Afzali Sardoo, Roghaye (2019) Prediction of ultimate strength of steel fiber reinforced concrete beams using hybrid artificial intelligence. Masters thesis, University of Zabol.
|
Text
Prediction of ultimate strength of steel fiber reinforced concrete beams using hybrid artificial intelligence.pdf Download (849kB) | Preview |
Abstract
Currently, The use of modern composite materials in the construction industry has increased due to their greater capacity. Reinforced concrete beams with fibers are one of these materials. These fibers are available in different materials, including steel. Since adding more than steel fibers to concrete reduces the ultimate capacity, investigation of failure of these beams is an important issue. In this study, the ultimate capacity of reinforced concrete beams reinforced with steel fibers (SFRCB) was investigated. For these materials, two modes of bending and shear have been investigated using artificial intelligence methods, such as the genetic algorithm (GA), the particle swarm algorithm (PSO), the harmony search algorithm (HSA) and gray wolf optimization (GWO). In order to train these algorithms 385 experimental data for shear and 210 experimental data were used for bending. The results indicate superiority of artificial intelligence methods than empirical methods. The best neural network structure was achieved by gray wolf algorithm with 8 neuron. The results of training and testing of gray wolf algorithm by using error statistics are compared with several empirical models and other artificial intelligence models. Deductive statistics such as mean-square error (MSE) and compliance coefficient (
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | ultimate strengths, reinforced concrete beams, steel fibers, artificial intelligence |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) |
| Depositing User: | Mrs najmeh khajeh |
| Date Deposited: | 23 Oct 2021 07:15 |
| Last Modified: | 23 Oct 2021 07:15 |
| URI: | http://eprints.uoz.ac.ir/id/eprint/2852 |
Actions (login required)
 |
View Item |
