Variance-Based Sensitivity Analysis of the Composite Dynamic Load Model

As dynamic load models become more granular, the number of parameters increases and sensitivity analysis studies become necessary to rank the importance each parameter has on the transient quality. One important objective of these studies is to determine how model parameters interact with each other and the effect of this interaction on the transient. In contrast with traditional local sensitivity analysis, global, variance-based sensitivity analysis attempts to consider the whole variation range of the parameter inputs. Variance-based sensitivity analysis has been recently introduced in the power system literature in a variety of settings such as power flow and small-signal stability studies. In this work we introduce variance-based sensitivity analysis for the study of the composite dynamic load model. In particular, we will employ variance-based sensitivity to analyze the parameter dependencies of the stalling behavior of the single-phase motor (Motor D) on the transient voltage recovery.