Title: Imperialist competitive algorithm approach to solve structural fuzzy random reliability
Authors: Mohammad Haranji; Mehdi Yazdani
Addresses: Department of Civil Engineering, Allameh Mohaddes Nouri University, Noor, Iran ' Department of Civil Engineering, Islamic Azad University, South Tehran Branch, Tehran, Iran
Abstract: In the presence of aleatory and epistemic uncertainties, the reliability index is a fuzzy number, instead of an ordinary (single-valued) one. A well-known approach to determine the fuzzy reliability is α-cut. The fuzzy random variable is transformed into an interval random variable for each membership degree α. In this method, the maximum and minimum of the reliability index corresponding to the membership degree α must be obtained. The proposed fuzzy random reliability is more accurate in comparison with the classical reliability. This article applies a metaheuristic algorithm, imperialist competitive algorithm (ICA), to search the maximum and minimum of the reliability index. The obtained results using the proposed algorithm are compared with outputs of a genetic algorithm. Results show the high performance of the ICA in obtaining desirable solutions. Considering the method applied and the algorithm hybridised to obtain reliability index, this is a new issue in structural reliability.
Keywords: imperialist competitive algorithm; ICA; structural reliability; fuzzy reliability index; fuzzy sets; uncertainty.
International Journal of Structural Engineering, 2022 Vol.12 No.1, pp.14 - 30
Received: 01 Nov 2020
Accepted: 05 Dec 2020
Published online: 30 Nov 2021 *