Independent Estimation of Generator Clustering and Islanding Conditions in Power System with Microgrid and Inverter-Based Generation

The use of phasor measurement units (PMU) allows us to obtain synchronized measurements of various points in the network and whit them analyze the stability of power systems. This chapter presents an algorithm based on participation factors to estimate generator clustering and to evaluate its application on controlled islanding on a power system, with distributed generation, using the data from PMUs after a severe disturbance. The proposed islanding detection method uses the data obtained from PMUs to represent the dynamics of the entire power system and form a measurement matrix, updated using a sliding window, containing the angles of the voltage phasors. Then, a covariance matrix is computed, and the eigenvalues and eigenvectors of this matrix are obtained. Subsequently, the most energetic eigenvalue is identified, and its participation factors are calculated. The participation factors are used as a contribution measurement of each generator into the most energetic eigenvalue, i.e., they will show the contribution made by each one after a disturbance. The clusters will be formed by generators sharing the same participation level. Controlled islanding condition of the system will be evaluated by using the clustering schemes proposed in the literature.