Phasor measurement unit selection for unobservable electric power data integrity attack detection

Electric power system operators make critical decisions based on remote measurements. If those measurements are compromised, it is possible to make decisions that could lead to critical consequences. Of particular concern are unobservable attacks where compromised measurements are not flagged as erroneous by bad data detection algorithms. The use of secure measurement devices, such as PMUs, is one way to recognize such attacks. In this paper we present an algorithm based on integer programming for optimal placement of PMUs to detect unobservable electric power SCADA data integrity attacks. Alternatively, this algorithm can be used to identify minimal sets of existing PMUs whose data is needed to detect unobservable bad data attacks. We show that the algorithm is efficient on practical examples drawn from the power engineering literature.