An approach to assess structural reliability in presence of random and fuzzy variables

Abstract

   Assessment of structural reliability in case of the mixture of random and fuzzy input variables always exists in engineering problems, but has a lot of challenge, because there is no unique solution in mathematics. A lot of approaches had been proposed to solve this problem. This article proposes a new appoach, based on innovating and combining the insufficient reason and the maximum specifity principles for transformations between random and fuzzy variables, to determine the normal random variables, that are equivalent to the symmetric triangular fuzzy number. From these random variables, the original problem is converted to the basis structural reliability problems, and can apply the familiar methods of the traditional reliablility theory to calculate. Meanwhile, this article proposes the way to determine the central reliability and the marginal reliability. Numerical results are compared with the results of the existing method, to demonstrate the acuracy and effectiveness of the proposed method.