Protection Placement for Power System State Estimation Measurement Data Integrity

Abstract

In this article, a protection placement problem for power network measurement system data integrity is considered. The placement problem is motivated from [G. Dan and H. Sandberg, "Stealth attacks and protection schemes for state estimators in power systems," in Proc. 1st IEEE Int. Conf. Smart Grid Commun., 2010, pp. 214-219], where supervisory-control-and-data-acqui-sition and phasor measurement unit measurements can be selectively encrypted and protected to prevent false data injection attacks. The problem is shown to be NP-hard and an integer linear programming formulation is provided based on the topological observability condition by [G. Krumpholz, K. Clements, and P. Davis, "Power system observability: A practical algorithm using network topology," IEEE Trans. Power App. Syst., vol. PAS99, no. 4, pp. 1534-1542, Jul. 1980.]. The description of the observability condition requires a graph connectivity constraint, which is found to be most effectively described by the Miller-Tucker-Zemlin conditions. For a specialization without line power flow measurements, the protection placement problem can be modeled by a domination-type integer linear program that is much easier to solve than the general formulation requiring graph connectivity. Numerical studies with IEEE benchmark and other large power systems with more than 2000 buses indicate that by using the proposed formulations, the protection placement problem can be solved in a negligible amount of time in realistic application settings.