Effects of correlations on sufficient controllability of complex networks

If a dynamic system can be driven from any initial state to any desired state in finite time by applying appropriate external signals to it, the system is controllable. When it comes to the sufficient controllability of a network, its purpose is to minimize the number of external controllers, so as to ensure that a sufficient number of nodes in the network can be driven to any desired state in finite time and energy input. Here, we study the influence of various network characteristics on the sufficient controllability of complex networks. The three most commonly used correlation coefficients in network research are degree correlation coefficient, modularity correlation coefficient, and clustering correlation coefficient. In this dissertation, we carry out rewiring operations based on existing algorithms on Erdős-Rényi networks to generate the needed degree correlation coefficients and analyze the influence of degree correlation coefficients on the sufficient controllability of complex networks, and we provide two improved algorithms based on existing algorithms to study and analyze the influence of modularity correlation coefficients and clustering correlation coefficients on the sufficient controllability of complex networks. Also, it should be noted that all of the algorithms used or proposed in this dissertation are degree-preserving algorithms. The purpose is to avoid the influence of the degree change of nodes on the research.Master of Science (Computer Control and Automation