.png){kind=icon}
Return
An improved reliability-based robust design optimization method using Bayesian seemingly unrelated regression and multivariate loss function
.png){kind=icon}
PRE
AI
DOI：10.1007/s00158-022-03172-6.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
At present, loss function-based reliability-based robust design optimization (RBRDO) methods have been widely used to ensure reliable and robust product designs. However, conventional methods suffer from a problem that they did not consider the variance-covariance structure of responses in the loss function-based optimization strategy, which leads to an undesirable design. To address the problem, this paper proposes an improved loss function-based RBRDO method. First, the proposed method adopts Bayesian inference for the seemingly unrelated regression (SUR) model. Bayesian SUR models are applied to construct accurate response models when responses are correlated and model parameters are under uncertainty. Second, the proposed method integrates Bayesian SUR models and multivariate loss function approach into the loss function-based optimization strategy to consider the variance-covariance structure of responses. Thus, the three properties (i.e., bias, robustness, and quality of predictions) are simultaneously considered for objective functions in the loss function-based optimization strategy, in which the expected responses optimized to be close to their design targets, and the variances of both the observed and the predicted responses to be small at an optimal design. The proposed method was applied to the design of a bridge-type amplification mechanism to demonstrate its effectiveness. Results revealed that the proposed method provides a better solution than existing methods in terms of optimal design.
Keywords:
Reliability-based robust design optimization
Loss function
Seemingly unrelated regression
Metamodeling
Journal
IF:
4
Papers:
4.9K
Citations:
1.7W
Organization
Cited Papers
.png){kind=icon}
ShareSave
ShareSaveComplementing human judgment of essays written by English language learners with e-rater® scoring
PREAI
ShareSaveShareSave
ShareSave
View more