Genetic algorithms approach for the evaluation of composite generation-transmission systems reliability worth

This paper presents a genetic algorithm (GA) based method for evaluating reliability worth indices of composite power systems. The GA is used as a state sampling tool for the composite power system network. Two different approaches have been developed. In the first approach, GA samples failure states for each load level separately. Thus reliability worth indices are calculated for each load level and then combined to obtain the annual reliability worth indices. In the second approach, GA samples failure states with load buses assigned the maximum load state. Failures states are then reevaluated with lower load states until a success state is obtained or all load states have been considered. In both approaches GA is able to trace failure states in a more intelligent and faster manner than conventional methods. An optimization model based on linearized load flow is used for the evaluation of sampled states. Two different objectives are used in state evaluation. The first objective is to minimize load curtailment considering load category and load bus relative importance. The second objective is to minimize load interruption cost. Instead of using the raw interruption cost associated with failure state mean duration time, random sampling is used to calculate mean interruption cost associated with each failure state. Case studies on the RBTS test system considering different state evaluation methods and cost calculation methods are introduced. Results obtained from the two different approaches are compared and analyzed.