Abstract:There are not only random uncertainties in structural engineering, but also fuzzy uncertainties. Aiming at fuzzy reliability problem of implicit structure performance function, an ellipsoidal convex model-response surface method based on fuzzy decomposition theorem is proposed in this paper. Firstly, the fuzzy variable is transformed into the interval variable at the confidence level [0, 1] by determining the probability distribution function of the structural reliability. Secondly, the super ellipsoidal convex model is established, and the super ellipsoid is converted into a unit hyper sphere, and the non-probabilistic reliability index is obtained, which is expressed as the shortest distance from the origin to the limit state surface. Finally, the response surface method is used to solve the structural non-probabilistic reliability. The central composite design is used to generate the test points. The reliability index is obtained by finding the shortest distance point from the origin to the limit state surface through multiple iterations, and the structural fuzzy reliability of the traversal confidence level is obtained. The simulation of the examples shows the effectiveness and accuracy of the proposed method, which provides a feasible way for structural fuzzy reliability analysis and calculation.