In this research we conducted a sensitivity experimental study where we explored the dependency of the shear strain on the seismic properties of bearings, namely lead rubber bearing (LRB) and super high damping rubber bearing (SHDR). The factors studied were vertical pressure, temperature, shear modulus of the inner rubber (G value), loading frequency, and loading sequence. Six specimens were adopted, i.e., three LRBs and three SHDR bearings. A series of test plans were designed. The seismic characteristics of the bearings were captured through a cyclic loading test, which included post-yield stiffness, characteristic strength, area of a single cycle of the hysteretic loop, equivalent stiffness, and equivalent damping ratio. A whole analysis of variances was then conducted. At the same time, to explore certain phenomena caused by the factors, an extended discussion was carried out. Test results showed that the temperature is the most dominant feature, whereas the G value is the least contributing factor, with the effect of the loading frequency and the loading sequence found between these two. The increment of the post-yielded stiffness for LRB from 100% to 25% is a significant reduction from a low temperature to high one. The slope of the characteristic strength versus the shear strain for LRB under high temperature is larger than the one under low temperature.

A response surface method for reliability analysis based on iteratively-reweighted-least-square extreme learning machines (IRLS-ELM) is explored in this paper, in which, highly nonlinear implicit performance functions of structures are approximated by the IRLS-ELM. Monte Carlo simulation is then carried out on the approximate IRLS-ELM for structural reliability analysis. Some numerical examples are given to illustrate the proposed method. The effects of parameters involved in the IRLS-ELM on accuracy in reliability analysis are respectively discussed. The results exhibit that a proper number of samples and neurons in hidden layer nodes, an appropriate regularization parameter, and the number of iterations for reweighting are of important assurance to obtain reasonable precision in estimating structural failure probability.