A radial basis function (RBF) based sequential surrogate reliability method(SSRM) is proposed, in which a special optimization problem is solved to updatethe surrogate model of the limit state function (LSF) iteratively. Theobjective of the optimization problem is to find a new point to maximize theprobability density function (PDF), subject to the constraints that the newpoint is on the approximated LSF and the minimum distance to the existingpoints is greater than or equal to the given distance. By updating thesurrogate model with the new points, the surrogate model of the LSF becomesmore and more accurate in the important region with a high failure probabilityand on the LSF boundary. Moreover, the accuracy of the unimportant region isalso improved within the iteration due to the minimum distance constraint. SSRMtakes advantage of the information of PDF and LSF to capture the failurefeatures, which decreases the number of the expensive LSF evaluations. Sixnumerical examples show that SSRM improves the accuracy of the surrogate modelin the important region around the failure boundary with small number ofsamples and has better adaptability to the nonlinear LSF, hence increases theaccuracy and efficiency of the reliability analysis.
展开▼
