An alternating least squares approach is developed in this paper to identify the exponential recovery dynamic load model of wide-area power systems. The nonlinear optimization problem is decomposed to two linear least squares problems, and solved in an alternating way. Then, a new algorithm for numerical derivative calculation using discrete Fourier transform is proposed to attenuate the effect of noises in the process of parameter estimation. Based on the estimated dynamic load characteristics, the application on voltage stability is analyzed. Finally, numerical and laboratory examples are conducted to demonstrate the effectiveness of the proposed methods.
