In the paper, the results of investigations on the location of generating units most affecting the angular stability of a large power system (PS) are presented. For their location, the eigenvalues of the PS model state matrix associated with electromechanical phenomena (electromechanical eigenvalues) were used. The eigenvalues were calculated on the basis of the analysis of the disturbance waveforms of instantaneous power of the generating units operating in the PS. The used method of calculating eigenvalues consists in approximation of the disturbance waveforms of generating units by the waveforms being the superposition of modal components. The parameters of these components depend on the sought eigenvalues and their participation factors. The objective function was defined as the mean square error between the approximated and approximating waveforms. To minimize it, a hybrid algorithm, being a combination of genetic and gradient algorithms, was used. In the instantaneous power waveforms of generating units most affecting the PS angular stability, the least damped or undamped modal components dominate. They are related to eigenvalues with the largest values of real parts. The impact of individual modal components on the disturbance waveforms of subsequent generating units was determined with the use of participation factors and correlation coefficients of electromechanical eigenvalues.

This study suggests a new algorithm based on a combination of fuzzy logic and genetic algorithm (GA) to improve voltage profile in a microgrid. The considered microgrid includes control variables such as onload tap changer (OLTC), active power output from distributed generators (DG) and reactive power output from feeder switched capacitors that are controlled in a microgrid controller (MGC) by communication links. The proposed method was used to obtain the optimum value of control variables to establish voltage stabilization in varying load condition as online. For establishing voltage stabilization at the microgrid, an objective function is defined and is tried to minimize it by control variables. The control variables were changed based on fuzzy logic and the GA was employed for finding the optimum shape of membership functions. In order to verify the proposed method, a 34 buses microgrid in varying load condition was analyzed and was compared with previous works.