Modeling and Mitigating Cascading Failures in Interdependent Power Grids and Communication Networks

Many of today’s critical infrastructures are organized in the form of a network, which are dependent on one another. A particular example is the power grid and the communication network used to control the grid. While this dependence is beneficial during normal operation, as it allows for more efficient operation, it can be harmful when the networks are under stress. Indeed, in such interdependent network infrastructures, a cascade of failures may occur where power failures can lead to communication failures, which, in turn, lead to cascading power failures.
The goal is to develop models for both the impact of power failures on the communication networks, and the impact of loss of communications and control on the power grid; analyze the vulnerability of interdependent networks to natural disasters and attacks and finally design network protection and dynamic control techniques to minimize the impact of failure cascades in interdependent power grids and communication networks. In the following, we describe some of our preliminary results.

We developed a basic interdependence model for the power grid and the supporting communication and control network. Under this model, a failure in the power grid may cause a failure in the communication network and vice versa. The result of showed that interdependent networks are more vulnerable to failures than networks in isolation. Further, in a power grid the flows are driven by Kirchoff’s laws, and cannot be described by a network flow model. Thus, when a failure occurs in a power grid, the power flow is redistributed through the rest of the network and some elements could overload and fail, leading to “Cascading Failures” within the power grid. We showed that it is critical to consider the actual power flow in analyzing the behavior of the power grid. We also proposed a two-phase load shedding scheme to control the cascade of failures both inside and between the networks, and showed that our control policy performs close to optimal for many scenarios.

We studied the reliability of power grid under regional (geographical) disasters. We quantified the effect of large-scale non-targeted disasters and their resulting cascade effects on the reliability of the power grid and a dependent communication network.