The project will develop new methods for the distributed management of large physically connected systems with distributed autonomous management and global coordination. The research will be driven by case studies in electrical grid management and control, including the charging of electric vehicles, and industrial production management.
The methods that will be investigated for the management of large physically coupled systems of systems are:
- Population-control techniques that are motivated by the behavior of biological systems;
- Market-like mechanisms that try to achieve global optimality by the iterative setting of prices or resource utilization constraints
- Coalition games, where agents group dynamically to pursue common goals.
The properties of the distributed management and control techniques are investigated, and they are validated in large-scale simulations of case studies provided by industrial partners. From the validation for the case studies, general conclusions will be drawn about the suitability of the proposed distributed management and control mechanisms for certain classes of systems of systems. This will provide guidelines for the design of evolving systems of systems with respect to the interplay of local autonomy and global management. The project will include the development of methods and tools for the engineering and implementation of advanced management solutions for systems of systems.
The planned outcomes of the project are:
- Innovation in distributed management methods for complex interconnected systems of systems;
- Progress in methods for the rigorous analysis and validation of systems of systems;
- Demonstration of advanced methods in realistic large-scale simulations of real use cases of systems of systems;
- Reduction of the carbon footprint and of the resource consumption of industry and of electric power generation and distribution, including charging of electric vehicles;
- Identification of technology gaps in advanced management and coordination methods for SoS and their implementation.