Abstract: In order to solve the small sample reliability problems for structures and mechanisms in real applications,an analyzing framework named Bootstrap-support vector regression-second order cumulant (Bootstrap-SVR-SOC) method together with a sample size extension method combining multi-order moments that could make the extended samples more objective were proposed. Bootstrap was combined with support vector regression method in order to build the limit state function for reliability analysis based on the second order cumulant approach. Finally, numerical case of a crank shaper mechanism was analyzed to demonstrate the effectiveness of the proposed framework.

Key words: sample extension, structural and mechanism reliability, small sample, bootstrap, support vector regression, second order cumulant