General formulation of transfer function identification for power system based on WAMS

With interconnection of power grid, power system scale becomes larger, and operating condition becomes more variable. Rapid identification of system model is helpful for online analysis and wide-area control of power system. Based on wide-area measurements, a general formulation of Prony algorithm is brought forward to identify power system transfer function. Principles of input signals and sampling parameters selection are summed up. This method allows initial condition of output signal is nonzero, and it can deal with any input signal types. Most importance, computation time of this input-output data based identification will not evidently be affected by power system scale. Simulation test on numerical example shows some improvement of this general method. Example of transfer function identification is carried out on 16-machine 68-bus power system. Comparison of results obtained by the proposed approach and conventional eigenvalue analysis techniques is provided.