Adaptive Learning of Hybrid Models for Nonlinear Model Predictive Control of Distillation Columns. Issue 3 (2021)

• Record Type: Journal Article
• Authors: Lüthje, Jannik T.; Schulze, Jan C.; Caspari, Adrian; Mhamdi, Adel; Mitsos, Alexander; Schäfer, Pascal
• Is Part Of: IFAC-PapersOnLine. Volume 54:Issue 3(2021)
• Page Start: 37
• Page End: 42
• Publication Date: 2021
• Subjects: Nonlinear model reduction -- Model predictive -- optimization-based control -- Adaptive control -- Real time optimization -- control -- Neural networks in process control; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2021.08.215
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: Our previous work has shown that replacing parts of the classical compartmentalization model reduction approach for distillation columns by offline-trained artificial neural networks (ANNs) improves computational performance. In real-life applications, the absence of a high-fidelity model for data generation can, however, prevent the deployment of this approach. Therefore, we propose a method that utilizes solely plant measurement data, starting from a small initial data set and then continuously adapting to newly measured data. We demonstrate the approach in closed-loop simulations and compare to benchmarks using either the high-fidelity model or an offline trained reduced model for control.