A Framework for the Simulation and Validation of Distributed Control Architectures for Technical Systems of Systems*. Issue 1 (July 2017)

• Record Type: Journal Article
• Authors: Nazari, Shaghayegh; Wenzel, Simon; Samuel Maxeiner, Lukas; Sonntag, Christian; Engell, Sebastian
• Is Part Of: IFAC-PapersOnLine. Volume 50:Issue 1(2017)
• Page Start: 12458
• Page End: 12463
• Publication Date: 2017-07
• Subjects: Distributed management -- distributed control -- simulation -- model-based validation -- Modelica -- integrated petrochemical production site -- semi-batch reactor network; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2017.08.1921
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: There are many examples of modern technical systems that consist of many closely integrated physical and cyber subsystems. Centralized management and control strategies may not always be the best option, or may even be infeasible, for these so-called cyber-physical systems of systems (CPSoS) due to their large complexity or due to confidentiality restrictions between the subsystems. These challenges can be overcome by distributed management and coordination strategies that do not require a central controller. Instead, each subsystem solves its own local optimization problem and only reports limited information to a higher-level controller to ensure the satisfaction of the global constraints which are often defined as limitations on the utilization or production on some shared resources or raw materials. This paper presents a novel Modelica-based software framework that supports the development of such distributed management and control systems by providing a structured, plug-and-play approach for the validation of distributed management architectures on simulation models of the controlled CPSoS, and for the subsequent deployment into operational environments. The capabilities of the framework are demonstrated on two complex, industrial use cases, the distributed management of an integrated chemical production site and the distributed optimization of a network of multi-product semi-batch reactors.