EPPEI Power Seminar #5

Presenter:	Nader Samaan Ph.D. Student (Advisor: Dr. C. Singh)
Topic:	A New Method for Composite System Annualized Reliability Indices Based on Genetic Algorithms
Time:	October 11th, Friday at 2:00pm
Location:	128D ZEC

ABSTRACT:

Reliability analysis of large power networks when generation and transmission are considered together is a complex and computationally difficult problem. Two approaches have been established to evaluate adequacy indices of composite power systems. The first approach is based on analytical evaluation. The second is based on Monte Carlo simulation techniques of either random sampling or sequential sampling.

A new approach for state sampling of composite power system based on genetic algorithms (GA) is presented. Sampled states are used to assess annualized reliability indices. In the proposed method GA intelligently searches the enormous state space of a power system to find the most probable states contributing to system failure. Binary encoded GA is used to represent system states. Through its fitness function GA is able to trace failure states in a more intelligent manner than conventional methods. A linearized optimization load flow model is used for evaluation of sampled states. The model takes into consideration importance of load in calculating load curtailment at different buses in order to obtain a unique solution for each state.

The full set of composite system adequacy indices and load bus indices is calculated. The proposed method is applied to a simple test system to be validated. Obtained results are compared with other conventional methods.

Presentation slides