A new computational tool for composite reliability evaluation

The authors describe the modeling assumptions and computational aspects of a computational tool for composite reliability evaluation, CREAM. The reliability evaluation methodology for power systems is based on Monte-Carlo sampling with a variance reduction scheme which allows the incorporation of user experience or of analytical models as regression variables. The component of reliability indices explained by these regression variables is factored out of the Monte-Carlo sampling scheme, thus reducing computational effort. The approach is illustrated in case studies with 500-bus systems.<
>