Cyber-physical system failure analysis based on Complex Network theory

Cyber-physical system studies request revolutionary methods to identify the vulnerability of the system in order to develop risk assessment and countermeasures and consequently reduce the vulnerabilities. One possible approach to analyze cyber-physical system vulnerability is to study their topology and therefore to quantify the interdependencies on the physical side and on the cyber side. This paper presents a three-dimensional Complex Network model for a coupled electric power system and information and communication technology network, where power buses and communication network routers/multiplexers are modelled as nodes while power lines and communication channels are modelled as edges. The vulnerability of each system component may vary under different system operating conditions due to the change of system's structure. Therefore a Vulnerability-weighted Node Degree is introduced to assess the vulnerability of each component under two characteristic power system operating conditions. The most vulnerable components are identified using Vulnerability-weighted Node Degree and the consequences of failure of these components are demonstrated with Node Removal method.