Global sensitivity analysis of structural seismic demand based on information entropy

To improve the computational efficiency of global sensitivity analysis (GSA) for complex structures, this study proposed a new importance analysis method (IE) based on the low deviation sequences and orthogonal polynomials to study the influence of parameters' uncertainty on three structural seismic demands. A comparative investigation utilizing nonlinear time history analysis for these seismic demands was conducted using OpenSEES. The variance-based importance analysis method and the Tornado graphic sensitivity analysis method were employed to validate the accuracy of the proposed approach. The results regarding the order of importance are nearly consistent across methods, demonstrating the effectiveness of our proposed method. Notably, the sample size required by this new method is only 1024 to achieve reliable results, which is significantly lower than existing sampling methods that necessitate thousands of samples for effective importance analysis; thus, enhancing overall efficiency. Furthermore, the findings indicate that the influence of representative value of gravity load (M_s) on seismic demands is relatively substantial, whereas the influence of modulus of elasticity of concrete (E_c) is comparatively minor.