Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

Reliability-based design optimization (RBDO) has been used for optimizing engineering systems in presence of uncertainties in design variables, system parameters or both of them. RBDO involves reliability analysis, which requires a large amount of computational effort, especially in real-world application. To moderate this issue, a novel and efficient Surrogate-Assisted RBDO approach is proposed in this article. The computational intelligence and decomposition based RBDO procedures are combined to develop a fast RBDO method. This novel method is based on the artificial neural networks as a surrogate model and Sequential Optimization and Reliability Assessment (SORA) method as RBDO method. In SORA, the problem is decoupled into sequential deterministic optimization and reliability assessment. In order to improve the computational efficiency and extend the application of the original SORA method, an Augmented SORA (ASORA) method is proposed in this article. In developed method, A criterion is used for identification of inactive probabilistic constraints and refrain the satisfied constraints from reliability assessment to decrease computational costs associated with probabilistic constraints. Further, the variations of shifted vectors obtained for satisfied constraints are controlled to be exactly equal to zero for the next RBDO iteration. Several mathematical examples with different levels of complexity and a practical engineering example are solved and results are discussed to demonstrate efficiency and accuracy of the proposed methods.

Keywords: Optimization, Reliability-based Design, Computational Intelligence, Surrogate Model, Neural Networks

Full-Text [PDF 2067 kb]   (7895 Downloads)    

Add your comments about this article : Your username or Email:

---