A neural network approach to predicting the quality indicators of the purification department of a sugar factory
DOI:
https://doi.org/10.31548/energiya3(73).2024.037Abstract
The article proposes a model of the defecosaturation department of a sugar beet processing plant based on a neural network, which predicts the main indicators of the efficiency of diffusion juice purification. The predictive model allows you to adjust the operation mode of the department by analyzing the unmeasured qualitative indicator of the purification effect. The developed model also predicts the color of the semi-finished product, as well as the loss of sugar in the filter cake I and II saturation with an error of less than 5%. Unlike existing solutions, forecasting is performed in real time using a set of indicators of an automation system and an industrial laboratory. A predictive neural network model is an MLP universal for approximating non-linear complex functions with many variables. The model has 32 inputs, 4 outputs and one hidden layer with 23 neurons. The main technological variables from the automation system – temperatures, flow rates, pressure, pH in the apparatus, as well as from the industrial laboratory – alkalinity, SO content in the saturation gas will be fed into the network input. This allows to increase the informative support of the operator-technologist regarding the quality indicators of the enterprise without additional load on the industrial laboratory and without additional cost automatic quality devices.
Key words: defecosaturation department, neural network, forecast, sugar beet, quality assessment
References
Council Directive 2001/111/EC of December 20, 2001 on certain sugars intended for human consumption. Avalaible at: https://zakon.rada.gov.ua/laws/show/984_002-01#Text , last accessed 2022/07/13.
Zaiets, N., Vlasenko, L., Lutska, N., Shtepa, V. (2022). Resource Efficiency Forecasting Neural Network Model for the Sugar Plant Diffusion Station. In: Szewczyk, R., Zieliński, C., Kaliczyńska, M. (eds.) Automation 2022: New Solutions and Technologies for Automation, Robotics and Measurement Techniques. Advances in Intelligent Systems and Computing, vol. 1427. Springer, Cham (2022).
Ogando, F. I. B., de Aguiar, C. L., Viotto, J. V. N., Heredia, F. J., Hernanz, D. (2019). Removal of phenolic, turbidity and color in sugarcane juice by electrocoagulation as a sulfur-free process. Food Research International 122 (1337), 643–652.
Saska, M., Zossi, S., Liu, H. (2010). Colour behaviour in cane juice clarification by defecation, sulfitation and carbonation. In: International society of sugar cane technologists, 27, 1–14.
Begum, R., Nimbal, G. M., Halse, S. V. (2013). Fuzzy neural network modelling and pH value control in the clarifying process of sugar cane juice. International Journal of Electrical and Electronic Engineering & Telecommunications 2(2), 25–32.
Lin, X., Lei, S., Song, C., Song, S., Liu, D. (2008). ADHDP for the pH value control in the clarifying process of sugar cane juice. In: International Symposium on Neural Networks, 796–805. Springer, Berlin, Heidelberg.
Xiaofeng, Lin, Jiaran, Yang. (2009). HDP for the Neutralized pH Value Control in the Clarifying Process of Sugar Cane Juice. In: AIP Conference Proceedings, 1127, 162173.
Sunori, S. K., Negi, P. B., Arora, S., Khan, F., Maurya, S., Juneja, P., Ghai, K. (2022). Neuro-fuzzy Controller Design for pH Control in Sugar Refineries. In: 8th International Conference on Advanced Computing and Communication Systems (ICACCS), l (1), 197–202. IEEE.
Zaiets, N. A., Savchuk, O. V., Shtepa, V. M., Lutska, N. M., Vlasenko L.O. (2021). The synthesis of strategies for the efficient performance of sophisticated technological complexes based on the cognitive simulation modelling. Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu 2, 110–117.
Zaiets, N., Vlasenko, L., Lutska N., Usenko, S. (2019). System Modeling for Construction of the Diagnostic Subsystem of the Integrated Automated Control System for the Technological Complex of Food Industries. In: 5th International Conference on Mechatronics and Robotics Engineering «ICMRE 2019», 93–98. Rome, Italy.
Vukov, K. (1977), Physics and Chemistry of Sugar-Beet in Sugar Manufacture. Auflage, 595 Seiten, 177 Abb., 228 Tab. Verlag Akademiai Kiadó, Budapest.
Shaojian, S., Jinchuan, W., Xiaofeng, L., Huixia, L. (2012) Modeling of key production indices and operating parameters optimized set for sugar clarification process. In: Proceedings of the 31st Chinese Control Conference, 7113–7118. IEEE.
Song, C. N., Zhong, W. H. (2013). Study on control of the second carbonation in the clarifying process of sugar cane juice via generalized predictive control algorithm. In: Applied Mechanics and Materials, 303, 1257–1260. Trans Tech Publications Ltd.
Zarifpoor, M., Hakimzadeh, V. (2018). Investigating of artificial neural network potential to predict the properties of refined raw sugar beet juice by electrocoagulation process. Ukrainian Food Journal 7(4), 682–691.
Song, S., Wu, J., Lin, X., Liu, H. (2012). Predictive model of production index for sugar clarification process by GDFNN. In: International Symposium on Neural Networks, 585–593. Springer, Berlin, Heidelberg.
Downloads
Published
Issue
Section
License
Relationship between right holders and users shall be governed by the terms of the license Creative Commons Attribution – non-commercial – Distribution On Same Conditions 4.0 international (CC BY-NC-SA 4.0):https://creativecommons.org/licenses/by-nc-sa/4.0/deed.uk
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
