Distributed model predictive control using optimality condition decomposition and community detection. (March 2021)

• Record Type: Journal Article
• Authors: Segovia, P.; Puig, V.; Duviella, E.; Etienne, L.
• Is Part Of: Journal of process control. Volume 99(2021)
• Page Start: 54
• Page End: 68
• Publication Date: 2021-03
• Subjects: Large-scale systems -- Model predictive control -- Distributed control -- Optimality condition decomposition -- Community detection; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2021.01.007
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract: This work regards the development of a distributed model predictive control strategy for large-scale systems, as centralized implementations often suffer from non-scalability. The decomposition of the overall system into minimally coupled subsystems as well as their coordination are based on optimality condition decomposition (OCD) and community detection. The OCD approach allows to solve the associated control subproblems in parallel in an iterative manner until the required degree of accuracy is attained. The proposed strategy is tested on two different systems, the quadruple-tank system and the Barcelona drinking water network, which allow to highlight the effectiveness of the approach. Highlights: A distributed model predictive control approach for large-scale systems is derived. The proposed decomposition strategy employs the KKT optimality conditions. The use of community detection yields the optimal problem decomposition. The convergence and stability properties of the solution are analyzed. The approach is tested considering two different benchmarks.