Uncertainty analysis of an offshore jacket-type platform using a developed numerical model updating technique

Autor: Alireza Mojtahedi, Samira Mohammadyzadeh, Hamid Hokmabady
Rok vydání: 2020
Předmět:
Zdroj: Ocean Engineering. 211:107608
ISSN: 0029-8018
DOI: 10.1016/j.oceaneng.2020.107608
Popis: To investigate the dynamic behavior of a complex system, an accurate updated finite element (FE) model is essentially required. Numerical model updating, which is performed based on valid experimental data, is imprecise without taking the related uncertainties into account. In this study, both structural and parametric uncertainties are considered within model updating of an offshore jacket platform through time history analysis for the first time. The experimental test is implemented employing a shake table and a scaled hydro-elastic model of the case study. For updating the FE model, a three-step updating procedure is implemented through eight distinct scenarios. In the first step, three unknown structural parameters are calibrated over ten distinct groups of elements in the FE model by using particle swarm optimization algorithm (PSO). In the second step, the damping matrix coefficients are also updated through the optimization process. In the third step, the model form error is considered utilizing four mathematical functions to decrease the parametric uncertainty and improve the FE model accuracy. Based on the results, by employing the proposed methodology, approximately 98 percent fitness is observed between the experimental and updated numerical model. According to the results, not only reducing both structural and parametric uncertainties is essential, but also calibrating the damping matrix for updating a numerical model and improving the FE model accuracy is of great importance. The developed methodology, which is applied to a sophisticated structural system, is strongly recommended for updating the systems that existence of an accurate updated numerical model is vital.
Databáze: OpenAIRE
Externí odkaz: