Neural Model for the Prediction of Volume Losses in the Aging Process of Rums
Abstract
The rum aging process shows volume losses, called wastage. The numerical operation variables: product, boardwalk, horizontal and vertical positions, date, volume, alcoholic degree, temperature, humidity and aging time, recorded in databases, contain valuable information to study the process. MATLAB 2017 software was used to estimate volume losses. In the modeling of the rum aging process, the multilayer perceptron neuronal network with one and two hidden layers was used, varying the number of neurons in these between 4 and 10. The Levenberg-Marquadt (LM) and Bayesian training algorithms were compared (Bay) The increase in 6 consecutive iterations of the validation error and 1,000 as the maximum number of training cycles were the criteria used to stop the training. The input variables to the network were: numerical month, volume, temperature, humidity, initial alcoholic degree and aging time, while the output variable was wastage. 546 pairs of input/output data were processed. The statistical Friedman and Wilcoxon tests were performed to select the best neural architecture according to the mean square error (MSE) criteria. The selected topology has a 6-4-4-1 structure, with an MSE of 2.1∙10-3 and a correlation factor (R) with experimental data of 0.9898. The neural network obtained was used to simulate thirteen initial aging conditions that were not used for training and validation, detecting a coefficient of determination (R2) of 0.9961.
Keywords
rums, aging, volume losses, modeling, artificial neural networks, MATLAB
Author Biography
Beatriz García-Castellanos
Ingeniera Química
Centro de Referencia de Alcoholes y Bebidas (CERALBE)
Osney Pérez-Ones, Ph. D.
Profesor Auxiliar
Decano
Facultad de Ingeniería Química
Lourdes Zumalacárregui-de-Cárdenas, Ph. D.
Profesor Titular
Facultad de Ingeniería Química
Idania Blanco-Carvajal, M.Sc.
Investigador Auxiliar
Jefe de Producción
Centro de Referencia de Alcoholes y Bebidas (CERALBE)
Luis Eduardo López-de-la-Maza
Profesor Instructor
Facultad de Ingeniería Química
References
[1] Ministerio de la industria alimentaria. Resolución No. 12/19. La Habana, Cuba, 13-2-2019.
[2] J. Marcano, “El ron. Tradición en Las Antillas,” 2019. Available: https://mipais.jmarcano.com/economia/ron.html.
[3] I. Witten, and E. Frank. Data Mining. Amsterdam: Morgan Kaufmann Publishers, 2005.
[4] M. Servente, “Algoritmos TDIDT aplicados a la minería de datos inteligente”. Tesis de grado, Universidad de Buenos Aires, Buenos Aires. Feb. 2002.
[5] L. Zumalacárregui, O. Pérez, F. Hernández, G. Cruz, G. “Modelación del equilibrio líquido-vapor a presión constante de mezclas etanol-agua utilizando redes neuronales artificiales,” Tecnología Química, vol. 38 (3), pp. 527-548, Jul. 2018.
[6] H. Abdi, and L. J. Williams “Principal component analysis,” Wiley Interdisciplinary Reviews: Computational Statistics, vol. 2 (4), pp. 433-59, Jul.-Aug.2010. https://doi.org/10.1002/wics.101.
[7] E. F. Caicedo, and J. A. López, Una aproximación práctica a las redes neuronales artificiales Santiago de Cali: Programa Editorial Universidad del Valle, Edición digital, Jul. 2017.
[8] J. D. Terán-Villanueva, S. Ibarra-Martínez, J. Laria-Menchaca, J. A. Castán-Rocha, M. G. Treviño-Berrones, A. H. García-Ruiz, and J. E. Martínez-Infante, “Estudio de redes neuronales para el pronóstico de la demanda de asignaturas,” Revista Facultad de Ingeniería, vol. 28 (50), pp. 30-39, Ene. 2019. https://doi.org/10.19053/01211129.v28.n50.2019.8783.
[9] M. A Korany, H. Mahgoub, O. Fahmy, and H. Maher, “Application of artificial neural networks for response surface modelling in HPLC method development,” Journal of Advanced Research, vol. 3 (1), pp. 53-63, Jan. 2012. https://doi.org/10.1016/j.jare.2011.04.001.
[10] L. E. López, L. Zumalacárregui, O. Pérez, and O. Llanes, “Obtención de un modelo neuronal para la estimación de la concentración de etanol en la destilería Héctor Molina,” Tecnología Química, vol. 38 (2), pp. 315-325, May. 2018.