Estimation of Parameters of a Dynamic Generator Model from Modal PMU Measurements

This paper introduces a novel methodology of parameter estimation for the power system models using the values of modes and mode shapes extracted from the PMU measurements. The main idea is to force the state matrix A, which is a function of the unknown parameters, to satisfy as closely as possible the eigenequation linking the state matrix, the modes (eigenvalues of A) and the mode shapes (eigenvectors of A). The methodology is derived and analyzed for a general case of any model's parameters. The applications to the estimation of generator inertia M coefficients are then discussed in detail. The proposed methodology has been tested using two different system models. The results confirm the efficiency of estimating M using the PMU data, while the estimation of D is shown to be very sensitive to the measurement noise.