Analysis of Structural Properties of Complex and Networked Systems

Over the past decades, science and society have been experiencing systems that tend to be increasingly sophisticated and interconnected. Although it would be challenging to understand and control complex systems fully, the analysis and control of such systems can be partially realized only after applying some reasonable simplifications. In particular, for the analysis of certain control properties, such as controllability, a complex system can be simplified to a linear structured system capturing an essential part of the structural information in that system, such as the existence or absence of relations between components of the system. This thesis has studied the effect of the interconnection structure of complex systems on their control properties following a structural analysis approach. More explicitly, we have analyzed strong structural properties of complex systems. The main contributions have been split into two parts:1. We have introduced a new framework for linear structured systems in which the relations between the components of the systems are allowed to be unknown. This kind of systems has been formalized in terms of pattern matrices whose entries are either fixed zero, arbitrary nonzero, or arbitrary. We have dealt with strong structural controllability and the solvability of the FDI problem of this kind of linear structured systems.2. We have introduced a novel framework for linear structured systems in which a priori given entries in the system matrices are restricted to take arbitrary but identical values. Several sufficient algebraic and graph theoretic conditions were established under which these systems are strongly structurally controllable.Finally, in the outlook subsection, we have suggested some future research problems concerning the analysis of strong structural properties of complex systems.