Genetic based algorithms for estimating synchronizing and damping torque coefficients

This paper presents an efficient digital methodology for estimating synchronizing and damping torque coefficients of a synchronous machine. These coefficients are used as a measurement of power system stability. The proposed algorithm is based on Genetic Algorithms (GAs) optimization technique, that uses digital samples of the machine time responses to perfonn the estimation process. The problem is formulated as a dynamic estimation problem. The goal is to minimize the error in the estimated coefficients. The method is tested using simulated case study. Effects of bad data measurements on the estimation process are studied. Results are reported and compared with those obtained using the least error square estimation technique. The comparison shows that the proposed method can successfully estimate the required coefficients even in very critical stable cases where other methods may fail. It is also shown that bad data has no effect on the out come results of the proposed method. The method can be considered as a very reliable and accurate tool for estimating the damping and synchronizing torque coefficient for power system stability assessment.