|کد مقاله||کد نشریه||سال انتشار||مقاله انگلیسی||ترجمه فارسی||نسخه تمام متن|
|805272||1468214||2016||9 صفحه PDF||سفارش دهید||دانلود رایگان|
• A new RBDO by using adaptive response surface and first-order score function is developed.
• Bayesian metric is proposed to select the best response surface considering data uncertainty.
• First-order score function is proposed to compute the sensitivities of probabilistic response.
This study presents a new Reliability-based Design Optimization method using adaptive response surface and first-order score function analysis for complex system design optimization considering the variability of design variables. The adaptive response surface using Bayesian metric and Gaussian process based model bias correction method, is developed to represent the true performance functions and replace the true limit state function. First-order score function analysis is exploited to compute the sensitivities of probabilistic responses with respect to the design variables, which are the mean values of the random variables. Numerical results indicate that the proposed methods can produce the best response surface and estimate the sensitivities of probabilistic responses accurately. The proposed methodology is demonstrated by a vehicle crashworthiness design optimization problem with full frontal and offset frontal impacts.
Journal: Reliability Engineering & System Safety - Volume 156, December 2016, Pages 125–133