An efficient technique for reliability analysis of power systems including time dependent sources

Abstract

A method of reliability analysis of electric power systems with time-dependent sources, such as photovoltaic and wind generation is introduced. The fluctuating characteristic of unconventional generation units affects the reliability of the generation system in a different manner than the conventional units. The method proposed by the authors groups the units into several subsystems. One subsystem contains the conventional units and the remaining subsystem consist of unconventional units. A generation system model is built for each subsystem. The outputs for the unconventional units and the load are treated as correlated random variables. Using a clustering procedure states are identified wherein for a given value of load there are specific mean values of the outputs of the unconventional units. Reliability analysis is performed by combining the conventional subsystem with the unconventional subsystems in each state. The output from all the states is combined to compute the loss of load expectation and expected unserved energy.<>