EPSO vs. Monte Carlo

This paper reports the application of a population based method (EPSO – Evolutionary Particle Swarm Optimization) to calculate power system reliability. Population based methods appear as competitors to the traditional Monte Carlo simulation because they can be much more computationally efficient in estimating a number of reliability indices. The work reported in this paper demonstrates that EPSO variants, suited to the problem of exploring a zone in the state space instead of searching for a single optimizing state, are efficient in calculating a number of system reliability indices such as power not supplied. The results obtained with EPSO are compared with Monte Carlo and with work of other researchers in population based methods.

Epso reliability submitted v1.pdf — PDF document, 523Kb