Quasi Monte Carlo method for reliability evaluation of powerrnsystem based on Dimension Importance Sorting

摘要

Quasi Monte Carlo (QMC) is a type of improved Monte Carlo (MC) method,rnalthough its improvement over MC is known to degrade with large‐dimensionrnproblems. This study proposes a Dimension Importance Sorting (DIS) method tornsolve this degradation problem, so QMC becomes feasible for large‐dimensionrnreliability evaluation of power systems. First, the error estimation principle ofrnQMC is presented, and the uniformity degradation of low discrepancy sequencernon high dimension is analyzed. The concept of analysis of variance is then employedrnto derive the decomposition term of error bounds for QMC, and a sampling methodrnthat uses Spearman correlation coefficient to sort the dimension importance isrnproposed to reduce the error bounds. Finally, 2 methods of reliability evaluation,rnQMC based on DIS (QMC‐DIS) and cross entropy QMC‐DIS (CE‐QMC‐DIS),rnare built to improve the conventional MC and CE, respectively, on the accuracy ofrnreliability indices. The effectiveness of proposed methods is demonstrated on thernReliability Test System‐79 generation system, a 292‐unit generation system, andrnthe Roy Billinton Test System generation and transmission system.