Implementation of multilinear interpolators as surrogate models of ethanolic transesterification reactors in global simulator EMSO

• Grant number: 18/07546-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2018
• End date: December 31, 2019
• Field of knowledge: Engineering - Chemical Engineering - Chemical Process Industries
• Principal Investigator: Roberto de Campos Giordano
• Grantee: Christian de Oliveira Martins
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Associated research grant: 14/21252-0 - Equilibrium and production processes of biofuels and bioproducts, AP.TEM

Abstract

The increase in computational power in the last decades has made process simulators essential tools for evaluating the technical and economic viability of new processes. These software solve systems of equations (algebraic or differential-algebraic), of the order of tens of thousands of variables, in different ways. The sequential modular simulators partition the process and solve these sequentially (usually observing the topological features of the process). For this reason, such simulators (such as Aspen Plus®) use specific calculation routines for each equipment, which makes its resolution more robust. On the other hand, equation-oriented or global simulators solve the whole system of equations simultaneously. These do not allow the use of specific calculation routines, reducing their robustness, but they have advantages in the construction, reuse and maintenance of the models. The EMSO simulator (Environment for Modeling Simulation and Optimization) is the global simulator that will be used in the present work. To overcome the convergence problems that arise in this type of simulator, surrogate models can be used. These can be, among other options, based on multilinear inspection tables to represent the behavior of the rigorous models or the experimental data directly. Such models have the advantage of having its accuracy controlled by the number of points in the look-up table, so that the accuracy is adjusted based on the model application needs. The thematic project in which this scientific initiation project is inserted is focused on the biodiesel production process by transesterification. It is estimated that this process produces 25% less greenhouse gases than the methyl process. In this context, this project proposes the simulation of the ethyl biodiesel production process using rigorous models and the use of the data obtained to construct the surrogate model, based on interpolators, for integration in the complete biorefinery simulation. (AU)