in Model-Based Predictive Control
[9]   Non-linear predictive control of an extractive alcoholic fermentation process [J].
Costa, AC ;
Meleiro, LAC ;
Maciel, R .
PROCESS BIOCHEMISTRY, 2002, 38 (05) :743-750
[10]   Factorial design and simulation for the optimization and determination of control structures for an extractive alcoholic fermentation [J].
Costa, AC ;
Atala, DIP ;
Maugeri, F ;
Maciel, R .
PROCESS BIOCHEMISTRY, 2001, 37 (02) :125-137