Optimum step length selection rule in modified HL-RF method for structural reliability

This paper presents the implementation concepts of a step length selection rule in a modified Hasofer-Lind and Rackwitz-Fiessler (HL-RF) method in order to improve its performance. Several step length selection rules in the context of structural reliability were studied and implemented in order to make the modified HL-RF method more robust and efficient. It has been demonstrated that a variant of Goldstein's rule, which is popularly known as the Armijo rule, is one of the most efficient step length selection rules. It is observed in this study that a proper step length selection rule will not only make the algorithm more robust but also improves the other parameters such as efficiency, capacity, etc. An iterative procedure, which enables systematic implementation into a computer program has been given. Several case studies have been taken from the literature to demonstrate and validate the proposed algorithm. A general-purpose software has been developed to solve a variety of structural reliability problems including those, which are complex and nonlinear. In addition, non-normal distributions, such as lognormal, extreme value distributions, etc., are incorporated in this software. (c) 2006 Elsevier Ltd. All rights reserved.