The New Approach for Dynamic Optimization with Variability Constraints

Abstract

In this work a new optimization approach for processes modeled by differential-algebraic equations with variability constraints was presented. The designed procedure was based on the modified direct shooting method, which can transform the dynamic optimization problem into a large-scale nonlinear optimization task (NLP). The first-order KKT optimality conditions with complementarity constraints were obtained. Finally, to solve the optimality conditions with the complementarity constraints, the solution procedure combining SQP algorithm with the filter approach as a globalization procedure was designed. The efficiency of the presented methodology was tested on a production process in chemical engineering.