Constrained Optimization of Nonlinear, Dynamic Chemical Processes – A Normalized Form Approach

Abstract

In chemical manufacturing processes, optimization problems with differential-algebraic constraints are frequently encountered. In general, these problems are difficult to solve and solution approaches are usually based on discretization schemes. This paper proposes an alternative method to determine optimal setpoint trajectories for a class of dynamic systems with differential-algebraic constraints. The method exploits di ff erential flatness to explicitly eliminate the differential state equations. The resulting optimization problem is an algebraic, nonlinearly constrained optimization problem, and can be solved by nonlinear programming. The proposed approach has the potential to reduce the amount of on-line computation required to solve a specific class of these problems in real-time.