Influence of parametric uncertainties and their interactions on small-signal stability : a case example of parallel-connected active loads in a DC microgrid

Classical stability analysis techniques based on nominal models do not consider the uncertainty of system parameters, their interactions, and nonlinearity, which are important characteristics of practical highly coupled microgrids. In this work, variance-based sensitivity analysis is used to identify parameter combinations that have a significant impact on the small-signal stability of a microgrid featuring two parallel active loads. The analysis indicates that the effectiveness of source-side damping is reduced when resonant frequencies of load input filters become matched. Further results using derivative-based sensitivity analysis reveal that source-side resistance can exhibit drastically different effects on the stability if load input filter resonant frequencies are matched with respect to the case when they are well separated. These behaviours are verified using time-domain switching models.