Abstract: The main purpose of this paper is to develop an estimation algorithm by digital computer, that extracts parameters of dynamic model of synchronous generator from test data with noises. The approach taken by this algorithm is to model the generator, uses this model to calculate the current outputs based on the initial parameter value, and then to utilize the sensitivity matrixes of parameters, adjusting the parameter estimates systemtically' to minimize the square of the difference between the measured outputs of the generator and model outputs. Furthermore, incorporating the statistics of the noise which inevitably corrupts the measurements into the estimation algorithm leads to the maximum posteriori probability estimation.