TY - JOUR

T1 - Efficient reliability-based optimization of linear dynamic systems with random structural parameters

AU - Yuan, Xiukai

AU - Gu, Jian

AU - Wu, Mingying

AU - Zhang, Feng

N1 - Funding Information: Our deepest gratitude goes to Prof. Marcos A. Valdebenito (from Universidad Adolfo Ibáñez, Chile) for his careful review and revision that have helped improving this paper. Funding Information: This work was supported in part by the Natural Science Foundation of Shanxi Province (2019JM-377), the Fundamental Research Funds for the Central Universities (310202006zy007), NSAF (Grant No. U1530122), the Aeronautical Science Foundation of China (Grant No. ASFC-20170968002) and the Fundamental Research Funds for the Central Universities of China (XMU, 20720180072).

PY - 2021/10

Y1 - 2021/10

N2 - This paper proposes a novel method to address the challenges associated with reliability-based design optimization (RBDO) of a class of problems, namely design of a linear system with random structural parameters subject to stochastic excitation. This method effectively estimates the failure probability as an explicit function of the design variables by representing the failure probability function (FPF) as a weighted average of sample values, which are generated by means of a single reliability analysis. The resulting estimation of the FPF is then used to decouple the target RBDO problem into a deterministic optimization problem, which can be solved by any appropriate deterministic optimization algorithm. In addition, a sequential approximate optimization framework is adopted to iteratively seek the solution of the RBDO problem. Several examples are provided to demonstrate the high accuracy and efficiency of the proposed method.

AB - This paper proposes a novel method to address the challenges associated with reliability-based design optimization (RBDO) of a class of problems, namely design of a linear system with random structural parameters subject to stochastic excitation. This method effectively estimates the failure probability as an explicit function of the design variables by representing the failure probability function (FPF) as a weighted average of sample values, which are generated by means of a single reliability analysis. The resulting estimation of the FPF is then used to decouple the target RBDO problem into a deterministic optimization problem, which can be solved by any appropriate deterministic optimization algorithm. In addition, a sequential approximate optimization framework is adopted to iteratively seek the solution of the RBDO problem. Several examples are provided to demonstrate the high accuracy and efficiency of the proposed method.

KW - Decoupling approach

KW - High dimensions

KW - Importance sampling

KW - Linear dynamic system

KW - Reliability-based optimization

KW - Uncertain systems

UR - http://www.scopus.com/inward/record.url?scp=85112363928&partnerID=8YFLogxK

U2 - 10.1007/s00158-021-03011-0

DO - 10.1007/s00158-021-03011-0

M3 - Article

AN - SCOPUS:85112363928

VL - 64

SP - 2593

EP - 2608

JO - Structural and Multidisciplinary Optimization

JF - Structural and Multidisciplinary Optimization

SN - 1615-147X

IS - 4

ER -