Smart grid initiatives are becoming more and more achievable through the use of information infrastructures that feature peer-to-peer communication, monitoring, protection and automated control. The analysis of smart grid operation requires considering the reliability of the cyber network as it is neither invulnerable nor failure free. The objective of this dissertation is to categorize interdependencies between cyber and power networks and propose mathematical evaluation models to calculate the reliability of the power network when considering failures of the cyber network. This study categorizes interdependencies between cyber and power networks into direct and indirect. In this research direct interdependencies among cyber-power networks is studied and the concept of state mapping is proposed to map the failures in the cyber network to the failures of the power network. The impact of indirect interdependencies on the reliability of power system is different and more complicated than that of direct interdependencies. In this dissertation, various aspects of smart monitoring, as an application of indirect interdependency, are discussed and a mathematical model to assess its impact on power grid reliability is proposed. Based on a multiple-state Markov chain model, the failure and repair rates of power components with and without monitoring provisions are determined and compared. In addition, to model indirect interdependencies between cyber and power networks, the concept and formulations of state updating are proposed to update the probability of states due to failures in the cyber network. Furthermore, in order to evaluate the impact of both direct and indirect cyberpower interdependencies on the reliability indices, two optimization models are introduced to maximize the data connection in the cyber network and minimize the load shedding in the power network.

No description available.