Proceedings of the Third ACM International Conference on Distributed Event-Based Systems
| Hi-index | 0.00 |
Electric power grids are complex interconnected systems that in North America can span more than a thousand miles. They must be operated such that the dynamic balance between supply and demand is maintained. Grids exhibit many power dynamics that are global over the entire grid, yet their control mechanisms and almost all of their operational data on the current dynamics and configuration of the grid status data are local in nature, typically limited to a single substation. The reason for this is that the communication system that is used to coordinate and monitor utility operations was designed largely in response to the 1965 blackout in the US Northeast. However, since then network and related technologies have improved dramatically. Status dissemination middleware is a new specialization of the publish-subscribe model that takes advantage of the semantics of status data. This dissertation presents a status dissemination middleware framework named GridStat for the electric power grid and other critical infrastructures. It takes advantage of the semantics of status data to optimize delivery and to manage its subscriptions for quality of service. In doing so, GridStat allows status information to be disseminated over a wide-area. The architecture, design, and implementation of this framework are described, along with its baseline mechanisms to improve efficiency and resilience. In addition, the rationale, design, and implementation of four mechanisms are presented. These mechanisms are limited flooding mechanisms, operating modes mechanism, filtering with multicast mechanism, and condensation function mechanism. This dissertation also presents an experimental evaluation of the GridStat framework.