TY - JOUR ID - 57581 TI - Structural Reliability: An Assessment Using a New and Efficient Two-Phase Method Based on Artificial Neural Network and a Harmony Search Algorithm JO - Civil Engineering Infrastructures Journal JA - CEIJ LA - en SN - 2322-2093 AU - Kazemi Elaki, Naser AU - Shabakhty, Naser AU - Abbasi Kia, Mostafa AU - Sanayee Moghaddam, Soroosh AD - M.Sc. of Structural Engineering, University of Sistan and Baluchestan AD - PhD of Civil Engineering Department of Civil Engineering University of Sistan and Baluchetsan AD - Msc of Computer science. Department of Mathematics, university of Lorestan AD - M.Sc. of Hydrolic Structures, University of Sistan and Baluchestan Y1 - 2016 PY - 2016 VL - 49 IS - 1 SP - 1 EP - 20 KW - Artificial Neural Network KW - Failure Probability KW - Harmony Search Algorithm KW - Implicit Limit State Function KW - Reliability Index DO - 10.7508/ceij.2016.01.001 N2 - In this research, a two-phase algorithm based on the artificial neural network (ANN) and a harmony search (HS) algorithm has been developed with the aim of assessing the reliability of structures with implicit limit state functions. The proposed method involves the generation of datasets to be used specifically for training by Finite Element analysis, to establish an ANN model using a proven ANN model in the reliability assessment process as an analyzer for structures, and finally estimate the reliability index and failure probability by using the HS algorithm, without any requirements for the explicit form of limit state function. The proposed algorithm is investigated here, and its accuracy and efficiency are demonstrated by using several numerical examples. The results obtained show that the proposed algorithm gives an appropriate estimate for the assessment of reliability of structures. UR - https://ceij.ut.ac.ir/article_57581.html L1 - https://ceij.ut.ac.ir/article_57581_f3e945db438e9cdb612d5492a37a5a50.pdf ER -