RESEARCH ON IMPROVEMENT OF RELIABILITY INDEX BASED ON SAMPLING POINTS ON LIMIT STATE SURFACE

摘要

The analysis result is less accurate with the first order reliability method (FORM) when the limit state surface is highly nonlinear. The second order reliability method (SORM) can provide some accuracy, but its analysis precision is sensitive to the design point solution. In practical structural reliability analysis a certain function model is chosen to restructure the performance function since it is implicit. So the approximate design point will lessen the analysis accuracy with the second order reliability method. In this paper a method is proposed with the sampling point on the real limit state surface to improve reliability index based on the second order reliability method. It considers the feature of the limit state surface and approximate accuracy. Numerical examples show that the method is of high precision.