Prediction of the surface tension of hydrocarbons with regressors based on machine learning techniques

• Ano de defesa: 2023
• Autor(a) principal: Fonseca Cala, Paula Jimena
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Tensão superficial; Hidrocarbonetos; Machine Learning; CNPQ::CIENCIAS EXATAS E DA TERRA::FISICA::FISICA DA MATERIA CONDENSADA
• Link de acesso: http://repositorio.ufc.br/handle/riufc/75232

Surface tension is a critical parameter in the petrochemical industry. It plays a key role in designing and optimizing various processes related to the production, transportation, and refining of hydrocarbons. The importance of surface tension lies in its ability to influence the behavior of liquids during these processes, affecting factors such as flow rate, separation efficiency, and the stability of emulsions. In recent years, machine learning (ML) techniques have shown promise in predicting the physical and chemical properties of hydrocarbons. These techniques offer a data-driven approach to understanding complex systems, and they have the potential to significantly improve the efficiency and accuracy of predictions related to hydrocarbon properties. In this study, we compared various machine learning algorithms, including K-Nearest Neighbors (KNN), Random Forest (RF), and XGBoost (XGB), to determine their effectiveness in predicting the surface tension of hydrocarbons. These algorithms were chosen due to their popularity and proven effectiveness in a variety of applications. The results of our study indicate that XGBoost exhibited the best performance in predicting the surface tension of hydrocarbons, with a mean squared error (MSE) of 4.65 and an R² score of 0.88. The R² score, also known as the coefficient of determination, is a statistical measure that represents the proportion of the variance for a dependent variable that’s explained by an independent variable or variables in a regression model. An R² score of 0.88 indicates a high level of accuracy in the predictions made by the XGBoost model. This study provides promising evidence that machine learning techniques can be effectively applied to predict the surface tension of hydrocarbons. The successful application of these techniques could lead to significant improvements in the efficiency and accuracy of processes in the petrochemical industry.