A Two-Stage Data-Driven Algorithm to Estimate the System Inertia Utilizing Event-Driven Disturbed PMU Measurements

The power system inertia can determine the ability of power system to keep synchronization in the case of short-term power imbalance. With the increasing penetration of renewable energy resources with less system inertia, accurate estimation of the power system inertia has become critical. To solve this challenging task, a two-stage data-driven estimation algorithm is proposed in this paper. First, analytical expressions of the frequency response under the steady-state and that of transient oscillatory components are derived first by integrating the low-order system frequency response model with the first-order turbine model. Then, based on this new parametric model, a two-stage estimation algorithm is developed. System parameters of oscillatory components can be extracted from PMU measurements, signal parameters, and rotational invariance techniques (ESPRIT) at the first stage. A weighted nonlinear least square approach can be applied at the second stage to estimate the system inertia, damping coefficient, turbine time constant, and regulation coefficient simultaneously by utilizing frequency measurement data from PMUs and parameters estimated from the ESPRIT algorithm. Finally, in order to validate the effectiveness of the proposed method, simulation studies of IEEE 39-bus system will be investigated first. Historical PMU measurements from Taiwan Power Systems will also be studied. Comparison studies with other existing methods are also performed to demonstrate the advantage of the proposed method.