Synchronous machine parameter estimation in the short-circuit test using the damped Gauss-Newton method

This paper presents a digital simulation approach to obtain an adjusted set of parameters of a synchronous machine, during the short-circuit fault test. In the context of power system planning, security and operation, it is essential to determine the parameters that represent the power system generators with high accuracy. A technique involving unconstrained nonlinear programming is implemented, using a fifth order model of the synchronous machine, during a short-circuit fault. Testing the proposed procedure comprises a stage of experimental data generation using some known fixed parameters, while a different set of parameters are used to obtain a model response. These data sets are then employed to minimize the residues between responses, in the least squares sense, using the damped Gauss-Newton method. The results show that, even in presence of measurement noise, the error between adjusted and real parameters is minimum, and this is also corroborated by a low-magnitude residuals graph.