A genetic based technique for identification of symmetrical components for power system protection

This paper presents a novel technique for identifying and measuring the symmetrical components of three phase voltage or current waveforms in electrical power systems. The proposed technique is based on genetic algorithm optimization technique. The problem is formulated as an estimation problem and presented in state space form. The goal is to minimize the error associated with the estimated parameters. The proposed algorithm is then used to estimate the positive, negative and zero components of unsymmetrical waveforms. The method is tested using practical study cases. Different factors that affecting the identification process such as data window size, sampling frequency, and number of samples are studied. Effects of bad data on the solution accuracy are also studied. Results obtained show that the proposed technique can estimate and track the symmetrical components of non-stationary three phase unbalance voltage or current waveforms in noisy environments. Fast accurate solutions are guarantee regardless the initial conditions. It is also shown that bad measurements have no effects on the final accuracy of the estimation.