"Phasor measurement units (PMUs) provide measurements with high precision at a high resolution (up to 50 samples per second). These measurements are synchronized and time-stamped using the Global Positioning System. Despite these advantages, the industry has been slow in adopting PMU technology due to the high cost of installing PMUs. Therefore, PMU locations must be judiciously selected through optimal placement of PMUs (OPP), which enables the minimization of installation cost. This dissertation examines the OPP problem from several perspectives. First, the OPP problem definition is re-examined since most OPP literature associates the PMU installation cost with the PMU unit. Most techniques in the literature have proposed to minimize the number of PMUs while considering the complete observability of the system. However, PMUs require sufficient infrastructure to be in place before they can perform most of their intended functions. Therefore, the OPP problem should be reformulated to include the supporting infrastructure of PMUs. The proposed OPP formulation is implemented by using a bi-level framework. This framework can accommodate different varieties of OPP, such as single-stage, multistage, and application-based approaches. Moreover, the proposed framework achieves the optimal solution with the maximum observability in the case of multiple optima. Second, this dissertation examines application-based OPP approaches, where specific technical benefits are prioritized over the cost of the OPP. Three applications are proposed. The first application is a fault-tolerance based OPP approach, where the network fault-tolerance is enhanced by deploying PMUs to the vulnerable elements in the network. In the second application, a voltage stability criterion is developed and proposed, where the critical buses are identified and prioritized for PMU allocation. The third application addresses false data injection attacks (FDIAs), where the system topology is used by the adversary to bypass the bad data detection of state estimators. The proposed OPP approach enhances bad data detection against FDIAs by utilizing the PMUs as authenticators for each other."--Pages ii-iii.
Read
