Industry-Driven Rapid Prototyping for Comparing Nonlinear Control Schemes in Pyomo

Abstract

With ongoing digitalization, fast simulation of process dynamics offers new opportunities for model-based control schemes. This is eased by the availability of high-level, open-access, easy-to-use software able to simulate and optimize chemical processes, making rapid prototyping possible also for the chemical engineer. In this study, the capabilities of the Python-based open-source software package Pyomo towards industrial application is illustrated in modeling and comparing different control schemes for a simple Williams-Otto process. It is shown how to simulate the process dynamics and how to compute optimal control trajectories for minimizing waste and maximizing yield. Two approaches to setpoint tracking are compared: one based on proportional-integral feedback control and one based on optimal open-loop control.