Bootstrap methods for stochastic prediction of voltage sags

This paper illustrates the use of resampling bootstrap methods for stochastic prediction of voltage sags from limited monitoring data. Using results from phase II of EPRI´s distribution power quality study; we illustrate different resampling techniques that can be used to estimate a system´s power quality. These resampling methods are compared with parametric approach, which assumes a certain probability distribution. Resampling techniques are powerful tools in analyzing variability and uncertainty in nonnormal data sets in power distribution systems.