A New Tool Providing an Integrated Framework for Process Optimization

1. Abstract Nowadays, process optimization plays an important role in the chemical industries, providing many benefits. The aim of this work is to present a new framework for process optimization of chemical industries based on rigorous process models and with many integrated features, such as: customized thermodynamic data package, dynamic and steady-state simulation, optimization, parameter estimation, data reconciliation, gross error detection and integration with data bank. The integrated framework with all these features increases significantly the benefits of process optimization. This tool, named EMSO (Environment for Modeling, Simulation, and Optimization), is an equation-oriented dynamic and steadystate simulator and optimizer. It has an equipment library with rigorous models that can be viewed and edited by the user, using a high-level modeling language. The process flow diagram can be built either in text or graphical mode and the equipments can be modeled using an object-oriented approach. This software provides a set of sparse algebra and automatic differentiation feature to solve NLA (Non-Linear Algebraic Equations) systems in a fast and efficient way. NLP solvers, such as IPOPT and OPT++, were interfaced with this framework taking advantage of EMSO’s efficiency to solve NLA systems, reducing the computational effort required to achieve the solution of optimization problems. Other optimization solvers can be readily interfaced through a plug-in system. Besides steady-state optimization of process flow diagrams with rigorous models it is also possible to perform steady-state and dynamic parameter estimation using the optimization feature. Furthermore, given a set of process measurements, it is also possible to solve data reconciliation problems and perform gross error detection. A language to describe the formulation of the optimization problems was developed and is presented in detail. The implementations mentioned above were evaluated with literature and practical examples, showing the benefits of using the proposed framework. 2.