Abstract

The current performance-based building design considers maximum interstorey drift (MISD) ratio as the main structural performance indicator. Observations from past earthquake and reported studies, however, have highlighted that residual interstory drift (RISD) ratio has become an important factor in assessing post-earthquake safety of buildings, and decision in economic feasibility of repair and reconstruction. Improving post-earthquake performance evaluation of buildings enables decision-makers prioritise repair and tag high-risk buildings. The MISD and RISD are subject to uncertainties and have non-linear relation with the input parameters. Thus, in this paper, analytical surrogate model of MISD and RISD ratios are developed using Gaussian process (GP). To show utility of the GP model, a new hybrid building system, cross laminated timber (CLT)–steel moment resisting frame hybrid system, was considered. The hybrid building was design for the seismicity of Vancouver, BC, and meets the current steel design code. For the GP surrogate model, the hybrid building input parameters considered were: infill pattern of the CLT, bracket spacing of the connection between the CLT and steel frame and panel thickness and strength of the CLT. In addition, sensitivity of four ground motion indicators was considered as surrogate input into the GP model: peak ground acceleration, ratio of peak ground acceleration/peak ground velocity, Arias intensity measure and significant duration. In general, the GP model showed good predictive performance of MISD and RISD ratios. In particular, the best predictions were obtained using the ratio of peak ground acceleration/peak ground velocity as a covariate.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call