The electric power grid is one of the nation’s most important critical infrastructures and there is considerable concern about its vulnerability to malicious physical and cyber attacks. The detailed data describing grid topology and components is considered sensitive and can only be shared through a Critical Energy Infrastructure Information (CEII) nondisclosure agreement. This tightly controlled access to power grid information limits advances in power system research and in power system operations. Researchers who work with the data must mask their results in publication, and, therefore, their results cannot be openly and independently verified by other researchers. Operation of power systems must tend towards the use of centralized and dedicated computing and controls, around the model’s access points, essentially eliminating the possibility of using advanced shared computing platforms, such as the so-called cloud computing, and raising major concerns about distributed computing and control in this domain.
This seminar discusses the development of a means to mask power system information that will allow computation on power system models without revealing the sensitive details in the models. The work is motivated by research in cloud computing on masking optimization models through a series of transformations. In our work, we show that these techniques can be used to preserve confidentiality of the DC and certain AC optimal power flow problems.
Bernie Lesieutre is a Professor of Electrical Engineering at the University of Wisconsin-Madison. He received his B.S., M.S., and Ph.D. degrees from the University of Illinois. His research interests focus on the modeling, monitoring, and analysis of electric power systems.