Data-Driven Model Reduction and Nonlinear Model Predictive Control of an Air Separation Unit by Applied Koopman Theory

TL;DR

This paper demonstrates a data-driven Koopman-based model reduction combined with a tailored nonlinear model predictive control approach, enabling real-time control of an air separation unit with significant computational efficiency gains.

Contribution

It introduces a novel Koopman theory-based reduction strategy integrated with a customized NMPC implementation for efficient real-time control of complex industrial processes.

Findings

Achieved 98% reduction in CPU time for NMPC of an ASU.

Successfully generated low-order models using machine learning and Koopman theory.

Enabled real-time control of an air separation unit with reduced computational resources.

Abstract

Achieving real-time capability is an essential prerequisite for the industrial implementation of nonlinear model predictive control (NMPC). Data-driven model reduction offers a way to obtain low-order control models from complex digital twins. In particular, data-driven approaches require little expert knowledge of the particular process and its model, and provide reduced models of a well-defined generic structure. Herein, we apply our recently proposed data-driven reduction strategy based on Koopman theory [Schulze et al. (2022), Comput. Chem. Eng.] to generate a low-order control model of an air separation unit (ASU). The reduced Koopman model combines autoencoders and linear latent dynamics and is constructed using machine learning. Further, we present an NMPC implementation that uses derivative computation tailored to the fixed block structure of reduced Koopman models. Our reduction approach with tailored NMPC implementation enables real-time NMPC of an ASU at an average CPU time decrease by 98 %.