A composite generation-transmission system reliability assessment method based on clustering of optimal multiplier vector

Monte Carlo simulation (MCS) method is an important method for composite generation-transmission system reliability assessment. However, the precision of MCS results and the computing time required are contradictory to each other. Optimal multiplier vector can be used to change the probability distribution of the sample space, reduce the variance, and subsequently speed up the MCS process. Existing optimal multiplier methods either use a large number of multipliers to improve the accuracy and increase the computing time, or use a small number of multipliers to improve computing speed and reduce the results accuracy. In order to reflect the different types of components and improve computational efficiency, the sensitivity analyzing method and clustering components which have significant impact on system reliability is proposed. A clustering optimal multiplier vector with the minimum goal of pre-sample variance is proposed as well to obtain similar results calculated by conventional Monte Carlo method, but requires smaller sampling number and less computing time. The convergence speed of the presented method is verified using the IEEE-RTS 79 test system. © 2011 State Grid Electric Power Research Institute Press.