Data reduction via clustering and averaging for contingency and reliability analysis

摘要

Data reduction is necessary when the choice of analysis method cannot deal with large data sets. This is for instance the case when a power market model generates future generation and load scenarios, and one wants to use these scenarios as a basis for contingency and reliability analysis. A framework that forms a reduced data set, which keeps the important information from the full data set intact, is presented. The framework uses statistical methods to find patterns in the data, and use averaging to reduce the data set. A case study where the framework is used in a reliability analysis is included.