Decomposition of control and optimization problems by network structure: Concepts, methods, and inspirations from biology

Abstract

This is the author manuscript accepted for publication and has undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Modularity measures: Concepts, computation, and applications to manufacturing systems

“…Current techniques available include hierarchical clustering algorithms, k‐means clustering, spectral clustering, and techniques based on modularity maximization . An excellent review on community detection techniques is provided in Reference . In this work, we adopt a modularity maximization approach, as this provides an intuitive approach to analyze and design modular manufacturing systems.…”

“…We highlight that the proposed modularity measure and MIQP formulation can be used in other applications that go beyond manufacturing. For instance, these tools can be used to identify optimal configurations for control architectures and optimal decomposition strategies for optimization problems . In this context, constraints on the number and size of modules can be used to create balanced configurations (e.g., to handle computational load balancing issues).…”

Distributed monitoring of the absorption column of a post‐combustion CO2 capture plant

“…It is seen that the objective of community structure detection is well aligned with that of subsystem decomposition for distributed state estimation/monitoring/control, which expects minimal interaction among different subsystems. Motivated by this observation, community detection has been taken advantage of to develop subsystem decomposition approaches for distributed control and simultaneous distributed estimation and control . In this work, we resort to our method proposed in the work of Yin and Liu to perform community‐based subsystem decomposition.…”

Machine learning‐based distributed model predictive control of nonlinear processes

“…In other words, the controllers communicate with each other to calculate their distinct set of manipulated inputs that will collectively achieve the control objectives of the closed‐loop system. Many distributed MPC methods have been proposed in the literature addressing the coordination of multiple MPCs that communicate to calculate the optimal input trajectories in a distributed manner (the reader may refer to References 6‐8 for reviews of results on distributed MPC, and to Reference 9 for a review of network structure‐based decomposition of control and optimization problems). A robust distributed control approach to plant‐wide operations based on dissipativity was proposed in References 10 and 11.…”