Artificial Neural Networks for Predicting Pressure in High-Viscosity Two-Phase Flow: A Comparative Analysis

https://doi.org/10.58571/CNCA.AMCA.2023.098

Resumen: This article presents the procedure undertaken to discover the most suitable model for predicting the outlet pressure of a horizontally oriented pipe with high-viscosity two-phase flow. In order to achieve this, a collection of artificial neural networks (ANN) with diverse architectures was developed, utilizing experimental data for training and validation purposes. The data used in this study were obtained from experiments carried out in a 54-meter-long pipeline, where a mixture of air and glycerin was flowing. The various ANN were constructed based on three different combinations of input variables, namely, the air flow rate (Qa), glycerin flow rate (Qg), and pressure measurements at specific spatial points. Subsequently, a statistical analysis was performed, followed by a comparison of the models’ performance across different scenarios.

¿Cómo citar?

Ajbar, Wassila; Torres, Lizeth; Guzmán, Enrique; Palacio-Pérez, Arturo. Artificial Neural Networks for Predicting Pressure in High-Viscosity Two-Phase Flow: A Comparative Analysis. Memorias del Congreso Nacional de Control Automático, pp. 586-590, 2023. https://doi.org/10.58571/CNCA.AMCA.2023.098

Palabras clave

Redes Neuronales; Control Supervisor; Otros Tópicos Afines

• Ahmed, W.H., Bello, M.M., El Nakla, M., Al Sarkhi, A., and Badr, H.M. (2014). Experimental investigation of flow accelerated corrosion under two-phase Flow conditions. Nuclear Engineering and Design, 267, 34-43.
• Ajbar, W., Parrales, A., Silva-Martínez, S., Bassam, A., Jaramillo, O., and Hernández, J. (2021). Identification of the relevant input variables for predicting the parabolic trough solar collector’s outlet temperatura using an artificial neural network and a multiple linear regression model. Journal of Renewable and Sustainable Energy, 13(4), 043701.
• Thaker, J. and Banerjee, J. (2016). On intermittent Flow characteristics of gas–liquid two-phase flow. Nuclear engineering and design, 310, 363–377.
• Torres, L., Noguera, J., Guzmán, J.E.V., Hernández, J., Sanjuan, M., and Palacio-Pérez, A. (2020). Pressure signal analysis of the characterization of high-viscosity two-phase flows in horizontal pipes. Journal of Marine Science and Engineering, 8, 1–14.
• Wood, M., Vetere-Arellano, A., and Van-Wijk, L. (2013). Corrosion related accidents in petroleum refineries. JRC Sci. Poligy Report, 0, 1–100.