Uncertainty Analysis Applied to the Calculation of the Bubble Point Temperature of Mixtures

Abstract

The article presents the calculation of the bubble point of a mixture of nbutane, iso-butane, n-pentane and isopentane from the bottom of a distillation tower. This work aims to calculate the bubble point of a mixture using the Monte Carlo (MC) method. The calculation was performed under uncertainty in the Antoine parameters, which were obtained in the literature and optimized using Excel's Solver genetic algorithm and Stochastic simulation through MC method. In this way, the result is not a point temperature, but a range of probable values for the temperature of the bubble. This method proved to be promising and means more accurate process calculations, generating energy savings in industrial units.

Metamodeling of the deposition process in oil pre-processing to optimise the cleaning of the heat exchanger network: A systematic review

“…A common mistake made in process calculations is to treat problems on a one-off basis rather than as a range, without considering the uncertainties associated with the measurement parameters (Kalid et al, 2012). It is necessary to know the importance of data reliability and to understand how small uncertainties in this information can significantly affect the technical and economic performance of an industrial plant(Santana et al, 2021). The literature surveyed shows a proposal for an MDE (Model-Driven Engineering) infrastructure for developing operation, control and simulation applications in the petrochemical industry, more specifically in the field of defining industrial plant equipment classes (fig.2), where the infrastructure, called M4PIA (Model-Driven Engineering for Petrochemical Industry Automation), is made up of three metamodels at two levels of abstraction (independent and dependent on the target platform on which the application will be implemented) and the mappings needed to automatically transform the system modelled at the highest level of abstraction into a platform-specific model and then automatically generate the application source code to be used in the implementation on the defined platform.…”